From cb@lim.nl Mon Jan 8 21:45:37 2001 From: cb@lim.nl (Colin Brace) Date: Mon, 08 Jan 2001 22:45:37 +0100 Subject: [MT-List] IBM's WebSphere Message-ID: <20010108214324.PEMG12023.amsmta01-svc@[62.108.30.75]> In, on 01/08/01 at 08:07 AM, "NewsScan" said: > IBM UNVEILS INSTANT TRANSLATION SOFTWARE > IBM's WebSphere Translation Server provides instant translation of Web > pages, e-mail and online chat from English into Spanish, German, French, > and Italian, and from those languages back into English. It also can > translate English into traditional Chinese, simplified Chinese, Japanese > and Korean, but does not have the reverse capability for those > languages. "Machine translation allows for dynamic business content to > be created at a fraction of the cost of professional translation > services," says Steve McClure, VP of Speech and Natural Language > Software, IDC. "At present, the machine translation industry consists > mainly of small segmented players, with no clear leader. IBM's entry > signifies a major turning point and will help validate and accelerate > the adoption of machine translation, a software market projected to > reach $378 million by 2003." The WebSphere Translation Server will be > available in March, priced at about $10,000 per processor. > (Reuters/Forbes.com 8 Jan 2001) > http://www.forbes.com/newswire/2001/01/08/rtr158437.html -- Colin Brace Amsterdam http://www.lim.nl From cb@lim.nl Fri Jan 12 13:45:34 2001 From: cb@lim.nl (Colin Brace) Date: Fri, 12 Jan 2001 14:45:34 +0100 Subject: [MT-List] Logos changes hands In-Reply-To: <973996956907773162940@multilingual.com> Message-ID: <20010112134815.TYWG17739.amsmta05-svc@[62.108.30.75]> I see that Logos has changed hands. Does anyone know anything about this company global words? In <973996956907773162940@multilingual.com>, on 01/11/01 at 04:28 PM, Seth Thomas Schneider <1293915293> said: > global words takes over Logos Corporation > The German language services provider global words GmbH, located in > Eisenach/Thuringia, has finalized the takeover of the American-based > Logos Corporation. According to global words' managing director Joerg > Ott, this merger will bring about one of the leading foreign language > service provider groups worldwide. Logos has been developing automatic > translation software based on linguistic logics since 1969 and is deemed > a worldwide leader in the field. > global words GmbH, Hospitalstrasse 6, 99817 Eisenach, Germany, Tel: > 49-36-91-204-0, Fax: 49-03691-204-204, E-mail: info@globalwords.de, Web: > http://www.globalwords.de -- Colin Brace Amsterdam http://www.lim.nl From rita.temmerman@planetinternet.be Fri Jan 12 14:15:28 2001 From: rita.temmerman@planetinternet.be (Rita Temmerman) Date: Fri, 12 Jan 2001 15:15:28 +0100 Subject: [MT-List] colloquium language technology, Brussels, 29 & 30 March 2001 Message-ID: <001901c07ca2$1f4c1580$df91efd4@default> This is a multi-part message in MIME format. ------=_NextPart_000_0016_01C07CAA.7EF8E820 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable On-line registration available from: = http://ttk.ehb.be/colloquim/colloquium.htm=20 Special rates for those registering before 16 February International Colloquium on Trends in Special Language and Language = Technology=20 Thursday, March 29, 2001 and Friday, March 30, 2001 Location: Brussels, Belgium Sponsored by Erasmushogeschool Brussels, De Nederlandse Taalunie and = Vrije Universiteit Brussels, the International Colloquium on Trends in = Special Language and Language Technology will take place on March 29-30, = 2001, in Brussels, Belgium. Colloquium topics for trends in special language include text, data and = knowledge management; corpora; information retrieval; and terminology = management. Colloquium topics for trends in language technology include Internet = challenges; computer translation and translation memory tools; = localization technology; controlled writing tools; and speech = technology. Among the speakers are J=FCrgen Muthig, "Technical Communication at the = Watershed"; Andr=E9 Taes, "Introducing CAT at the Translation Service of = the European Commission"; Joris Goetschalckx, "Machine Translation for = Dutch: The NL-Translex Project"; Jos Bailleul, "Translation Memories: = The Cypresoft Approach"; and Herman Caeyers, "Towards an = Industrialization of Translation Services."=20 Dr. Rita Temmerman, Co-ordinator Centrum voor Vaktaal en Communicatie, = Applied Linguistics, Erasmushogeschool, Trierstraat 84, B-1040, = Brussels, Belgium, Tel: 32-2-230-12-60, E-mail: rita.temmerman@ehb.be, = Web: http://ttk.ehb.be/colloquim/colloquium.htm=20 ------=_NextPart_000_0016_01C07CAA.7EF8E820 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
On-line registration available from: http://ttk.ehb.be/col= loquim/colloquium.htm=20
 
Special rates for those registering before 16 February
 
International Colloquium on Trends in = Special=20 Language and Language Technology
Thursday, March 29, 2001 and Friday, March 30, 2001
Location: Brussels, Belgium
 
Sponsored by Erasmushogeschool Brussels, De Nederlandse Taalunie = and Vrije=20 Universiteit Brussels, the International Colloquium on Trends in Special = Language and Language Technology will take place on March 29-30, 2001, = in=20 Brussels, Belgium.
 
Colloquium topics for trends in special language include text, data = and=20 knowledge management; corpora; information retrieval; and terminology=20 management.
 
Colloquium topics for trends in language technology include = Internet=20 challenges; computer translation and translation memory tools; = localization=20 technology; controlled writing tools; and speech technology.
 
Among the speakers are Jürgen Muthig, "Technical = Communication at=20 the Watershed"; André Taes, "Introducing CAT at the = Translation=20 Service of the European Commission"; Joris Goetschalckx, = "Machine=20 Translation for Dutch: The NL-Translex Project"; Jos Bailleul,=20 "Translation Memories: The Cypresoft Approach"; and Herman = Caeyers,=20 "Towards an Industrialization of Translation Services."
 
Dr. Rita Temmerman, Co-ordinator Centrum voor Vaktaal en = Communicatie,=20 Applied Linguistics, Erasmushogeschool, Trierstraat 84, B-1040, = Brussels,=20 Belgium, Tel: 32-2-230-12-60, E-mail: rita.temmerman@ehb.be, Web: http://ttk.ehb.be/col= loquim/colloquium.htm=20
------=_NextPart_000_0016_01C07CAA.7EF8E820-- From DenisV@ectaco.ru Thu Jan 18 07:57:25 2001 From: DenisV@ectaco.ru (EDC-Denis Volkov) Date: Thu, 18 Jan 2001 10:57:25 +0300 Subject: [MT-List] FYI: New English-Polish bidirectional full-text automatic transla tor released by Ectaco, Inc. Message-ID: Ectaco, Inc. (www.ectaco.com), a leading manufacturer of handheld electronic dictionaries and translating devices, has recently released its new English-Polish bidirectional full text automatic translation system. Free online version of English-Polish translator is available at http://www.poltran.com. Webmasters can place a free online translation form onto their sites. We believe it will be an attractive benefit to English and Polish-speaking Internet community. If you need any further information on MT and other linguistic technologies developed by Ectaco, please consult http://www.softwareoutsourcing.net. Best regards, Denis Volkov MT Linguistic Research Ectaco Development Center http://www.ectaco.com From ref@cs.cmu.edu Fri Jan 19 21:39:47 2001 From: ref@cs.cmu.edu (Robert Frederking) Date: Fri, 19 Jan 2001 16:39:47 -0500 Subject: [MT-List] Looking for MT Consultant In-Reply-To: Your message of Mon, 27 Nov 2000 18:29:43 -0500. Message-ID: <8346.979940387@euclid.mt.cs.cmu.edu> A startup company here in the US, for which I'm doing some consulting, also needs some consulting regarding the quality of the various current commercial MT systems. That's not really my forte, so I was wondering if anyone on this list was interested in such a consulting job, or knew of a contact that I could pass on to them. Thanks. Bob Frederking From reso.lang@sympatico.ca Mon Jan 22 04:44:01 2001 From: reso.lang@sympatico.ca (Nada Tamim) Date: Sun, 21 Jan 2001 23:44:01 -0500 Subject: [MT-List] Dutch-English or Dutch -French References: <8346.979940387@euclid.mt.cs.cmu.edu> Message-ID: <000801c0842d$f43ff1c0$8879e2d1@default> Hi all, A friend of mine is in Erethria and has to send translation requests any time his team needs a report or document from Dutch speaking teamworkers. He asked me if there is any translation tool that would assist him for a draft version while human translation is being processed. My knowledge of tools and translation memories is limited to French and English combination. Any recommendations as for the Dutch/English or Dutch /French? Anybody from PWGC or works for the Fed in Ottawa Canada would recommend a tool? Any reply is appreciated, Nada S. Tamim From steven.krauwer@let.uu.nl Mon Jan 22 09:45:54 2001 From: steven.krauwer@let.uu.nl (Steven Krauwer) Date: Mon, 22 Jan 2001 10:45:54 +0100 Subject: [MT-List] Dutch-English or Dutch -French References: <8346.979940387@euclid.mt.cs.cmu.edu> <000801c0842d$f43ff1c0$8879e2d1@default> Message-ID: <3A6C0152.65BFD4F8@let.uu.nl> Nada Tamim wrote: > A friend of mine is in Erethria and has to send translation requests any > time his team needs a report or document from Dutch speaking teamworkers. > He asked me if there is any translation tool that would assist him for a > draft version while human translation is being processed. My knowledge of > tools and translation memories is limited to French and English combination. > Any recommendations as for the Dutch/English or Dutch /French? Anybody from > PWGC or works for the Fed in Ottawa Canada would recommend a tool? > > Any reply is appreciated, I haven't seen the product myself, but Lernout & Hauspie's Simply Translating Deluxe might offer you something half-way useful for the purpose you sketched. The URL is http://www.lhsl.com/simplytranslating/deluxe/ It is only $99, but I must add that the company is in financial trouble at this moment, and I have no idea whether their sales channels still work. For Dutch there aren't any real alternatives, to my knowledge. Steven ______________________________________________________________________ Steven Krauwer, ELSNET / UiL OTS, Trans 10, 3512 JK Utrecht, Nederland phone: +31 30 2536050, fax: +31 30 2536000, email: s.krauwer@let.uu.nl http://www-sk.let.uu.nl http://www.elsnet.org From WJHutchins@compuserve.com Mon Jan 22 10:39:03 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 22 Jan 2001 05:39:03 -0500 Subject: [MT-List] Dutch-English or Dutch -French Message-ID: <200101220539_MC2-C28A-A3AC@compuserve.com> Dear Nada, Further to Steven Krauwer's answer. According to my data (in the "Compendium of translation software", available on the EAMT website www.eamt.org), translation software for Dutch to/from English is also offered by: PC-Translator (Linguistic Products) World Translator I have not seen either product (both cheap) so do not know the quality. URLs: www.PCTranslator.com www.worldlanguage.com As far as I can tell, the L&H product 'Simply Translating' is no more than a dictionary look-up system for many of its language pairs, and this includes Dutch. For Dutch-French there is only the old METAL system available on special terms from LANTMARK (probably not suiting your needs.) John Hutchins 22 Jan =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Nada Tamim wrote: > A friend of mine is in Erethria and has to send translation requests an= y > time his team needs a report or document from Dutch speaking teamworker= s. > He asked me if there is any translation tool that would assist him for = a > draft version while human translation is being processed. My knowledge= of > tools and translation memories is limited to French and English combination. > Any recommendations as for the Dutch/English or Dutch /French? Anybody from > PWGC or works for the Fed in Ottawa Canada would recommend a tool? > = > Any reply is appreciated, Steev Krauwer replied: I haven't seen the product myself, but Lernout & Hauspie's Simply Translating Deluxe might offer you something half-way useful for the purpose you sketched. = From ilidju@cal2.vsnl.net.in Tue Jan 23 11:27:43 2001 From: ilidju@cal2.vsnl.net.in (Dr. S. Bandyopadhyay) Date: Tue, 23 Jan 2001 16:57:43 +0530 Subject: [MT-List] WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION Message-ID: <000401c0852f$83119aa0$0e0101ca@notebook> This is a multi-part message in MIME format. ------=_NextPart_000_0001_01C0855D.9A65F600 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION Dear All, A Workshop on Example-Based Machine Translation is going to be held = as part of the MT-SUMMIT VIII. The details are given below. For further = details keep watching the Web-Site = http://www.compapp.dcu.ie/~away/EBMT.html or mail back to me at = sivaji_ju@vsnl.com. Dr. Sivaji Bandyopadhyay Reader, Computer Science & Engineering Department Jadavpur University, Calcutta - 700 032. India Workshop on Example-Based Machine Translation: 1st Call for Papers Hosted by MT-SUMMIT VIII Santiago de Compostella, Spain, September 18-22, 2001 http://www.eamt.org/summitVIII/index.html Co-chairs: Michael Carl, IAI, Saarbr=FCcken; Andy Way, Computer = Applications, Dublin City University=20 In recent years, corpora of multilingual translated texts have become = widely available for a number of languages. Notwithstanding the seminal = paper by Nagao (84), it is primarily since the early 90's that such = bilingual texts have been exploited in the area of Machine Translation = (MT).=20 The two main paradigmatic approaches which have been proposed are = Statistics-based Machine Translation (SBMT) and Example-Based Machine = Translation (EBMT). A related variant of EBMT that we ignore here, = depsite being widely used in the localisation area, is that of = Translation Memories (TM). No new translations are created afresh from = previously existing examples in the system database: rather, the closest = translation matches are proposed to the user for post-editing into the = correct translation.=20 While translation memory systems are used in restricted domains, SBMT = systems require training on huge, good quality bilingual corpora. As a = consequence TMs can hardly be applied as a general purpose solution to = MT and SBMT as yet cannot produce complex translations to the desired = quality, even if such translations are given to the system in the = training phase. EBMT seeks to exploit and integrate a number of = knowledge resources, such as linguistics and statistics, and symbolic = and numerical techniques, for integration into one framework. In this = way, rule-based morphological, syntactic and/or semantic information is = combined with knowledge extracted from bilingual texts which is then = re-used in the translation process.=20 However, it is unclear how one might combine the different knowledge = resources and techniques in an optimal way. In EBMT, therefore, the = question is asked: what can be learned from a bilingual corpus and what = needs to be manually provided? Furthermore, we remain uncertain as to = how far the EBMT methodology can be pushed with respect to translation = quality and/or translation purpose. Finally, one wonders what the = implications and consequences are for size and quality of the reference = translations, (computational) complexity of the system, sizeability and = transportability, if such an approach is taken.=20 Given this background, we propose to organize a workshop in order to = shed some light on these open questions, among others. We are seeking = contributions which go beyond the purely statistical and/or rule-based = approaches to MT. We welcome visionary and technical descriptions, = reports of empirical research as well as feasibility studies and system = demonstrations. We would welcome contributions on any of the following = topics and sub-headings:=20 a.. (semi-)automatic preparation of existing bi/multilingual corpora = for EBMT=20 a.. extraction of bi/multilingual texts from the web=20 b.. preparation of treebanks for EBMT=20 c.. bi/multilingual alignment/bracketing/parsing=20 d.. inference of bi/multilingual grammar and transfer rules=20 b.. description of `pure' EBMT systems=20 a.. knowledge resources used=20 b.. representation of numeric and symbolic knowledge=20 c.. descriptions of `hybrid' systems integrating EBMT with rule-based = or other methodologies=20 d.. evaluation of EBMT results and/or comparison with other MT systems = e.. considerations on domain-(in)dependence of EBMT systems=20 f.. computational and/or system complexity of EBMT systems=20 Submissions We expect to be able to accomodate 12 30-minute talks, in addition to = the invited guest speaker (to be arranged). There may also be poster = sessions, subject to demand. We also strongly encourage system = demonstrations, either in conjunction with contentful paper = presentations or as stand-alone demos during the lunch and coffee = intervals.=20 Submitted papers must describe original, previously unpublished work. = Submissions must not exceed 12 pages. Conference stylesheets are now = available.=20 Publication We plan to adhere to the publication criteria of the main conference = proceedings. That is, the workshop contributions will appear on CD-ROM. = However, it is anticipated that relevant publishers will be approached = to see if there would be interest in publishing the proceedings more = widely.=20 Important Dates a.. January 2001 First call for papers/demos=20 b.. 15.4.2001 Deadline for receipt of papers=20 c.. 30.5.2001 Notification of acceptance=20 d.. 01.7.2001 Final Paper due=20 e.. 18 or 19.9.2001 Workshop takes place=20 Attendance Fee This is likely to be of the order of Euro 100. This would cover = facilities and refreshments (though probably not lunch), and may include = some of the cost of proceedings.=20 Organizing Committee a.. Sivaji Bandyopadhyay, India=20 b.. Ralf Brown, USA=20 c.. Michael Carl, Germany=20 d.. Ilyas Cicekli, Turkey=20 e.. Brona Collins, Belgium=20 f.. Oliver Streiter, Taiwan=20 g.. Stephan Vogel, Germany=20 h.. Andy Way, Ireland=20 -------------------------------------------------------------------------= ------- Andy Way, 22nd January 2001. ------=_NextPart_000_0001_01C0855D.9A65F600 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION
 
     Dear=20 All,
 
    A = Workshop on=20 Example-Based Machine Translation is going to be held as part of the = MT-SUMMIT=20 VIII. The details are given below. For further details keep watching=20     the Web-Site http://www.compapp.dcu= .ie/~away/EBMT.html=20 or mail back to me at sivaji_ju@vsnl.com.
 
    Dr. = Sivaji=20 Bandyopadhyay
    = Reader, Computer=20 Science & Engineering Department
    = Jadavpur=20 University, Calcutta - 700 032.
    = India

Workshop on Example-Based Machine Translation: 1st Call for=20 Papers

Hosted by MT-SUMMIT=20 VIII
Santiago de Compostella, Spain,
September 18-22, 2001
http://www.eamt.org/summitVIII/in= dex.html

Co-chairs: Michael Carl, = IAI,=20 Saarbr=FCcken; Andy Way, = Computer=20 Applications, Dublin City University=20

In recent years, corpora of multilingual translated texts have become = widely=20 available for a number of languages. Notwithstanding the seminal paper = by Nagao=20 (84), it is primarily since the early 90's that such bilingual texts = have been=20 exploited in the area of Machine Translation (MT).=20

The two main paradigmatic approaches which have been proposed are=20 Statistics-based Machine Translation (SBMT) and Example-Based Machine=20 Translation (EBMT). A related variant of EBMT that we ignore here, = depsite being=20 widely used in the localisation area, is that of Translation Memories = (TM). No=20 new translations are created afresh from previously existing examples in = the=20 system database: rather, the closest translation matches are proposed to = the=20 user for post-editing into the correct translation.=20

While translation memory systems are used in restricted domains, SBMT = systems=20 require training on huge, good quality bilingual corpora. As a = consequence TMs=20 can hardly be applied as a general purpose solution to MT and SBMT as = yet cannot=20 produce complex translations to the desired quality, even if such = translations=20 are given to the system in the training phase. EBMT seeks to exploit and = integrate a number of knowledge resources, such as linguistics and = statistics,=20 and symbolic and numerical techniques, for integration into one = framework. In=20 this way, rule-based morphological, syntactic and/or semantic = information is=20 combined with knowledge extracted from bilingual texts which is then = re-used in=20 the translation process.=20

However, it is unclear how one might combine the different knowledge=20 resources and techniques in an optimal way. In EBMT, therefore, the = question is=20 asked: what can be learned from a bilingual corpus and what needs to be = manually=20 provided? Furthermore, we remain uncertain as to how far the EBMT = methodology=20 can be pushed with respect to translation quality and/or translation = purpose.=20 Finally, one wonders what the implications and consequences are for size = and=20 quality of the reference translations, (computational) complexity of the = system,=20 sizeability and transportability, if such an approach is taken.=20

Given this background, we propose to organize a workshop in order to = shed=20 some light on these open questions, among others. We are seeking = contributions=20 which go beyond the purely statistical and/or rule-based approaches to = MT. We=20 welcome visionary and technical descriptions, reports of empirical = research as=20 well as feasibility studies and system demonstrations. We would welcome=20 contributions on any of the following topics and sub-headings:=20

  • (semi-)automatic preparation of existing bi/multilingual corpora = for EBMT=20
    • extraction of bi/multilingual texts from the web=20
    • preparation of treebanks for EBMT=20
    • bi/multilingual alignment/bracketing/parsing=20
    • inference of bi/multilingual grammar and transfer rules =
  • description of `pure' EBMT systems=20
    • knowledge resources used=20
    • representation of numeric and symbolic knowledge
  • descriptions of `hybrid' systems integrating EBMT with rule-based = or other=20 methodologies=20
  • evaluation of EBMT results and/or comparison with other MT systems =
  • considerations on domain-(in)dependence of EBMT systems=20
  • computational and/or system complexity of EBMT systems

Submissions

We expect to be able to accomodate 12 30-minute talks, in addition to = the=20 invited guest speaker (to be arranged). There may also be poster = sessions,=20 subject to demand. We also strongly encourage system demonstrations, = either in=20 conjunction with contentful paper presentations or as stand-alone demos = during=20 the lunch and coffee intervals.=20

Submitted papers must describe original, previously unpublished work. = Submissions must not exceed 12 pages. Conference = stylesheets are=20 now available.=20

Publication

We plan to adhere to the publication criteria of the main conference=20 proceedings. That is, the workshop contributions will appear on CD-ROM. = However,=20 it is anticipated that relevant publishers will be approached to see if = there=20 would be interest in publishing the proceedings more widely.=20

Important Dates

  • January 2001 First call for papers/demos=20
  • 15.4.2001 Deadline for receipt of papers=20
  • 30.5.2001 Notification of acceptance=20
  • 01.7.2001 Final Paper due=20
  • 18 or 19.9.2001 Workshop takes place

Attendance Fee

This is likely to be of the order of Euro 100. This would cover = facilities=20 and refreshments (though probably not lunch), and may include some of = the cost=20 of proceedings.=20

Organizing Committee

Andy Way, 22nd January 2001.

 

------=_NextPart_000_0001_01C0855D.9A65F600-- From Jo_Grant@lotus.com Wed Jan 24 15:53:02 2001 From: Jo_Grant@lotus.com (Jo_Grant@lotus.com) Date: Wed, 24 Jan 2001 10:53:02 -0500 Subject: [MT-List] L&H Message-ID: This is a multipart message in MIME format. --=_alternative 00583F9C852569DE_= Content-Type: text/plain; charset="us-ascii" Stephen writes: >I must add that the company [L&H] is in financial >trouble at this moment, and I have no idea whether their sales >channels still work. Hiya, I'm just back from showcasing the Lotus Translation Components at LotusSphere in Orlando. In our Globalization Lab we had IBM, Transparent Language, Alis Technologies, Systran, and L&H showing their Machine Translation technologies interfacing through the LTC to Domino, Sametime and WebSphere. (Trados was also showing their Translation Memory product, and WorldPoint their Human Translation product.) I used the opportunity to chat with the L&H reps about the recent going ons in L&H. Apparently what is happening is that the MT bit is being split off into another subsidiary company. Since they have a sound product, business plan and management, not to mention the largest installed base of MT users, I think they are going to be in the game for quite some time. [For more information on the LTC, the industry leaders who have adopted it as their standard, and how to adopt it, see www.lotus.com/international.] Cheers, Jo Grant Software Architect, Lotus Translation Components Lotus Development Corp, an IBM company tel. 617 693 5703 fax. 617 693 5542 url: www.lotus.com/international email: jo_grant@lotus.com AIM: DTOGuy If as one people speaking the same language they have begun to do this, then nothing they plan to do will be impossible for them. - Genesis 11:6 --=_alternative 00583F9C852569DE_= Content-Type: text/html; charset="us-ascii"
Stephen writes:
>I must add that the company [L&H] is in financial
>trouble at this moment, and I have no idea whether their sales
>channels still work.

Hiya,
        I'm just back from showcasing the Lotus Translation Components at LotusSphere in Orlando. In our Globalization Lab we had IBM, Transparent Language, Alis Technologies, Systran, and L&H showing their Machine Translation technologies interfacing through the LTC to Domino, Sametime and WebSphere. (Trados was also showing their Translation Memory product, and WorldPoint their Human Translation product.)
        I used the opportunity to chat with the L&H reps about the recent going ons in L&H. Apparently what is happening is that the MT bit is being split off into another subsidiary company. Since they have a sound product, business plan and management, not to mention the largest installed base of MT users, I think they are going to be in the game for quite some time.
        [For more information on the LTC, the industry leaders who have adopted it as their standard, and how to adopt it, see www.lotus.com/international.]
        Cheers,
                        Jo Grant
Software Architect, Lotus Translation Components
Lotus Development Corp, an IBM company
tel. 617 693 5703
fax. 617 693 5542
url: www.lotus.com/international
email: jo_grant@lotus.com
AIM: DTOGuy

If as one people speaking the same language they have begun to do this, then nothing they plan to do will be impossible for them.
- Genesis 11:6 --=_alternative 00583F9C852569DE_=-- From mlf@dlsi.ua.es Thu Jan 25 08:55:03 2001 From: mlf@dlsi.ua.es (Mikel L. Forcada) Date: Thu, 25 Jan 2001 09:55:03 +0100 Subject: [MT-List] CFP: "Teaching MT" workshop at MT Summit VIII Message-ID: <3A6FE9E7.6D1AA9CC@dlsi.ua.es> [please accept our apologies if you receive multiple copies of this message] Teaching Machine Translation A workshop at Machine Translation Summit VIII September 18-22, 2001, Santiago de Compostela, Spain CALL FOR PAPERS Introduction In September 2001, the workshop Teaching Machine Translation will be held as part of the VIII Machine Translation Summit in Santiago de Compostela, Galicia, Spain. The workshop will provide an opportunity for machine translation instructors attending MT Summit VIII to exchange their experience by presenting papers or demonstrations describing the tools and techniques they use in the classroom or in the laboratory. Schedule and venue This one-day workshop will most probably take place either on September 18 or 19, 2001, before the regular conference starts. Please visit the Workshop website at http://www.dlsi.ua.es/tmt/ for updates. Papers The Teaching Machine Translation workshops seeks original papers in all aspects of the instruction of machine translation. Topics of interest include, but are not limited to: * why and to whom should MT be taught? * teaching the theoretical background of MT: linguistics, computer science, translation theory * addressing preconceptions about MT in the classroom * the use of commercial MT programs in hands-on teaching * teaching machine translation strategies to non-computer science majors * web-based distance learning of MT * teaching MT as a component of multilingual information resources and as a part of the entire documentation process * MT education and industry: bridging the gap between academia and the real world * teaching pre- and post-MT skills to MT users * teaching MT-related techniques: translation memory, alignment tools, terminological databases, use of corpora, etc. * teaching MT evaluation * building modules or `toy' MT systems in the laboratory * experiences on the evaluation of MT instruction The working language of the workshop will be English. Papers should describe research or experiences in any of the topics mentioned, and should not be longer than 3000 words. In view of the main topics of the workhop, presentations may have a substantial demonstrative component. Papers should be submitted to the address below, preferably by e-mail, and must be received by the indicated date. Papers should include a cover page with the following information: * paper title, * author(s)' name(s), affiliation(s), address(es), and e-mail address(es), * 100 word abstract, * up to 5 keywords, * for demonstrative presentations: the hardware, software and network requirements for the demonstration. Please submit electronically to tmt@dlsi.ua.es, following the guidelines for authors provided at the Teaching Machine Translation website http://www.dlsi.ua.es/tmt/. If you cannot submit electronically, please mail 4 hardcopies of the paper to: Mikel L. Forcada Teaching Machine Translation at MT Summit VIII Dept. Llenguatges i Sistemes Informàtics Universitat d'Alacant E-03071 Alacant, Spain Important Dates 15 April 2001 Paper submission deadline 31 May 2001 Notifications 1 July 2001 Final camera-ready copy deadline About Santiago and the Venue Santiago de Compostela in North West Spain (Galicia) was the most visited pilgrimage shrine in the Middle Ages, and to this day the Road to Santiago is a popular route for walkers through France and northern Spain. The goal is the imposing 11th century cathedral containing the relics of St James the Great. But the well preserved historical centre of the city offers much more: medieval streets, excellent restaurants, ancient churches, monasteries, palaces, etc. The venue for the conference is the Hostal de los Reyes Católicos, founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style, and now a hotel (the oldest and one of the most beautiful in the world). Those staying at the Hostal will have a once-in-a-lifetime experience, but there is also accommodation nearby in many other good hotels, all within easy walking distance of the centre and the Hostal. Santiago de Compostela has its own airport with daily direct flights from Barcelona, Madrid, Brussels, and London. Workshop Organization Workshop Co-chairmen Derek R. Lewis, University of Exeter, UK. d.r.lewis@exeter.ac.uk Mikel L. Forcada, Universitat d'Alacant, Spain. mlf@dlsi.ua.es Organizing committee F. Gaspari, Univ. Bologna, Italy M.H. Moeller, Syddansk Univ., Denmark J.A. Pérez-Ortiz, Univ. Alacant Further Information For more details, please visit the workshop website: http://www.dlsi.ua.es/tmt/ . You may also send a request for information to tmt@dlsi.ua.es . -- _____________________________________________________________________ Mikel L. Forcada E-mail: mlf@dlsi.ua.es Departament de Llenguatges Phone: +34-96-590-3400 ext. 3384; i Sistemes Informàtics also +34-96-590-3772. UNIVERSITAT D'ALACANT Fax: +34-96-590-9326, -3464 E-03071 ALACANT, Spain. URL: http://www.dlsi.ua.es/~mlf From WJHutchins@compuserve.com Mon Feb 5 11:18:23 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 5 Feb 2001 06:18:23 -0500 Subject: [MT-List] update of Compendium Message-ID: <200102050618_MC2-C43F-31FA@compuserve.com> The latest version of the "Compendium of translation software" has been recently loaded on the EAMT website (www.eamt.org). Subscribers can continue to use the same ID and passwords. Note that there have been relatively few additions since the printed first edition (and first Web version) in April 2000. The changes affecting Lernout & Hauspie and Logos Corporation have been noted, but no changes to their entries will be made until their situations have been fully clarified. The next update is scheduled for April. Any information about new products or changes to any entries in the Compendium will be most welcome. Please contact me at: WJHutchins@compuserve.com = John Hutchins 5 Feb From Sandra Manzi Thu Feb 8 09:54:23 2001 From: Sandra Manzi (Sandra Manzi) Date: Thu, 08 Feb 2001 10:54:23 +0100 (MET) Subject: [MT-List] MT evaluation workshop: call for participation Message-ID: MT Evaluation : An invitation to get your hands dirty! Maghi King, ISSCO and Flo Reeder, Mitre (in conjunction with the MT Evaluation Working Group of the ISLE project) Background. A workshop on MT evaluation organised during the AMTA conference in Cuernavaca in 2000 included a series of practical exercises on machine translation evalauation. Carrying out the exercises provided insights into the difficulties and subtleties of MT evaluation, thus inspiring several of those present to suggest the organisation of a longer workshop whose primary focus would be to design and carry out portions of a thorough evaluation. At the same time, the Evaluation Working Group of the ISLE project (funded by the EU, the NSF in the USA and by the Swiss and Danish governments) has been working on the provision of support material for those involved in MT evaluation. This material takes the form of classification schemes intended to be helpful in the definition of user needs, the choice of system characteristics of importance to the specific evaluation and the choice of metrics to be applied to system characteristics. The current version of the ISLE proposals can be seen at http://www.isi.edu/natural-language/mteval/. Date and Place. We invite you to a practical workshop to be held in Geneva between April 19th and 24th 2001. Organisation and activities. Participants in the workshop will be provided with a scenario describing a practical situation in which an evaluation of an MT system or systems might be undertaken. The organisers will ensure that the scenario(s) reflect real life situations. Particpants will then spend two days designing an evaluation which is appropriate to their scenario, using a unified framework (ISLE) described in the introductory talks. They may choose to work alone or in small groups. Participants will have free access to the machine translation systems available on the web, and to the considerable computing support available at the University of Geneva School of Translation and Interpretation. As much as possible, the evaluations will be carried out. Results and experience will be pooled and discussed in the final day of the workshop. This workshop can be seen as one in an-going series (LREC 2000, AMTA 2000, NAACL 2001, MT Summit VIII 2001), where each workshop builds on the experience and the results of previous workshops. Potential participants need not however have participated in earlier workshops, although we would of course like to encourage them to participate in later ones? More information about the conferences where workshops have been held or will be held can be found at LREC : http://www.icp.grenet.fr/ELRA/lrec2000.html AMTA : http://www.isi.edu/natural-language/conferences/AMTA2000.html NAACL 2001 : http://www.cs.cmu.edu/?ref/NAACL2001 MT Summit VIII : http://www.eamt.org/summitVIII Week outline timetable. * April 19th, morning: Introduction of the ISLE proposals. Distribution and discussion of scenarios, formation of working groups. * April 19th afternoon, April 20th: Design of evaluations. * April 21st, morning: Execution of evaluations. * April 21st, afternoon: Free time. * April 22nd: Free time. * April 23rd: Interpretation of evaluation scores and metrics. * April 24th: Reports and discussion of results. Major themes of interest. * What metrics are suitable for assessing what system characteristics ? * What system characteristics reflect what user needs ? * Is there a radical difference between evaluation focusing on research or development needs and evaluation focusing on end-user needs? * When should real world data be used, and what is the impact of using it? * What constitutes a valid metric? How can you demonstrate that a metric is valid? * What metrics can be automated? * What are the advantages and disadvantages of specific metrics? * For the metric(s) selected for the evaluation, what are the difficulties in applying them? * For a given metric, what variations in scores are typically produced? What are the statistical error variances? * For a given metric, what are the score ranges for 'good' and for 'bad' systems? * Are there metrics which correlate with one another ? Are there metrics which indicate an overall quality score? * Are there metrics which work better with specific language pairs? Participation. Participation in the workshop is free of charge, although particpants must pay their own travel and living expenses. Because of the nature of the exercise, participation is limited to a maximum of 20 persons, and will be on a first come first served basis. Note, though, that if there is a team which would like to participate as a team, these restrictions may be relaxed in order to accomodate them. Further information. Further information can be obtained from Maghi King at Margaret.King@issco.unige.ch or from Florence Reeder at freeder@mitre.org. How to register. Send your registration request to Gisella Anspach at Gisella.Anspach@issco.unige.ch as soon as possible and at the absolute latest by March 15th. Gisella will also be able to help you if necessary with finding accomodation in Geneva, as will the Geneva Tourist Office whose site at http://www.geneva-tourism.ch/eng/ will provide you with much information about the city. DEADLINES REGISTRATION -- MARCH 15TH 2001 LATEST NOTIFICATION OF ACCEPTANCE -- MARCH 22ND 2001 WORKSHOP -- APRIL 19-24 2001 ----------------------------------------------------------------- Sandra Manzi | e-mail: Sandra.Manzi@issco.unige.ch ******* New Address as of Sept 28th, 1999 ******* TIM/ISSCO | Tel: +41-22-705.8680 University of Geneva | Fax: +41-22-705.8689 40 blvd du Pont d'Arve | WWW: http://issco-www.unige.ch/ CH-1211 GENEVA 4 (Switzerland) | From jsamuelson@wordstream.com Thu Feb 8 22:27:08 2001 From: jsamuelson@wordstream.com (Joanne Samuelson) Date: Thu, 08 Feb 2001 17:27:08 -0500 Subject: [MT-List] a job posting PLEASE Message-ID: <3A831D3C.6A8E6E8E@wordstream.com> --------------FEA22A9DAD8836B5C7028A15 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hello. We at WordStream would be very grateful if you would be so kind as to post the below employment opportunity for us. Feel free to call or write with questions, of course, and many thanks in advance. If it isn't too much trouble I'd be appreciative of a confirmation. Kind regards, Joanne Samuelson Director of Language Development, Computational Linguist Opportunity to join an exciting young company with a creative approach to language translation. WordStream isn’t trying to be another Machine Translation company but rather has redefined the problem and is solving it with a unique technology based on an Interlingua. Competitive salary and benefits. Metropolitan Boston location. Job Description: Computational linguist to lead development and design of commercial interlingua MT system and to manage a team of linguists. The job will involve determining the conceptual framework for the development of a lexical database and grammar system. Candidate will need to work closely with Software Development team to ensure coordination of linguistic and technical requirements. Job Requirements: Master’s (PhD preferred) in Computational Linguistics or related areas. Fluent in English and at least one other language, preferably one of the following: German, French, Spanish, Italian, Japanese or Chinese. Excellent communication skills as well as the ability to both motivate a team and be a team player. Experience in grammar and lexicon development. Familiarity with linguistic resources (corpora, grammars, parsers, etc.). Highly Desirable: Experience with commercial MT systems along with project management and implementation of MT applications. Experience managing linguists and technology specialists. Programming skills (knowledge of perl, java, or C++). Interested parties should contact : www.wordstream.com careers@wordstream.com WordStream Inc. 48 Grove St. Suite 201 Somerville MA 02144 Tel 617-666-7700 X231 Fax 617-666-2850 --------------FEA22A9DAD8836B5C7028A15 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hello.  We at WordStream would be very grateful if you would be so kind as to post
the below employment opportunity for us.  Feel free to call or write with questions, of course,
and many thanks in advance.  If it isn't too much trouble I'd be appreciative of a confirmation.

Kind regards, Joanne Samuelson

Director of Language Development, Computational Linguist
Opportunity to join an exciting young company with a creative approach to
language translation.   WordStream isn’t trying to be another Machine
Translation company but rather has redefined the problem and is solving
it with a unique technology based on an Interlingua.
Competitive salary and benefits.  Metropolitan Boston location.

Job Description:
Computational linguist to lead development and design of commercial
interlingua MT system and to manage a team of linguists. The job will
involve determining the conceptual framework for the development of a
lexical database and grammar system. Candidate will need to work closely
with Software Development team to ensure coordination of linguistic and
technical requirements.

Job Requirements:
Master’s (PhD preferred) in Computational Linguistics or related areas.
Fluent in English and at least one other language, preferably
one of the following: German, French, Spanish, Italian, Japanese or Chinese.
Excellent communication skills as well as the ability to both motivate a team and
be a team player.
Experience in grammar and lexicon development.
Familiarity with linguistic resources (corpora, grammars, parsers, etc.).

Highly Desirable:
Experience with commercial MT systems along with project management and
implementation of MT applications.
Experience managing linguists and technology specialists.
Programming skills (knowledge of perl, java, or C++).

Interested parties should contact :

www.wordstream.com
careers@wordstream.com
WordStream Inc.
48 Grove St. Suite 201
Somerville MA 02144
Tel 617-666-7700 X231
Fax 617-666-2850
  --------------FEA22A9DAD8836B5C7028A15-- From Margaret.King@issco.unige.ch Sun Feb 18 14:43:24 2001 From: Margaret.King@issco.unige.ch (Maghi King) Date: Sun, 18 Feb 2001 15:43:24 +0100 Subject: [MT-List] Big software companies and MT Message-ID: <3A8FDF8C.432DB0B@issco.unige.ch> Hello Everybody! As some of you know, I'm working on a project that involves assessing the present state of the art in the use of technology to support translation and localisation. One thing that has struck me is that some big companies are getting involved in different ways: translation buttons appearing all over the place, for example, or IBM re-launching its TM system and launching a commercial MT system. Does anybody have a feeling that there is in general a growing interest amongst the big software companies, or even know of interesting recent developments? Maybe L&H's problems will put them off? Thanks a lot for any answers! Maghi -- IMPORTANT: ISSCO has moved (from 28th September 1999) AND my phone number keeps changing. Please check that you have the latest one below. Maghi King | E-mail: Margaret.King@issco.unige.ch TIM/ISSCO, ETI University of Geneva | WWW: http://issco-www.unige.ch/ 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 From stopping@rochester.rr.com Sun Feb 18 16:22:07 2001 From: stopping@rochester.rr.com (Suzanne Topping) Date: Sun, 18 Feb 2001 11:22:07 -0500 Subject: [MT-List] Big software companies and MT References: <3A8FDF8C.432DB0B@issco.unige.ch> Message-ID: <005c01c099c6$f04e1fc0$351d5d18@rochester.rr.com> ----- Original Message ----- From: "Maghi King" > > One thing that has struck me is that some big companies are getting > involved in different ways: translation buttons appearing all over the > place, for example, or IBM re-launching its TM system and launching a > commercial MT system. Does anybody have a feeling that there is in general > a growing interest amongst the big software companies, or even know of > interesting recent developments? Maybe L&H's problems will put them off? Are you interested specifically in MT, or in CAT technologies as well? From Margaret.King@issco.unige.ch Sun Feb 18 18:27:52 2001 From: Margaret.King@issco.unige.ch (Maghi King) Date: Sun, 18 Feb 2001 19:27:52 +0100 Subject: [MT-List] Big software companies and MT References: <3A8FDF8C.432DB0B@issco.unige.ch> <005c01c099c6$f04e1fc0$351d5d18@rochester.rr.com> Message-ID: <3A901428.C59B544C@issco.unige.ch> In anything really. I was just wondering if there was a trend developing. Maghi Suzanne Topping wrote: > > ----- Original Message ----- > From: "Maghi King" > > > > One thing that has struck me is that some big companies are getting > > involved in different ways: translation buttons appearing all over the > > place, for example, or IBM re-launching its TM system and launching a > > commercial MT system. Does anybody have a feeling that there is in general > > a growing interest amongst the big software companies, or even know of > > interesting recent developments? Maybe L&H's problems will put them off? > > Are you interested specifically in MT, or in CAT technologies as well? > > -- > For MT-List info, see http://www.eamt.org/mt-list.html -- IMPORTANT: ISSCO has moved (from 28th September 1999) AND my phone number keeps changing. Please check that you have the latest one below. Maghi King | E-mail: Margaret.King@issco.unige.ch TIM/ISSCO, ETI University of Geneva | WWW: http://issco-www.unige.ch/ 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 From John.Kohl@sas.com Sun Feb 18 21:18:11 2001 From: John.Kohl@sas.com (John Kohl) Date: Sun, 18 Feb 2001 16:18:11 -0500 Subject: [MT-List] Big software companies and MT Message-ID: <4F2457EC921ED31188C20090276DC8C402553628@merc02.us.sas.com> Hello Maghi and others, My employer, SAS, is a fairly large, privately owned software company with annual sales of about USD 1.1 billion (that's billion as in thousand million). I have been investigating Controlled English authoring tools (mainly those developed by Smart, LANT, and the Carnegie Mellon Language Technologies Institute) on the company's behalf and have found out in the course of my investigation that a couple of much larger, publicly held software companies are also investigating CE tools. I don't know whether these other companies are also investigating MT, but here at SAS, the new director of our Software Globalization Division is very interested in it. Like a lot of other software companies, we are doing more and more translation, and the director is alarmed by how much it costs. Our translators have been using TM for a couple of years, and we are just starting to reap some cost savings from that. Although I have urged SAS for eight years to investigate CE and MT, I am not sure the language technology industry is mature enough for me to want to see SAS actually start using these technologies yet. It's still much too difficult to get the kind of detailed information that I need from the CE and MT vendors. (It's hard to figure out what questions to ask, but you would think that they would offer more details to help educate their prospective clients and to alleviate some of the clients' uncertainties.) There are still far too many unknowns, and far too many "pieces" to integrate. It seems you have to do a pilot project before you can know what you are getting into, and the question is, what combination of tools do you use for the pilot? The ISLE framework is interesting, but what I need/want is not just a list of criteria and considerations, but the facts as they pertain to the different vendors' products. I think -that- information is a long way from being made readily available. Nevertheless, I have been directed to pursue these investigations, so I will have a better idea of "what's out there" in a few months, I suppose. Regards, John Kohl, Senior Editor SAS J2122 SAS Campus Drive Cary, NC 27513 USA 919.531.6149 fax: 919.677.4444 http://www.sas.com SAS... The Power to Know > -----Original Message----- > From: Maghi King [mailto:Margaret.King@issco.unige.ch] > Sent: Sunday, February 18, 2001 9:43 AM > To: MT-list@eamt.org > Subject: [MT-List] Big software companies and MT > > > Hello Everybody! > > As some of you know, I'm working on a project that involves > assessing the > present state of the art in the use of technology to support > translation > and localisation. > > One thing that has struck me is that some big companies are getting > involved in different ways: translation buttons appearing all over the > place, for example, or IBM re-launching its TM system and launching a > commercial MT system. Does anybody have a feeling that there > is in general > a growing interest amongst the big software companies, or even know of > interesting recent developments? Maybe L&H's problems will > put them off? > > Thanks a lot for any answers! > > Maghi > -- > IMPORTANT: ISSCO has moved (from 28th September 1999) > AND my phone number keeps changing. Please check that you > have the latest > one below. > > Maghi King | E-mail: Margaret.King@issco.unige.ch > TIM/ISSCO, ETI > University of Geneva | WWW: http://issco-www.unige.ch/ > 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 > CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > From gpl@di.fct.unl.pt Mon Feb 19 10:36:30 2001 From: gpl@di.fct.unl.pt (Gabriel Pereira Lopes) Date: Mon, 19 Feb 2001 10:36:30 +0000 Subject: [MT-List] Big software companies and MT References: <3A8FDF8C.432DB0B@issco.unige.ch> Message-ID: <3A90F72E.4DCFCBB4@di.fct.unl.pt> Maghi, I am also working in an European project (MLIS) with Systran. My group is mainly working on automatic extraction of linguistic information from parallel and non-parallel texts. This will enable the inhancement of lexicons production, be either for single word and multiword lexical units. The quality of translation will also improve quite rapidly. About the interest of various companies on MT I have also observed that but I do not have any plausible explanation except that new challenges are created for CLIR e-learning, etc., etc. Best regards, Gabriel Pereira Lopes Departamento de Informática Faculdade de Ciências e Tecnologia Universidade Nova de Lisboa Quinta da Torre 2825-114 Caparica Portugal Phone: 351 21 294 8536 fax: 351 21 294 8541 Maghi King wrote: > Hello Everybody! > > As some of you know, I'm working on a project that involves assessing the > present state of the art in the use of technology to support translation > and localisation. > > One thing that has struck me is that some big companies are getting > involved in different ways: translation buttons appearing all over the > place, for example, or IBM re-launching its TM system and launching a > commercial MT system. Does anybody have a feeling that there is in general > a growing interest amongst the big software companies, or even know of > interesting recent developments? Maybe L&H's problems will put them off? > > Thanks a lot for any answers! > > Maghi > -- > IMPORTANT: ISSCO has moved (from 28th September 1999) > AND my phone number keeps changing. Please check that you have the latest > one below. > > Maghi King | E-mail: Margaret.King@issco.unige.ch > TIM/ISSCO, ETI > University of Geneva | WWW: http://issco-www.unige.ch/ > 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 > CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From esteam@otenet.gr Mon Feb 19 11:47:07 2001 From: esteam@otenet.gr (ESTeam) Date: Mon, 19 Feb 2001 13:47:07 +0200 Subject: [MT-List] Big software companies and MT References: <3A8FDF8C.432DB0B@issco.unige.ch> Message-ID: <3A9107BB.AAAC70CA@otenet.gr> Dear Maghi and all, I can report serious growth and I think if you check how much personnel the translation software developers are adding in the last couple of years you will understand that translation software (I don't like either MT or TM as a term, it certainly doesn't fit our product) is a fast growing market and by no means surviving off research funds. In fact people like me find it completely wrong that industry is sponsored by any public funds and feel sometimes that I'm being forced to apply, especially with the new EU call on E-Content. My view is that translation software Research should be in universities, since the issues that aren't solved are so many. All evaluation at this stage is very local - no one can be on top of things that are happening anymore and developers like me have no wish to inform the world about what they are doing and there is no feel that this will provide market exposure. The main issue is that things are getting serious, official evaluations warn our competition as to what we are doing and we judge it to be more important to avoid this. You ask about large companies, yes, very big companies are seriously interested in translation software (even from the development perspective, not only the user prespective like SAS). My contacts like their privacy as I do and thus I will not name them. Finally, the L&H crisis has had no negative effect, on the other hand I think that the publicity they managed to provide to the market before the crisis was very beneficial and perhaps the main reason for the upswing we are getting. The integration of Systran into Web engines is another factor. No one is laughing at bad translation jokes anymore and we have internet companies standing in line to become multilingual. Finally, stories like our own, with fully automatic translation solutions into and from all the EU languages helps. Our old system has now been producing translations daily for almost 3 years - and there isn't a IPR lawyer globally that hasn't received our translations at one time or another. As for news about us without revealing any secrets - our new system has a final delivery to the EU next week and will be in production by May - touch wood :-)) - so even the EU has 3 contracted translation software systems inside its walls - Systran, Trados and ESTeam Translator. Our staff is three times bigger than February 2000. hopefully see you at the MT-Summit Gudrun Magnusdottir ESTeam AB Maghi King wrote: > Hello Everybody! > > As some of you know, I'm working on a project that involves assessing the > present state of the art in the use of technology to support translation > and localisation. > > One thing that has struck me is that some big companies are getting > involved in different ways: translation buttons appearing all over the > place, for example, or IBM re-launching its TM system and launching a > commercial MT system. Does anybody have a feeling that there is in general > a growing interest amongst the big software companies, or even know of > interesting recent developments? Maybe L&H's problems will put them off? > > Thanks a lot for any answers! > > Maghi > -- > IMPORTANT: ISSCO has moved (from 28th September 1999) > AND my phone number keeps changing. Please check that you have the latest > one below. > > Maghi King | E-mail: Margaret.King@issco.unige.ch > TIM/ISSCO, ETI > University of Geneva | WWW: http://issco-www.unige.ch/ > 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 > CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From Steven.Krauwer@let.uu.nl Mon Feb 19 14:14:31 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Mon, 19 Feb 2001 15:14:31 +0100 (MET) Subject: [MT-List] CfP: MT 2010 -- Towards a Road Map for MT (MT Summit Workshop) Message-ID: <200102191414.PAA01872@salto.let.uu.nl> Call for Workshop Contributions MT 2010 -- Towards a Road Map for MT September 19, 2001, afternoon Workshop organised by ELSNET and hosted by MT Summit VIII Santiago de Compostela, Spain September 18-22 2001 Aim and context: The aim of the workshop is to contribute to ELSNET's ongoing action to establish a road map for MT for the next decade. A road map comprises an analysis of the present situation, a vision of where we want to be in ten years from now, and a number of intermediate milestones that would help in setting intermediate goals and in measuring our progress towards our goals. The function of the road map is not to impose anything on anyone, but rather to provide a broadly supported definition of a context in which to position the MT community's efforts, which would allow us to identify common priorities for joint activities in e.g. research, resources and training. Scope: Papers are invited that * give critical analyses of the present state of the art in machine translation of written and spoken language, * present visions of the future of machine translation, both from a technological and from an application point of view, or * identify major milestones and challenges on our way towards the future, and/or ways to measure our progress along the road. Format: It will be a half-day workshop, with three sessions: * The first session will aim at giving a critical analysis of the present situation. * The second session will be dedicated to visions of the future. * The third session will aim at establishing intermediate milestones (e.g. timelines, necessary research achievements), methods to evaluate progress along the road, and training requirements to be met in order to reach our goals. Audience: The primary audience consists of people with an analytical or future oriented, programmatic interest, both from research and from industry. Workshop schedule: * 15 April 2001: Submission deadline * 31 May 2001: Notification of acceptance * 15 July 2001: Submission of final versions Submission details: Papers should be sent in electronic form to steven.krauwer@elsnet.org. For the submission format we follow the same guidelines as the main conference: http://www.eamt.org/summitVIII/author.html Core Programme/Organisation Committee: * Steven Krauwer (ELSNET, Netherlands) Chair * Niels Ole Bernsen (NIS, Denmark) * Paul Heisterkamp (Daimler-Chrysler, Germany) * Jose Pardo (UPM, Spain) * Pierre Isabelle (XEROX, France) * Arjan van Hessen (IP Globalnet, Netherlands) * Antonio Zampolli (U. Pisa, Italy) * Hans Uszkoreit (DFKI, Germany) * Oliviero Stock (IRST, Italy) * Susan Armstrong (ISSCO/ETI, Switzerland) * Herman Caeyers (LANT, Belgium) Contact details: Steven Krauwer email: steven.krauwer@elsnet.org ELSNET / UiL OTS www: http://www.elsnet.org Trans 10 phone: +31 30 253 6050 3512 JK Utrecht, NL fax: +31 30 253 6000 From Margaret.King@issco.unige.ch Mon Feb 19 19:26:03 2001 From: Margaret.King@issco.unige.ch (Maghi King) Date: Mon, 19 Feb 2001 20:26:03 +0100 Subject: [MT-List] Thank you all. Message-ID: <3A91734B.3882C483@issco.unige.ch> Lots of people have replied to me about software companies getting involved in MT. For a while I managed to keep up with thanking everybody individually, but eventually couldn't. So please, if you have not had a thank you from me, accept it now. I am very grateful to you all. As a proper thank you, when I can I will try to make a synopsis of the replies and put it on the list, avoiding what I think anybody might consider sensitive material: they were all very interesting. Maghi -- IMPORTANT: ISSCO has moved (from 28th September 1999) AND my phone number keeps changing. Please check that you have the latest one below. Maghi King | E-mail: Margaret.King@issco.unige.ch TIM/ISSCO, ETI University of Geneva | WWW: http://issco-www.unige.ch/ 40 blvd du Pont d'Arve | Tel: +41/22/705 87 55 CH-1211 GENEVA 4 (Switzerland) | Fax: +41/22/781 86 89 From stopping@rochester.rr.com Mon Feb 19 21:01:58 2001 From: stopping@rochester.rr.com (Suzanne Topping) Date: Mon, 19 Feb 2001 16:01:58 -0500 Subject: [MT-List] Big software companies and MT References: <3A8FDF8C.432DB0B@issco.unige.ch> Message-ID: <00c001c09ab7$350e5bc0$351d5d18@rochester.rr.com> ----- Original Message ----- From: "Maghi King" > One thing that has struck me is that some big companies are getting > involved in different ways: translation buttons appearing all over the > place, for example, or IBM re-launching its TM system and launching a > commercial MT system. Does anybody have a feeling that there is in general > a growing interest amongst the big software companies, or even know of > interesting recent developments? Maybe L&H's problems will put them off? I think there are two major factors related to increased interest in automated translation methods. The first is the continued massive growth in internet use for sales, support, and infrastructure. Companies can reach customers, suppliers, and employees located around the world faster and easier than ever before. Since the web is designed for global access, posting any item makes in (theoretically) available to the entire world. No more printing costs, shipment of marketing materials etc. to reach regional customers. Given the shrinking world and globally expanding economy, most big companies -must- reach customers around the world, and find that the web helps save money in some areas. But localization for reaching those targets is still required, and dealing with multilingual web sites is complex. Workflow is a huge issue. Creation of tools to help deal with this complexity is on the rise. Money saved in other ways must now be spent on managing the multilingual sites, so optimization methods are in high demand. The second issue is the emergence of XML and related standards. XML will be one of the key technologies which will drive the development of all CAT tools, web workflow tools, and all web-based software applications. XML and it's accompanying technologies effectively performs the task of separating content/data from all delivery and processing components. Since separation of code and content is a huge piece to the localization puzzle, ongoing developments in this arena are helping to make localization processes more efficient and affordable. So developing automated methods is becoming easier. I therefore believe that some of the primary reasons you are seeing an increase of interest by large companies is that the technology timing is right. Market and business imperatives demand it, and the technology is getting closer and closer to being able to provide it. Just my two cents. Suzanne --++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Suzanne Topping BizWonk Inc. (TM) Solutions for a Global E-conomy Phone: +1 716.454.4210 Fax: +1 716.454.4213 Email: stopping@bizwonk.com (Send me an email to join the North East Localization Special Interest Group, an email distribution list which acts as a discussion forum for localization issues.) From theo@softissimo.com Mon Feb 19 13:12:55 2001 From: theo@softissimo.com (=?iso-8859-1?Q?Th=E9o_Hoffenberg?=) Date: Mon, 19 Feb 2001 14:12:55 +0100 Subject: [MT-List] new translation software and service : Reverso.net Reverso Pro Message-ID: <03b101c09a75$d1a6e200$1401a8c0@SOFTISSIMO> C'est un message de format MIME en plusieurs parties. ------=_NextPart_000_03AC_01C09A7E.0D9B1D00 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear All, A new free service, based on Reverso software is now available and = allows translation of texts, e-mails, web pages, and it offers a = brand-new "My Translator" : a popup window always present which allows = you to translate words, phrases or full sentences from any site or = application. You can see it on www.reverso.net and get more information on the = underlying technology at www.softissimo.com=20 =20 For translators (professional or occasional), the appropriate tool is = Reverso Pro, a PC software which includes powerful features for = terminology management, layout management, post-editing... see details = on www.softissimo.com=20 Best regards Th=E9o Hoffenberg Softissimo 7 rue Auber, Paris tel(331)43128505 fax(331)42661154 email : theo@softissimo.com ------=_NextPart_000_03AC_01C09A7E.0D9B1D00 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Dear All,
 
A new free service, based on Reverso software is now = available=20 and allows translation of texts, e-mails, web pages, and it offers a = brand-new=20 "My Translator" : a popup window always present which allows you to = translate=20 words, phrases or full sentences from any site or = application.
 
You can see it on www.reverso.net and get more = information on=20 the underlying technology at www.softissimo.com
 
For translators (professional or occasional), the = appropriate=20 tool is Reverso Pro, a PC software which includes powerful features for=20 terminology management, layout management, post-editing... see details = on www.softissimo.com 
 
Best regards
 
Th=E9o Hoffenberg
Softissimo
7 rue Auber,=20 Paris
tel(331)43128505
fax(331)42661154
email : theo@softissimo.com
<= /FONT>
------=_NextPart_000_03AC_01C09A7E.0D9B1D00-- From ref@cs.cmu.edu Mon Feb 19 18:22:01 2001 From: ref@cs.cmu.edu (Robert Frederking) Date: Mon, 19 Feb 2001 13:22:01 -0500 Subject: [MT-List] Big software companies and MT In-Reply-To: Your message of Sun, 18 Feb 2001 16:18:11 -0500. <4F2457EC921ED31188C20090276DC8C402553628@merc02.us.sas.com> Message-ID: <9623.982606921@euclid.mt.cs.cmu.edu> > The ISLE framework is interesting, but what I need/want is not just a > list of criteria and considerations, but the facts as they pertain to > the different vendors' products. I think -that- information is a long > way from being made readily available. This is an interesting point. I have run into this problem in trying to teach our graduate MT course's lecture on "Commercial MT systems". It's very difficult to get concrete, useful information from MT vendors (other than that "ours is the best, you should buy it"). Since (it seems to me) any such information will necessarily be all wrapped up in marketting issues, I wonder how it would be possible to get straight, reliable information on a range of companies. (That is, every company clearly wants to look like they are the best.) I know there has been some discussion of some kind of Consumer Reports for MT, but this is bound to be expensive to do (independently) in a serious way. I find myself wondering whether anything like this exists for other commercial software fields. I suspect not, actually. After all, any serious quality comparison would be damaging to Microsoft. :-) So, is there any reliable, independent assessment of commercial software in other fields? If so, how do they manage to do it? (Or is it perhaps easier to evaluate other software copared to MT systems?) Bob -- Robert E. Frederking Senior Systems Scientist Language Technologies Institute/Center for Machine Translation Carnegie Mellon University 5000 Forbes Avenue Telephone: +1-412-268-6656 Pittsburgh, PA 15213 USA FAX: +1-412-268-6298 Email: ref@cs.cmu.edu WWW: http://www.cs.cmu.edu/~ref/ From KurtGodden@excite.com Mon Feb 19 23:56:08 2001 From: KurtGodden@excite.com (Kurt Godden) Date: Mon, 19 Feb 2001 15:56:08 -0800 (PST) Subject: MT Evaluation, Was: [MT-List] Big software companies and MT Message-ID: <15867308.982626968711.JavaMail.imail@batty.excite.com> This is a big topic, much discussed. Of course, to do an evaluation of MT or any other s/w one needs at least: 1. A decision on what dimensions of quality or performance or usability one is interested in. 2. A metric for each of those dimensions. No doubt one important dimension for any MT system is the "quality" of the raw translation, which is itself a veritable Pandora's Box of issues. Nonetheless, many people and organizations have developed translation quality metrics. I was involved in one for the Society of Automotive Engineers. But I think there are other important dimensions to consider as well. And just because it is difficult to arrive at a decision about what is valuable to measure, and then how to measure those things, it does not follow that it is impossible or not worth the effort. In fact, I think the value is well worth the high effort involved. And I *know* that it is possible to develop a reasonably objective metric. Kurt Godden Director of Linguistics JustTalk, Inc. Ann Arbor, MI www.justtalk.com On Mon, 19 Feb 2001 13:22:01 -0500, Bob Frederking ref@cs.cmu.edu wrote: >I know > there has been some discussion of some kind of Consumer Reports for > MT, but this is bound to be expensive to do (independently) in a > serious way. > > I find myself wondering whether anything like this exists for other > commercial software fields. I suspect not, actually. After all, any > serious quality comparison would be damaging to Microsoft. :-) > > So, is there any reliable, independent assessment of commercial > software in other fields? If so, how do they manage to do it? (Or is > it perhaps easier to evaluate other software copared to MT systems?) > > Bob > -- > Robert E. Frederking Senior Systems Scientist > Language Technologies Institute/Center for Machine Translation > Carnegie Mellon University > 5000 Forbes Avenue Telephone: +1-412-268-6656 > Pittsburgh, PA 15213 USA FAX: +1-412-268-6298 > Email: ref@cs.cmu.edu WWW: http://www.cs.cmu.edu/~ref/ > > -- > For MT-List info, see http://www.eamt.org/mt-list.html _______________________________________________________ Send a cool gift with your E-Card http://www.bluemountain.com/giftcenter/ From LCR@oce.nl Tue Feb 20 13:06:08 2001 From: LCR@oce.nl (Cremers LMG) Date: Tue, 20 Feb 2001 14:06:08 +0100 Subject: [MT-List] Big software companies and MT Message-ID: Perhaps one way to achieve to some extent a reasonable comparison between MT systems is, to select a group of current MT users with different systems, have them run a test set designed for this purpose, under specific conditions. This shouldn't be too expensive to do. Lou Cremers -------------------------------------------------------- Océ Technologies B.V. ITC - Translation Services name: Lou Cremers Dr Blumenkampstr 3 mailto:lcr@oce.nl 5914 PV Venlo Tel: +31 (0)77-359 34 44 The Netherlands Fax: +31 (0)77-359 54 27 http://www.oce.com > -----Original Message----- > From: Robert Frederking > [mailto:Robert_Frederking@mlist-3.sp.cs.cmu.edu] > Sent: Monday, February 19, 2001 7:22 PM > To: John Kohl > Cc: 'MT-list@eamt.org' > Subject: Re: [MT-List] Big software companies and MT > > > > The ISLE framework is interesting, but what I need/want is > not just a > > list of criteria and considerations, but the facts as they > pertain to > > the different vendors' products. I think -that- information > is a long > > way from being made readily available. > > This is an interesting point. I have run into this problem in trying > to teach our graduate MT course's lecture on "Commercial MT systems". > It's very difficult to get concrete, useful information from MT > vendors (other than that "ours is the best, you should buy it"). > > Since (it seems to me) any such information will necessarily be all > wrapped up in marketting issues, I wonder how it would be possible to > get straight, reliable information on a range of companies. (That is, > every company clearly wants to look like they are the best.) I know > there has been some discussion of some kind of Consumer Reports for > MT, but this is bound to be expensive to do (independently) in a > serious way. > > I find myself wondering whether anything like this exists for other > commercial software fields. I suspect not, actually. After all, any > serious quality comparison would be damaging to Microsoft. :-) > > So, is there any reliable, independent assessment of commercial > software in other fields? If so, how do they manage to do it? (Or is > it perhaps easier to evaluate other software copared to MT systems?) > > Bob > -- > Robert E. Frederking Senior Systems Scientist > Language Technologies Institute/Center for Machine Translation > Carnegie Mellon University > 5000 Forbes Avenue Telephone: +1-412-268-6656 > Pittsburgh, PA 15213 USA FAX: +1-412-268-6298 > Email: ref@cs.cmu.edu WWW: http://www.cs.cmu.edu/~ref/ > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > > From Giuseppe@efoodmanager.com Mon Feb 26 09:55:55 2001 From: Giuseppe@efoodmanager.com (Giuseppe Gentile) Date: Mon, 26 Feb 2001 10:55:55 +0100 Subject: [MT-List] Big software companies and MT Message-ID: Dear Ms. Maghi, I have asked recently to Lernout and Hauspie to make an offer regarding MT on our portal. My request was very general and we did not proceeded further. I work in a b2b company and we are "potential customers" of MT and translation memory... . I assume in the future there will be a higher request of MT applied on portals and consequently an increased request of this type of service to providers like ibm and others. However, in my opinion, this development is still not very near... . In fact we have to face a quantity of problems related to the coordination of multilingual information first. Moreover, MT has a specific place on the scene of ecommerce, it can be applied to non critical content and offered as value added service... . Nonetheless this technological development and request will be necessary at a later stage, when company will have to offer this "added" service to their portals to remain competitive. I believe one or two years and the time will be mature enough... for the moment, I think, we are still on the ground of an "hypothetical" use of MT on portals... . I hope this may help have a nice day. Giuseppe Giuseppe Gentile Language Manager efoodmanager AG "e-commerce for the food industry" Phone +49 (0) 211 43 90 3 520 Fax +49 (0) 211 43 90 3 100 www.efoodmanager.com From gojol@sunu.rnc.ro Mon Feb 26 14:28:04 2001 From: gojol@sunu.rnc.ro (Vlad V. Gojol) Date: Mon, 26 Feb 2001 14:28:04 +0000 Subject: [MT-List] Advanced Parser & Tagger Message-ID: <1.5.4.32.20010226142804.0092bbc4@u3.ici.ro> Dear Readers , Those interested in the two pieces of software presented below or / and by the possibility of building a translating system to Romanian ( e.g. from English ) are kindly asked to contact me per- sonally ( gojol@sunu.rnc.ro ) . Thank you , Dr.ing. Vlad Gojol ---------------------------------- Senior Research Engineer Institutul National de Informatica Bucuresti , Romania ............................................................................ LANGUAGE INDEPENDENT PART-OF-SPEECH TAGGER I created a part-of-speech tagger with an unusual capacity of dealing with large contexts , especially for German . I used Negra ( seemingly the best known German corpus , with free obtainable licence ) . The tagger currently reputed as being the most accurate for German is perhaps TnT . It reports upon this corpus an error rate of 3.4% . But I have found a syste- matic error in Negra : all the occurences of the auxilliary verbs are tagged as auxilliary ( VAFIN ) , though in 50% of the cases they function as finite verbs ( VVFIN ) . I corrected a part of the corpus ( cca 40,000 tokens ) . In this more correct environment ( where the performance of TnT should be probably around 4.5% ) , my tagger gets 1.7% . On another German corpus ( I call it X ) , with comparable contents ( news- paper articles ) and tagset , but with attached EXTERIOR lexicon ( i.e. not extracted from the corpus ) , the result is 2.4% . I also used Susanne ( the only English corpus I could get free ) . The re- ported result for TnT is 3.8% . Mine is 2.8% . On the "A" texts , best paral- lelable with those in Negra , as journalistic , it's 2.3% . By restricting the tagset to a more normal size ( cca 100 tags , determined as optimal af- ter lots of test runs ) , it's 1.3% . Initially I had used a Romanian corpus , with a result of 0.9% ( compared to 1.7% , 2.5% and 4.2% respectively got by the Xerow , Birmingham and Brill taggers ) . The speed is comparable to that of TnT and modifiable by parameter setting , in reverse proportion to the accuracy ( but without affecting it much ) . The incremental operating mode and the data structures segmentation allow running on very small memory computers . There is the advantage of an intuitive output ( no hostile binary matrix ) , in a form analogue to the input of some expert systems . Alternative taggings are output , with scores : unlike with other taggers , they don't refer to individual words , but to whole sentence parts ( repre- senting somehow phrase surfaces of minimum energy ) . Special facilities exist , such as virtual tags , or context essentialisa- tion ( permitting to get the minimal contexts set characteristic to a certain linguisic style , useful not only for maximum accuracy and speed ) etc. Recently added features : wide-character support , the possibility of being called as a simple library routine , or suspending the notion of a file ( a complex files system is emulated into the memory ) . For example ( for the second feature above ) , you can create an instance of the tagger ( let's say for English with the tagset Lancaster ) , call it to tag a cer- tain text buffer ( by writing the resulting tags into another buffer ) and finally kill the instance , all this without using any disc file : t = GojolTagger_new("lancaster"); error_code = GojolTagger_tag(t,input_buffer,output_buffer); GojolTagger_free(t); All is built on two essentially new concepts : organicity and context pro- pagation . I didn't publish anything about them , to keep up their commercial appeal . The accuracy comparable to that of manual tagging made me find many errors in the used corpora : 98 in Negra , 36 in Susanne ; Prof. G. Sampson replied gratefully , saying that it's the first time somebody reports more than 2 errors , and that my findings make necessary a new version of Susanne . The handling of very large contexts could even modify the current tagsets de- sign , by cancelling some unnatural decisions ( motivated only by the incapa- city of the existing taggers to see beyond a 3-tokens neighborhood ) , such as those concerning the auxilliary verbs , participles etc. - so removing some burden from the subsequent stages of text processing . It is written in C ( Linux ) . Demos for German ( Negra ) and English ( Susanne ) are available . .............................................................................. LANGUAGE INDEPENDENT STATISTIC PARSER After learning from a 46,000 words pos-tagged corpus and a 32,000 words parsed ( treebank ) corpus , a 2,000 words text ( not included in any of the two corpora ) is parsed ( tagging excluded ) in 6 seconds ( on a 200 MHz machine ) with 2% incomplete trees ( but for these declared failures , are also provided well formed trees sufficient for a subse- quent translator ) - the extracted grammar having cca 12,000 rules . The Negra corpus of German was used . After learning from a 17,000 words parsed corpus and from the same 46,000 words pos-tagged one , a 2,000 words text included into the first ( but excluded from the second ) , to warrant that the grammar is complete relative to it ( i.e. contains all the rules necessary for its correct parsing ) , is processed in 2 seconds with no incomplete tree - the extracted grammar having cca 7,000 rules . The system is language independent - for English , upon the Susanne corpus , comparable results are obtained . To have an acceptable parser for any other i- diom , you need essentially simply a corpus with 30,000 tag- ged words , from which only 20,000 parsed as well - and for optimal results , 50,000 and 30,000 respectively . The parser may accept a set of rules intended to modify the statistical grammar deduced from the corpus . Moreover , it can take as input only a context-free grammar ( in which case it ceases to be a statistical parser ) , but in this operating mode it requires much time and memory ( during the learning , not during the parsing as such ) if the grammar is over-dimensioned . The statistical grammar is refined not by simply adding the proposed rules , but by modifying the corpus , to exploit all the real contexts possible for them . Semantic processing could be easily inserted at rule re- duction points . Actually this generalized parser can also work as a compiler generator : by appending specific semantic routines , you get efficient compilers for C , Pascal etc. This versatile system has more than 40 parameters which tune the accuracy and speed according to the target language sam- ple . The output is in treebank format and optionally in gra- phic ( with the trees effectively drawn ) one . Linux demos exist for German and English . As only the mi- nimal definition of C is used , it is easily adaptable to any machine ( for other Unix-like operating systems , probably a simple recompilation would be sufficient ) . From gojol@sunu.rnc.ro Mon Feb 26 14:37:47 2001 From: gojol@sunu.rnc.ro (Vlad V. Gojol) Date: Mon, 26 Feb 2001 14:37:47 +0000 Subject: [MT-List] Advanced Parser & Tagger Message-ID: <1.5.4.32.20010226143747.00922ed8@u3.ici.ro> Dear Sir , Please tell me whether would it be possible inserting within your publication a presentation of the two pieces of software mentioned below , without disclosing their theoretical principles ( I could rewrite according to your guidelines , but still omit- ting theoretical explanations ) . Best regards , Vlad Gojol ---------------------------------- Senior Research Engineer Institutul National de Informatica Bucuresti , Romania ............................................................................ LANGUAGE INDEPENDENT PART-OF-SPEECH TAGGER I created a part-of-speech tagger with an unusual capacity of dealing with large contexts , especially for German . I used Negra ( seemingly the best known German corpus , with free obtainable licence ) . The tagger currently reputed as being the most accurate for German is perhaps TnT . It reports upon this corpus an error rate of 3.4% . But I have found a syste- matic error in Negra : all the occurences of the auxilliary verbs are tagged as auxilliary ( VAFIN ) , though in 50% of the cases they function as finite verbs ( VVFIN ) . I corrected a part of the corpus ( cca 40,000 tokens ) . In this more correct environment ( where the performance of TnT should be probably around 4.5% ) , my tagger gets 1.7% . On another German corpus ( I call it X ) , with comparable contents ( news- paper articles ) and tagset , but with attached EXTERIOR lexicon ( i.e. not extracted from the corpus ) , the result is 2.4% . I also used Susanne ( the only English corpus I could get free ) . The re- ported result for TnT is 3.8% . Mine is 2.8% . On the "A" texts , best paral- lelable with those in Negra , as journalistic , it's 2.3% . By restricting the tagset to a more normal size ( cca 100 tags , determined as optimal af- ter lots of test runs ) , it's 1.3% . Initially I had used a Romanian corpus , with a result of 0.9% ( compared to 1.7% , 2.5% and 4.2% respectively got by the Xerow , Birmingham and Brill taggers ) . The speed is comparable to that of TnT and modifiable by parameter setting , in reverse proportion to the accuracy ( but without affecting it much ) . The incremental operating mode and the data structures segmentation allow running on very small memory computers . There is the advantage of an intuitive output ( no hostile binary matrix ) , in a form analogue to the input of some expert systems . Alternative taggings are output , with scores : unlike with other taggers , they don't refer to individual words , but to whole sentence parts ( repre- senting somehow phrase surfaces of minimum energy ) . Special facilities exist , such as virtual tags , or context essentialisa- tion ( permitting to get the minimal contexts set characteristic to a certain linguisic style , useful not only for maximum accuracy and speed ) etc. Recently added features : wide-character support , the possibility of being called as a simple library routine , or suspending the notion of a file ( a complex files system is emulated into the memory ) . For example ( for the second feature above ) , you can create an instance of the tagger ( let's say for English with the tagset Lancaster ) , call it to tag a cer- tain text buffer ( by writing the resulting tags into another buffer ) and finally kill the instance , all this without using any disc file : t = GojolTagger_new("lancaster"); error_code = GojolTagger_tag(t,input_buffer,output_buffer); GojolTagger_free(t); All is built on two essentially new concepts : organicity and context pro- pagation . I didn't publish anything about them , to keep up their commercial appeal . The accuracy comparable to that of manual tagging made me find many errors in the used corpora : 98 in Negra , 36 in Susanne ; Prof. G. Sampson replied gratefully , saying that it's the first time somebody reports more than 2 errors , and that my findings make necessary a new version of Susanne . The handling of very large contexts could even modify the current tagsets de- sign , by cancelling some unnatural decisions ( motivated only by the incapa- city of the existing taggers to see beyond a 3-tokens neighborhood ) , such as those concerning the auxilliary verbs , participles etc. - so removing some burden from the subsequent stages of text processing . It is written in C ( Linux ) . Demos for German ( Negra ) and English ( Susanne ) are available . .............................................................................. LANGUAGE INDEPENDENT STATISTIC PARSER After learning from a 46,000 words pos-tagged corpus and a 32,000 words parsed ( treebank ) corpus , a 2,000 words text ( not included in any of the two corpora ) is parsed ( tagging excluded ) in 6 seconds ( on a 200 MHz machine ) with 2% incomplete trees ( but for these declared failures , are also provided well formed trees sufficient for a subse- quent translator ) - the extracted grammar having cca 12,000 rules . The Negra corpus of German was used . After learning from a 17,000 words parsed corpus and from the same 46,000 words pos-tagged one , a 2,000 words text included into the first ( but excluded from the second ) , to warrant that the grammar is complete relative to it ( i.e. contains all the rules necessary for its correct parsing ) , is processed in 2 seconds with no incomplete tree - the extracted grammar having cca 7,000 rules . The system is language independent - for English , upon the Susanne corpus , comparable results are obtained . To have an acceptable parser for any other i- diom , you need essentially simply a corpus with 30,000 tag- ged words , from which only 20,000 parsed as well - and for optimal results , 50,000 and 30,000 respectively . The parser may accept a set of rules intended to modify the statistical grammar deduced from the corpus . Moreover , it can take as input only a context-free grammar ( in which case it ceases to be a statistical parser ) , but in this operating mode it requires much time and memory ( during the learning , not during the parsing as such ) if the grammar is over-dimensioned . The statistical grammar is refined not by simply adding the proposed rules , but by modifying the corpus , to exploit all the real contexts possible for them . Semantic processing could be easily inserted at rule re- duction points . Actually this generalized parser can also work as a compiler generator : by appending specific semantic routines , you get efficient compilers for C , Pascal etc. This versatile system has more than 40 parameters which tune the accuracy and speed according to the target language sam- ple . The output is in treebank format and optionally in gra- phic ( with the trees effectively drawn ) one . Linux demos exist for German and English . As only the mi- nimal definition of C is used , it is easily adaptable to any machine ( for other Unix-like operating systems , probably a simple recompilation would be sufficient ) . From gojol@sunu.rnc.ro Mon Feb 26 14:43:22 2001 From: gojol@sunu.rnc.ro (Vlad V. Gojol) Date: Mon, 26 Feb 2001 14:43:22 +0000 Subject: [MT-List] Advanced Parser & Tagger Message-ID: <1.5.4.32.20010226144322.0093f500@u3.ici.ro> Dear List Members , Sorry for the second inadvertent message . Vlad Gojol From Steven.Krauwer@let.uu.nl Tue Feb 27 14:52:20 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Tue, 27 Feb 2001 15:52:20 +0100 (MET) Subject: [MT-List] CfP: ACL2001 Workshop on Sharing Tools and Resources Message-ID: <200102271452.PAA25821@salto.let.uu.nl> CALL FOR CONTRIBUTIONS ACL/EACL Workshop on Sharing Tools and Resources for Research and Education Co-organised by ELSNET Toulouse, Saturday 7th July 2001 BACKGROUND: At a workshop at ACL 2000 in Hong Kong dedicated to Infrastructures for Global Collaboration there was an agreement between the main professional organisations in NLP and Speech (ACL and ISCA), and ELSNET, and the other meeting participants, that it would be useful to aim at a broadly supported, joint repository or catalogue for tools and materials for the language and speech communities. An ELSNET-sponsored workshop on educational issues held at EACL99 concluded that certain non-transient infrastructures needed to be instigated to raise the public perception of educational issues in NLP. It also concluded that a repository of shared materials, appropriately indexed for educational usage, would be a useful point of departure. This workshop will build on the consensus reached at these previous workshops. There will be two clear foci: one upon instruments for sharing tools and resources in general that addresses practical problems, and the other upon the technological and infrastructural issues surrounding the educational uses of repositories. Good examples of existing initiatives in this area are among others the ACL Natural Language Software Registry (hosted at DFKI, registry.dfki.de) which was set up as a repository for tools for the distinct fields of Human Language Techology (HLT), the ELRA/ELDA, LDC, TELRI and Elsnet resources catalogues and repositories (www.icp.inpg.fr/ELRA, www.ldc.upenn.edu, www.telri.de and www.elsnet.org/resources.html), OLAC (a worldwide network of language archives at www.language-archives.org), and JEWELS (www.elsnet.org/jewels), an as-yet incomplete EU funded website for educational materials in Language and Speech. A third theme concerns how to build upon existing initiatives as sources of data or inspiration. AIM AND SCOPE OF THE WORKSHOP: The main goal of the workshop is to discuss methods for the improvement and extension of existing repositories; the educational uses of repositories; the closer interlinking between different kinds of repositories (tools and resources); global infrastructures for the achievement of joint actions. However, we expect the scope of the workshop to be much wider than that, as the issues addressed are of general interest to everybody who believes that sharing tools and resources is essential for the progress of research and education in our field. Contributions of papers and demonstrations are solicited that address the above themes. The following list of topics is suggestive rather than exhaustive: * Repositories versus catalogues * Mechanisms and infrastructures for sharing and describing content * Repository management * Standards for exchange, description, and annotation * Metadata descriptions * Quality assessment * Structure and content of an NLP/CL repository * Tools and materials for NLP/CL education * Web-based teaching methods for NLP/CL SUBMISSION REQUIREMENTS FOR PAPERS * Electronic submissions only (PostScript, Word, or PDF), following the appropriate ACL latex style or Microsoft Word style. Submissions should not exceed eight (8) pages, including references. You can download the appropriate style or template files using the following link: acl2001.dfki.de/style. In case of problems with the submission format, please contact one of the co-chairs. * Submissions to either co-chair (Mike Rosner and Thierry Declerck). All submissions will be acknowledged. * Please provide a list of keywords in the separate header page and indicate the best fitting subtopic(s) from the above list. SUBMISSION REQUIREMENTS FOR DEMONSTRATIONS * Demos may be submitted with or without an accompanying paper. * Please write a 2-page description of the demo and send to either co-chair. Please let us know about special hardware requirements over and above the standard PC + beamer without internet access provided by default CONFIRMED COMMITTEE MEMBERS - Thierry Declerck (DFKI) Co-chair (Repository) declerck@dfki.de - Mike Rosner (Malta) Co-chair (Education) mros@cs.um.edu.mt - Steven Bird (U. Penn) - Bill Black (UMIST) (UMIST, Manchester, UK) wjb@co.umist.ac.uk - Gosse Bouma (University of Groningen) gosse@let.rug.nl - Koenraad de Smedt (University of Bergen) desmedt@uib.no - Claire Gardent (CNRS, Nancy) claire.gardent@loria.fr - Steven Krauwer (Utrecht University) steven.krauwer@elsnet.org - Donna Harman (NIST) - Julia Hirschberg (ATT, ISCA) - Jun'ichi Tsujii (Tokyo) - Andy Way (Dublin City University) away@compapp.dcu.ie - Antonio Zampolli (Univ. of Pisa) pisa@ilc.pi.cnr.it DEADLINES * Submission Deadline: 6th April 2001 * Notification Date: 27th April 2001 * Camera ready copy due: 16th May 2001 WORKSHOP URL http://www.elsnet.org/acl2001-tools.html FOR FURTHER INFORMATION CONTACT Michael Rosner mros@cs.um.edu.mt Thierry Declerck declerck@dfki.de From Steven.Krauwer@let.uu.nl Tue Feb 27 15:08:22 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Tue, 27 Feb 2001 16:08:22 +0100 (MET) Subject: [MT-List] CfP: ARABIC NLP Workshop at ACL2001 Message-ID: <200102271508.QAA26511@salto.let.uu.nl> ACL/EACL 2001 Workshop ARABIC Language Processing: Status and Prospects Toulouse, France, Friday 6 July 2001 Co-organized by: ELSNET NAPLUS WORKSHOP OBJECTIVES AND DESCRIPTION: The objective of the workshop is threefold. * First of all we want to bring together people who are actively involved in Arabic language and/or speech processing in a mono- or multilingual context, and give them an opportunity to report on completed and ongoing work as well as on the availability of products and core technologies. This should enable the participants to develop a common view on where we stand with respect to Arabic language processing. * Secondly, we want to identify problems of common interest, and possible mechanisms to move towards solutions, such as sharing of tools and resources, moving towards standards, sharing and dissemination of information and expertise, adoption of current best practices, setting up joint projects and technology transfer mechanisms, etc. * Third, we would like to enhance collaboration between the Arabic NLP community and the NLP community at large. The workshop program will include the following components: * Introduction * Overview talks * Scientific papers * Short presentations of projects, core technologies and products * A panel session and/or a round table discussion * Conclusions SUBMISSION REQUIREMENTS FOR SCIENTIFIC PAPERS: Papers are solicited that address all aspects of Arabic language processing in a mono- or multilingual context, including tools, resources and standards. Papers will have to be original and report on completed research. Submissions of scientific papers should not exceed eight (8) pages. Please provide a list of keywords in the separate header page. Further submission details below. SUBMISSION REQUIREMENTS FOR SHORT PRESENTATIONS OF PROJECTS, CORE TECHNOLOGIES AND PRODUCTS: Short presentations serve to give the audience an impression of ongoing activities and projects, and of existing core technologies and products, with a view to possible collaboration and synergies (i.e. NO commercial product presentations). Submissions of short presentations should not exceed two (2) pages. Short presentations will be reviewed on the basis of relevance and clarity of presentation. REQUIREMENTS FOR ALL SUBMISSIONS: Electronic submissions only (PostScript, Word, or PDF), following the appropriate ACL latex style or Microsoft Word style. Submissions should not exceed the length indicated above, including references. You can download the appropriate style or template files using the following link: http://acl2001.dfki.de/style/. Submission and presentation language is English. In case of problems with the submission format, please contact steven.krauwer@elsnet.org. Submissions should be sent to steven.krauwer@elsnet.org. All submissions will be acknowledged. DEADLINES FOR ALL SUBMISSIONS: * Submission deadline: 6th April 2001 * Notification date: 27th April 2001 * Camera-ready papers due: 16th May 2001 * Workshop date: 6th July 2001 CONFIRMED CORE PROGRAMME/ORGANISATION COMMITTEE: * Mustafa Yaseen, Amman University, Jordan (Co-chair, myaseen@cbj.gov.jo) * Joseph Dichy, Universite Lumiere-Lyon 2, France (Co-chair, dichy@univ-lyon2.fr) * Steven Krauwer, Utrecht University / ELSNET, The Netherlands (Contact person, steven.krauwer@elsnet.org) * Adnane Zribi, University of Tunis, Tunisia (adn@gnet.tn) * Salem Ghazali, IRSIT, Tunisia (ghazali@irsit.rnrt.tn) * Humoud Al-Sadoun, Ministry of Education, Kuwait (hbh@moe.edu.kw) * Jean Senellart, SYSTRAN, France (senellart@systran.fr) * Nadia Hegazy, ERI, Egypt (nhegazy@idsc.gov.eg) * Khalid Choukri, ELRA/ELDA, France (choukri@elda.fr) * Malek Boualem, FTRD/DMI/LAN, France (malek.boualem@rd.francetelecom.fr) * Everhard Ditters, University of Nijmegen, The Netherlands, (e.ditters@let.kun.nl) WORKSHOP URL: http://www.elsnet.org/acl2001-arabic.html CONTACT INFO: Steven Krauwer email: steven.krauwer@elsnet.org ELSNET / UiL OTS www: http://www.elsnet.org Trans 10 phone: +31 30 253 6050 3512 JK Utrecht, NL fax: +31 30 253 6000 From Steven.Krauwer@let.uu.nl Tue Feb 27 15:11:49 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Tue, 27 Feb 2001 16:11:49 +0100 (MET) Subject: [MT-List] CfP: ACL2001 Workshop on HLT and Knowledge Management Message-ID: <200102271511.QAA26735@salto.let.uu.nl> WORKSHOP ON HUMAN LANGUAGE TECHNOLOGY AND KNOWLEDGE MANAGEMENT ACL'2001 Conference Toulouse, France July 6-7, 2001 Human language technologies promise solutions to challenges in human computer interaction, information access, and knowledge management. Advances in technology areas such as indexing, retrieval, transcription, extraction, translation, and summarization offer new capabilities for learning, playing and conducting business. This includes enhanced awareness, creation and dissemination of enterprise expertise and know-how. This workshop aims to bring together the community of computational linguists working in a range of areas (e.g., speech and language processing, translation, summarization, multimedia presentation, content extraction, dialog tracking) both to report advances in human language technology, their application to knowledge management and to establish a road map for the Human Language Technologies for the next decade. The road map will comprise an analysis of the present situation, a vision of where we want to be in ten years from now, and a number of inter-mediate milestones that would help in setting intermediate goals and in measuring our progress towards our goals. The workshop will be structured into two days, the first which will address new research in human language tech-nology for knowledge management that addresses problems including but not limited to: * Expert Discovery: Modeling, cataloguing and tracking of distributed organizations and communities of experts. * Knowledge Discovery: Identification and classification of knowledge from unstructured multimedia data. * Knowledge Sharing: Awareness of and access to enterprise expertise and know-how. Human language technology promises solutions to these challenges through technologies such as: * Automated retrieval, extraction, and enrichment of information and knowledge from multimedia, multilin-gual, and multiparty information sources. * Translingual or crosslingual retrieval, presentation, and sharing of knowledge. * Automated detection and tracking of emerging topics from unstructured multimedia data (e.g., documents, web, video news broadcasts). * Use of knowledge sources to facilitate knowledge mapping and access (e.g., lexicosemantic such as Word-Net, semantic such as geospatial Gazetteers, semistructured such as thesauri, encyclopedia, fact books) * Automated question-answering from heterogeneous source * Intelligent tools that support the automated bibliometrics and document analysis/understanding in support of discovery of distributed experts and communities of expertise * Summarization and presentation generation of knowledge (e.g., knowledge maps, lessons learned). * Modeling of user knowledge, beliefs, plans, (dis)abilities and preferences from queries, created artifacts, and human computer interactions. The second day of the workshop will target the formulation and refinement of a road map for the Human Language Technologies for the next decade. Participants will help formulate grand challenge problems, discuss possible data sets and/or evaluation metrics/methods that could form the basis of more scientific methods, articulate the role of and necessary advances in human language technology to solve these challenges, as well as identify and characterize early innovations and issues (e.g., robustness, scalability, ontology, privacy). PROGRAM COMMITTEE * Dr. Mark Maybury (Chair), The MITRE Corporation, maybury@mitre.org * Niels Ole Bernsen (Co-chair), University of Southern Denmark, nob@nis.sdu.dk * Steven Krauwer, ELSNET, U. Utrecht, steven.krauwer@let.uu.nl * Irma Becerra-Fernandez, Florida International University, becferi@fiu.edu * Paul Heisterkamp, Daimler-Chrysler Research Ulm, paul.heisterkamp@daimlerchrysler.com * Arjan van Hessen, COMSYS / U. Twente, hessen@cs.utwente.nl * Pierre Isabelle, XEROX Grenoble, pierre.isabelle@xrce.xerox.com * Enrico Motta, The Open University, e.motta@open.ac.uk * Jose Pardo, ELSNET, Univ.Politecnica Madrid, pardo@die.upm.es * Oliviero Stock, IRST Trento, stock@itc.it * Henry Thompson HCRC LTG, University of Edinburgh, ht@cogsci.ed.ac.uk * Hans Uszkoreit, DFKI Saarbruecken, uszkoreit@dfki.de * Yorick Wilks, University of Sheffield, yorick@dcs.shef.ac.uk * Rick Wojcik, Boeing Phantom Works, richard.h.wojcik@boeing.com * Antonio Zampolli, ELSNET, U. Pisa, pisa@ilc.pi.cnr.it TARGET AUDIENCE The target audience of the workshop includes active researchers, developers, appliers/entrepreneurs and funders of human language technology in general as well as how it is applied to knowledge management applications. While we project a high degree of interest in this topic, we intend to restrict attendance based upon the quality of paper submissions to foster high quality interchange and progress. SUBMISSION FORMAT AND INSTRUCTIONS Both papers and demonstration submissions are encouraged, either on HLT in general or its application to KM systems. Papers targeted at the first day on HLT for KM should clearly articulate the knowledge management problem addressed, the technical approach to solving that, the novelty of the approach, its relation to previous work, the evaluation or performance of the system or method, and discussion of limitations. Papers targeted at the second day of on human language technology direction should be authored so they could be integrated into a more general HLT roadmap and so should include a definition of the HLT area addressed (e.g., information ex-traction, translation, speech recognition), a statement of the grand challenges or problems in the subfield, an ar-ticulation/analysis of the current state of the art, a vision of where the community wants to be in ten years from now, a set of intermediate milestones that would help to set intermediate goals and measure/evaluate progress toward these goals. Submissions must be in English, no more than 8 pages long, and in the two-column format prescribed by ACL'2001. Please see http://acl2001.dfki.de/style/ for the detailed guidelines. Submissions should be sent elec-tronically in Word (preferably) or PDF or ASCII text format to arrive no later than April 2, 2001 to Paula MacDonald (pmmmac@mitre.org). As soon as possible, authors are encouraged to send a brief email indicating their intention to participate to include their contact information and the topic they intend to address in their submission. Submissions will be evaluated on the basis of their relevance, innovation, quality, and presentation according to the schedule below. SCHEDULE + Submission Deadline: 2 April 2001 + Notification : 30 April 2001 + Camera Ready Papers Due: 16 May 2001 WORKSHOP DATE July 6 and 7, 2001 WORKSHOP URL http://www.elsnet.org/acl2001-hlt+km.html From Sandra Manzi Fri Feb 23 11:13:31 2001 From: Sandra Manzi (Sandra Manzi) Date: Fri, 23 Feb 2001 12:13:31 +0100 (MET) Subject: [MT-List] CFP -- MT Evaluation Workshop Message-ID: MT Evaluation: Who Did What To Whom A workshop held in conjunction with Machine Translation Summit VIII Saturday, September 22nd, 2001, Santiago de Compostela, Spain CALL FOR PAPERS Introduction Many measures and methods of evaluating MT systems have been developed over the years. The ISLE project, funded jointly by the European Union and the National Science Foundation of the USA, is continuing with work started in the EU's EAGLES project on systematizing these measures and procedures. This workshop will be the fourth in a series that report on and flesh out portions of the systematization. The main thrust of the work is to build up schemes that classify various aspects of import for MT, including user needs, the suggested system characteristics and associated metrics for measuring each of these. The classification schemes relate to ISO work on software evaluation. The work is intended to be useful to those who are considering using machine translation, those interested in comparative evaluation of several MT systems and to MT system developers. A fuller account of the ISLE evaluation work and an overview of the current classification schemes can be found at http://www.isi.edu/natural-language/mteval/. By the time of the MT Summit, the ISLE project will have organized or been involved in a number of workshops on MT evaluation, namely at LREC in Athens, AMTA in Mexico, Geneva Workshop, and NAACL in Pittsburgh. Participants are encouraged, but not required, to get involved with the Geneva Workshop or NAACL Workshop as these are hands-on exercises in MT. The results of the previous workshops, as possible, will be presented here and feedback based on them serves directly the goal of developing proposals of concrete usefulness to the whole community. This workshop will be seen as a preliminary attempt at reaching conclusions which synthesize results. Information about the previous workshops can be found at: LREC: http://www.icp.grenet.fr/ELRA/lrec2000.html AMTA: http://www.isi.edu/natural-language/conferences/AMTA2000.html NAACL 2001: http://www.cs.cmu.edu/~ref/naacl2001.html MT Evaluation workshop: An invitation to get your hands dirty! http://www.issco.unige.ch/projects/isle/mt-eval-workshop.html, and finally at the MT Summit VIII: http://www.eamt.org/summitVIII. Papers We invite submission of papers along the themes discussed above. The questions and issues to be answered are diverse. All papers on evaluation and evaluation issues will be considered, but preference will be given to papers relating to the ISLE framework. The following questions suggest possible evaluation threads within the framework. * What kind of metrics are useful for what system characteristics? * What system characteristics reflect what user needs? * Is there a radical difference between evaluation focusing on research or development needs and evaluation focusing on end-user needs? * When should real-world data be used, and what is the impact of using it? * What constitutes a valid metric? How can you demonstrate that a metric is valid? * What are the advantages/disadvantages of specific metrics? * What kinds of tools automate the evaluation process? Can the process (or any part of it) be automated? What are the difficulties inherent in choosing particular metrics for automation? * What kinds of tasks are suited to which evaluation schemes? * How can we use the evaluation process to speed or improve the MT development process? * How can we evaluate MT when MT is a small part of the data flow? How independent is MT of the subsequent processing? Cleaning up the data improves performance, but does it improve it enough? How do we quantify that? Submission Guidelines Papers may be submitted electronically to Sandra Manzi, (Sandra.Manzi@issco.unige.ch) by the submission date. The format may be RTF, Microsoft Word, or PostScript. Any specialized fonts should be included in the submission package. Users are advised to follow the main conference formatting as described at: http://www.eamt.org/summitVIII/guidelines.html. Submission for review: papers of no more than 6000 words are expected. Submission for publication: A template will be provided for accepted papers. Important Dates Paper submission deadline April 15, 2001 Notification of acceptance May 31, 2001 Final camera-ready due July 1, 2001 Schedule and Venue This one-day workshop will take place on September 22nd, 2001, after the regular conference. The main conference site contains information about registration and accommodation for both the conference and any satellite events, including this workshop. About Santiago Santiago de Compostela in North West Spain (Galicia) was the most visited pilgrimage shrine in the Middle Ages, and to this day the Road to Santiago is a popular route for walkers through France and northern Spain. The goal is the imposing 11th century cathedral containing the relics of St. James the Great. But the well-preserved historical center of the city offers much more: medieval streets, excellent restaurants, ancient churches, monasteries, palaces, etc. The venue for the conference is the Hostal de los Reyes Católicos, founded in 1499 as a Royal hospital for pilgrims; a magnificent building in the Spanish Renaissance style, and now a hotel (the oldest and one of the most beautiful in the world). Those staying at the Hostal will have a once-in-a-lifetime experience, but there is also accommodation nearby in many other good hotels, all within easy walking distance of the center and the Hostal. Santiago de Compostela has its own airport with daily direct flights from Barcelona, Madrid, Brussels and London. Contact Points and Organizing Committee: For any further information, please contact Sandra Manzi, (sandra.manzi@issco.unige.ch) TIM/ISSCO -- University of Geneva 40 blvd du Pont d'Arve CH-1211 GENEVA 4 (Switzerland) Tel: +41-22-705.8680 Fax: +41-22-705.8689 Organizing Committee E. Hovy, ISI, hovy@isi.edu B. Maegaard, CST, bente@cst.ku.dk S. Manzi, ISSCO, sandra.manzi@issco.unige.ch M. King, ISSCO, margaret.king@issco.unige.ch F. Reeder, MITRE, freeder@mitre.org From ivarea@info-ab.uclm.es Fri Mar 2 12:13:32 2001 From: ivarea@info-ab.uclm.es (Ismael Garcia-Varea) Date: Fri, 2 Mar 2001 13:13:32 +0100 Subject: [MT-List] Maximum Entropy Ristad toolkit Message-ID: <20010302131332.A1723@info-ab.uclm.es> Dear list members, My name is Ismael García. I am assistant professor for technical studies in the Computer Science Department at the Univesity of Castilla-La Mancha (Albacete Spain) and a member of the PRHLT (Pattern Recognition and Human Language Technology) group at the UPV (Politechnical Univesity of Valencia Spain). For my PhD I am working in Statistical Machine Translation. Some months ago I started to use the Ristad Maximum Entropy Toolkit (version 1.5) in order to construct refined lexicon models using the ME approach. So far I used a compiled version of the toolkit for ALPHA machines but two months ago the last available alpha machine here broke down. After that, I tried to download an intel version of this software at the Princeton University but it seems that the toolkit is not available anymore. Then now I have no possibility to continue my work in this direction because the only available machines in the department are all INTEL ones. Do you have an INTEL version of the this toolkit that was available some time (maybe years) ago ? Could you provide me it or tell me if there exist any place on where to find/download it? I want to notice that this stopped work is very important for finish my PhD, so any kind of help will be very very appreciated. Thank you very much in advance. Ismael García. -- Ismael García Varea Departamento de Informática Escuela Politécnica Superior de Albacete Universidad de Castilla-La Mancha Telf: +34 967 59 92 00 - Ext.- 2479 Fax: +34 967 59 92 24 mailto: ivarea@info-ab.uclm.es From VM131652@exchange.DAYTONOH.NCR.com Mon Mar 5 17:11:16 2001 From: VM131652@exchange.DAYTONOH.NCR.com (Morland, Verne) Date: Mon, 5 Mar 2001 12:11:16 -0500 Subject: [MT-List] Global English - Not! Message-ID: <936BA3429F9FD411B2A100E029265C3B01C92962@susdayte11.daytonoh.ncr.com> Fellow MT Aficionados, I just got around to reading an excellent article in The Atlantic Monthly's November 2000 issue, "What Global Language?" by Barbara Wallraff, a senior editor of The Atlantic. Conveniently the article is on the web, together with some supplementary material including an interview with the English Company U.K.'s David Graddol and a forum discussion on the topic, "Has the globalization of English affected your work -- or your life?" Wallraff's main point is that despite the continuing growth of English as the lingua franca of commerce and science, when it comes to reaching the general population of the world, English is, in fact, shrinking as a native tongue (in percentage terms). She writes: > According to "The Decline of the Native Speaker," a paper > David Graddol published last year in the AILA Review (AILA > is the French acronym for the International Association of > Applied Linguistics; the review belongs to the minority of inter- > national scholarly journals that still make use of another > language in addition to English), the proportion of native > English-speakers in the world population can be expected > to shrink over the century 1950-2050 from more than eight to > less than five percent. You will find the entire article at: http://www.theatlantic.com/globalenglish (Wow, how I love the web for this sort of thing!) Wallraff discusses MT briefly near the end of the article and her assessment, while cautious, is upbeat: > That's the theory, anyway. I have my doubts about how > exquisite the actual results will be for quite some time. > The interchanging of speech with writing, writing with speech, > and English with other languages may, however, yield > serviceable results very soon. ...[E]ach of these steps is > already being accomplished approximately, and imple- > mented not just in experimental settings but in real life. This is from the section called "The Web in my Own Language" and "English By Accident" at: http://www.theatlantic.com/issues/2000/11/wallraff3.htm Regards, Verne P.S. The links at the bottom of the article's first page to the continuation pages are broken. They should be: http://www.theatlantic.com/issues/2000/11/wallraff2.htm/ http://www.theatlantic.com/issues/2000/11/wallraff3.htm/ From tufis@racai.ro Mon Mar 5 17:30:21 2001 From: tufis@racai.ro (Dan Tufis) Date: Mon, 5 Mar 2001 19:30:21 +0200 Subject: [MT-List] Advanced Parser & Tagger References: <1.5.4.32.20010226142804.0092bbc4@u3.ici.ro> Message-ID: <007701c0a599$f4cd04b0$1327e6c1@racai.ro> Dear Mr. Gojol, I reiterate, this time publicly, the invitation to evaluate your amazing tools on some arbitrary texts for one of the languages you mention. As the results which you report for Romanian are incredibly high, it would be really interesting to compare the outputs from your tagger and our tagger. As I assume you are using a similar tagset as we do from the time you worked in my team and had access to all our resources, I guess that the comparison would be quite simple. This assessment would not require any disclosure of the inner solutions, but would be a good opportunity to persuade a speechless NLP community. Sincerely yours, Dan Tufis ----- Original Message ----- From: "Vlad V. Gojol" To: Sent: Monday, February 26, 2001 4:28 PM Subject: [MT-List] Advanced Parser & Tagger > > Dear Readers , > > Those interested in the two pieces of software presented below > or / and by the possibility of building a translating system to > Romanian ( e.g. from English ) are kindly asked to contact me per- > sonally ( gojol@sunu.rnc.ro ) . > Thank you , > Dr.ing. Vlad Gojol > > ---------------------------------- > Senior Research Engineer > Institutul National de Informatica > Bucuresti , Romania > > > ............................................................................ > > LANGUAGE INDEPENDENT PART-OF-SPEECH TAGGER > > I created a part-of-speech tagger with an unusual capacity of dealing > with large contexts , especially for German . I used Negra ( seemingly the > best known German corpus , with free obtainable licence ) . The tagger > currently reputed as being the most accurate for German is perhaps TnT . It > reports upon this corpus an error rate of 3.4% . But I have found a syste- > matic error in Negra : all the occurences of the auxilliary verbs are tagged > as auxilliary ( VAFIN ) , though in 50% of the cases they function as finite > verbs ( VVFIN ) . I corrected a part of the corpus ( cca 40,000 tokens ) . > In this more correct environment ( where the performance of TnT should be > probably around 4.5% ) , my tagger gets 1.7% . > On another German corpus ( I call it X ) , with comparable contents ( news- > paper articles ) and tagset , but with attached EXTERIOR lexicon ( i.e. not > extracted from the corpus ) , the result is 2.4% . > I also used Susanne ( the only English corpus I could get free ) . The re- > ported result for TnT is 3.8% . Mine is 2.8% . On the "A" texts , best paral- > lelable with those in Negra , as journalistic , it's 2.3% . By restricting > the tagset to a more normal size ( cca 100 tags , determined as optimal af- > ter lots of test runs ) , it's 1.3% . > Initially I had used a Romanian corpus , with a result of 0.9% ( compared > to 1.7% , 2.5% and 4.2% respectively got by the Xerow , Birmingham and Brill > taggers ) . > > The speed is comparable to that of TnT and modifiable by parameter setting , > in reverse proportion to the accuracy ( but without affecting it much ) . > The incremental operating mode and the data structures segmentation allow > running on very small memory computers . > There is the advantage of an intuitive output ( no hostile binary matrix ) , > in a form analogue to the input of some expert systems . > Alternative taggings are output , with scores : unlike with other taggers , > they don't refer to individual words , but to whole sentence parts ( repre- > senting somehow phrase surfaces of minimum energy ) . > Special facilities exist , such as virtual tags , or context essentialisa- > tion ( permitting to get the minimal contexts set characteristic to a certain > linguisic style , useful not only for maximum accuracy and speed ) etc. > Recently added features : wide-character support , the possibility of > being called as a simple library routine , or suspending the notion of a > file ( a complex files system is emulated into the memory ) . For example > ( for the second feature above ) , you can create an instance of the tagger > ( let's say for English with the tagset Lancaster ) , call it to tag a cer- > tain text buffer ( by writing the resulting tags into another buffer ) and > finally kill the instance , all this without using any disc file : > t = GojolTagger_new("lancaster"); > error_code = GojolTagger_tag(t,input_buffer,output_buffer); > GojolTagger_free(t); > > All is built on two essentially new concepts : organicity and context pro- > pagation . I didn't publish anything about them , to keep up their commercial > appeal . The accuracy comparable to that of manual tagging made me find many > errors in the used corpora : 98 in Negra , 36 in Susanne ; Prof. G. Sampson > replied gratefully , saying that it's the first time somebody reports more > than 2 errors , and that my findings make necessary a new version of Susanne . > The handling of very large contexts could even modify the current tagsets de- > sign , by cancelling some unnatural decisions ( motivated only by the incapa- > city of the existing taggers to see beyond a 3-tokens neighborhood ) , such as > those concerning the auxilliary verbs , participles etc. - so removing some > burden from the subsequent stages of text processing . > It is written in C ( Linux ) . Demos for German ( Negra ) and English > ( Susanne ) are available . > > ............................................................................ .. > > LANGUAGE INDEPENDENT STATISTIC PARSER > > After learning from a 46,000 words pos-tagged corpus and > a 32,000 words parsed ( treebank ) corpus , a 2,000 words > text ( not included in any of the two corpora ) is parsed > ( tagging excluded ) in 6 seconds ( on a 200 MHz machine ) > with 2% incomplete trees ( but for these declared failures , > are also provided well formed trees sufficient for a subse- > quent translator ) - the extracted grammar having cca 12,000 > rules . The Negra corpus of German was used . After learning > from a 17,000 words parsed corpus and from the same 46,000 > words pos-tagged one , a 2,000 words text included into the > first ( but excluded from the second ) , to warrant that the > grammar is complete relative to it ( i.e. contains all the > rules necessary for its correct parsing ) , is processed in > 2 seconds with no incomplete tree - the extracted grammar > having cca 7,000 rules . The system is language independent > - for English , upon the Susanne corpus , comparable results > are obtained . To have an acceptable parser for any other i- > diom , you need essentially simply a corpus with 30,000 tag- > ged words , from which only 20,000 parsed as well - and for > optimal results , 50,000 and 30,000 respectively . > The parser may accept a set of rules intended to modify > the statistical grammar deduced from the corpus . Moreover , > it can take as input only a context-free grammar ( in which > case it ceases to be a statistical parser ) , but in this > operating mode it requires much time and memory ( during the > learning , not during the parsing as such ) if the grammar > is over-dimensioned . The statistical grammar is refined not > by simply adding the proposed rules , but by modifying the > corpus , to exploit all the real contexts possible for them . > Semantic processing could be easily inserted at rule re- > duction points . Actually this generalized parser can also > work as a compiler generator : by appending specific semantic > routines , you get efficient compilers for C , Pascal etc. > This versatile system has more than 40 parameters which tune > the accuracy and speed according to the target language sam- > ple . The output is in treebank format and optionally in gra- > phic ( with the trees effectively drawn ) one . > Linux demos exist for German and English . As only the mi- > nimal definition of C is used , it is easily adaptable to any > machine ( for other Unix-like operating systems , probably a > simple recompilation would be sufficient ) . > > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > From wolfskil@MIT.EDU Mon Mar 5 19:54:52 2001 From: wolfskil@MIT.EDU (Jud Wolfskill) Date: Mon, 5 Mar 2001 14:54:52 -0500 Subject: [MT-List] book announcement--Melamed Message-ID: --============_-1228300390==_ma============ Content-Type: text/plain; charset="us-ascii" Dear Moderator, I wondered if the following book announcement would be appropriate for posting to the MT List. I'd be happy to edit the announcement to meet your specifications. Please let me know whether or not you use the announcement. Thank you! Best, Jud I thought readers of MT List might be interested in this book. For more information please visit http://mitpress.mit.edu/promotions/books/MELEHF00. Empirical Methods for Exploiting Parallel Texts I. Dan Melamed Parallel texts (bitexts) are a goldmine of linguistic knowledge, because the translation of a text into another language can be viewed as a detailed annotation of what that text means. Knowledge about translational equivalence, which can be gleaned from bitexts, is of central importance for applications such as manual and machine translation, cross-language information retrieval, and corpus linguistics. The availability of bitexts has increased dramatically since the advent of the Web, making their study an exciting new area of research in natural language processing. This book lays out the theory and the practical techniques for discovering and applying translational equivalence at the lexical level. It is a start-to-finish guide to designing and evaluating many translingual applications. I. Dan Melamed is a research scientist in the Computer Science Research Department of West Group, Minnesota. 6 x 9, 198pp., 75 illus., cloth ISBN 0-262-13380-6 -------------------------------------------------------------------------------- Jud Wolfskill 617.253.2079 phone Associate Publicist 617.253.1709 fax MIT Press wolfskil@mit.edu 5 Cambridge Center http://mitpress.mit.edu Fourth Floor Cambridge, MA 02142 --============_-1228300390==_ma============ Content-Type: text/enriched; charset="us-ascii" MonacoDear Moderator, I wondered if the following book announcement would be appropriate for posting to the MT List. I'd be happy to edit the announcement to meet your specifications. Please let me know whether or not you use the announcement. Thank you! Best, Jud I thought readers of MT List might be interested in this book. For more information please visit http://mitpress.mit.edu/promotions/books/MELEHF00. Empirical Methods for Exploiting Parallel Texts I. Dan Melamed Parallel texts (bitexts) are a goldmine of linguistic knowledge, because the translation of a text into another language can be viewed as a detailed annotation of what that text means. Knowledge about translational equivalence, which can be gleaned from bitexts, is of central importance for applications such as manual and machine translation, cross-language information retrieval, and corpus linguistics. The availability of bitexts has increased dramatically since the advent of the Web, making their study an exciting new area of research in natural language processing. This book lays out the theory and the practical techniques for discovering and applying translational equivalence at the lexical level. It is a start-to-finish guide to designing and evaluating many translingual applications. I. Dan Melamed is a research scientist in the Computer Science Research Department of West Group, Minnesota. 6 x 9, 198pp., 75 illus., cloth ISBN 0-262-13380-6 -------------------------------------------------------------------------------- Jud Wolfskill 617.253.2079 phone Associate Publicist 617.253.1709 fax MIT Press wolfskil@mit.edu 5 Cambridge Center http://mitpress.mit.edu Fourth Floor Cambridge, MA 02142 --============_-1228300390==_ma============-- From gojol@sunu.rnc.ro Wed Mar 7 18:00:35 2001 From: gojol@sunu.rnc.ro (Vlad V. Gojol) Date: Wed, 07 Mar 2001 18:00:35 +0000 Subject: [MT-List] Advanced Parser & Tagger Message-ID: <1.5.4.32.20010307180035.00909470@u3.ici.ro> At 07:30 PM 3/5/01 +0200, you wrote: >I reiterate, this time publicly, the invitation to evaluate your amazing >tools on some arbitrary texts for one of the languages you mention. Mr Tufis , It's not really necessary to express by a public complaint the frustration for not having been accepted as an interlocutor . First I decided to ignore such an exotic petition , but insinuations should be cleared when produced before an audience , however . >As I assume you are using a similar tagset as we do from the time you worked >in my team I don't remember having ever "worked in your team" : I simply got a two man-months collaboration contract in 1997 , because nobody in that team seemed able to read the sources of Brill's tagger ( in or- der to correct the errors inside and to make it fit into your small memory computers , by rewriting a significant part ) . If I ordered some software to Microsoft , I would not feel entitled to proclaim I was Bill Gates' boss for a while . Anyway thanks for this entertai- ning vision : toiling alongside with some fresh graduates , a grey beard personage ( a bit more aged than you ) , which happened to have led during the 80's the software team around the first Romanian mini- computer . >and had access to all our resources I confess I used for the preliminary version of my tagger your "1984" corpus . A couple of months later ( 1998 ) , finding no per- spective in dealing with Romanian texts , I deleted it accidentally ( and without regrets ) - I had already two German corpora and an En- glish one . You waste your energy around a trifle : virtually nobo- dy ( outside Romania ) is interested in results upon Romanian , and the respective line of my advertisement was remarked probably only by you ( if I had put "language Y" instead , the general impact would have been identical ) : the only meaningful thing there was the superiority upon Brill's and Birmingham taggers , which was al- ready pleonastic after the comparison with TnT , a much better ra- ted tagger . >As the results which you report for Romanian are incredibly high Sometimes suspicion is the nicest compliment . You could have a- voided doing me this favour by a simple exercise of logic . German is obviously more difficult to tag than Romanian because of its scattered syntax ( lots of words between an auxilliary and a par- ticiple , or between an article and a noun etc ) . The Negra corpus has at least an average difficulty , being carefully selected from quite different newspaper domains - while "1984" is a text of big syntactic simplicity , way below the average . The Susanne corpus aims explicitly at the complexity ( sentences of more than 40 words are not uncommon ) . If from the former three statements you derive the conclusion that a tagger which gets 1.7% on Negra and 1.3% on Sue is not supposed to get 0.9% on "1984" , then that's your pro- blem . >would be a good opportunity to persuade a speechless NLP >community. Probably by speechless community you designate your team ( since I don't see who else would comission you as a speaker ) . As for the colleagues abroad , several tens of them got demos ( and I don't re- member to have had contestations about ) during 1999-2001 . By the way the tagger is already licensed in Germany . -- Vlad V. Gojol From steven.krauwer@let.uu.nl Wed Mar 7 16:49:36 2001 From: steven.krauwer@let.uu.nl (Steven Krauwer) Date: Wed, 07 Mar 2001 17:49:36 +0100 Subject: [MT-List] Advanced Parser & Tagger References: <1.5.4.32.20010307180035.00909470@u3.ici.ro> Message-ID: <3AA666A0.2B608FD3@let.uu.nl> "Vlad V. Gojol" wrote: > > At 07:30 PM 3/5/01 +0200, you wrote: > >I reiterate, this time publicly, the invitation to evaluate your amazing > >tools on some arbitrary texts for one of the languages you mention. > > Mr Tufis , > > It's not really necessary to express by a public complaint the > frustration for not having been accepted as an interlocutor . First > I decided to ignore such an exotic petition , but insinuations > should be cleared when produced before an audience , however . > ... Hey guys, do we really have to listen to your internal fights? I've really got better things to do -- and I wish you had too. Steven ______________________________________________________________________ Steven Krauwer, ELSNET / UiL OTS, Trans 10, 3512 JK Utrecht, Nederland phone: +31 30 2536050, fax: +31 30 2536000, email: s.krauwer@let.uu.nl http://www-sk.let.uu.nl http://www.elsnet.org From Steven.Krauwer@let.uu.nl Fri Mar 23 13:45:21 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Fri, 23 Mar 2001 14:45:21 +0100 (MET) Subject: [MT-List] ACL/EACL2001 Tools Workshop -- Final CfP Message-ID: <200103231345.OAA04899@salto.let.uu.nl> Apologies for multiple listings **FINAL CALL FOR CONTRIBUTIONS** ACL/EACL2001 Workshop on Sharing Tools and Resources for Research and Education Co-organised by ELSNET Toulouse, Saturday 7th July 2001 BACKGROUND: At a workshop at ACL 2000 in Hong Kong dedicated to Infrastructures for Global Collaboration there was an agreement between the main professional organisations in NLP and Speech (ACL and ISCA), and ELSNET, and the other meeting participants, that it would be useful to aim at a broadly supported, joint repository or catalogue for tools and materials for the language and speech communities. An ELSNET-sponsored workshop on educational issues held at EACL99 concluded that certain non-transient infrastructures needed to be instigated to raise the public perception of educational issues in NLP. It also concluded that a repository of shared materials, appropriately indexed for educational usage, would be a useful point of departure. This workshop will build on the consensus reached at these previous workshops. There will be two clear foci: one upon instruments for sharing tools and resources in general that addresses practical problems, and the other upon the technological and infrastructural issues surrounding the educational uses of repositories. Good examples of existing initiatives in this area are among others the ACL Natural Language Software Registry (hosted at DFKI, registry.dfki.de) which was set up as a repository for tools for the distinct fields of Human Language Techology (HLT), the ELRA/ELDA, LDC, TELRI and Elsnet resources catalogues and repositories (www.icp.inpg.fr/ELRA, www.ldc.upenn.edu, www.telri.de and www.elsnet.org/resources.html), OLAC (a worldwide network of language archives at www.language-archives.org), and JEWELS (www.elsnet.org/jewels), an as-yet incomplete EU funded website for educational materials in Language and Speech. A third theme concerns how to build upon existing initiatives as sources of data or inspiration. AIM AND SCOPE OF THE WORKSHOP: The main goal of the workshop is to discuss methods for the improvement and extension of existing repositories; the educational uses of repositories; the closer interlinking between different kinds of repositories (tools and resources); global infrastructures for the achievement of joint actions. However, we expect the scope of the workshop to be much wider than that, as the issues addressed are of general interest to everybody who believes that sharing tools and resources is essential for the progress of research and education in our field. Contributions of papers and demonstrations are solicited that address the above themes. The following list of topics is suggestive rather than exhaustive: * Repositories versus catalogues * Mechanisms and infrastructures for sharing and describing content * Repository management * Standards for exchange, description, and annotation * Metadata descriptions * Quality assessment * Structure and content of an NLP/CL repository * Tools and materials for NLP/CL education * Web-based teaching methods for NLP/CL SUBMISSION REQUIREMENTS FOR PAPERS * Electronic submissions only (PostScript, Word, or PDF), following the appropriate ACL latex style or Microsoft Word style. Submissions should not exceed eight (8) pages, including references. You can download the appropriate style or template files using the following link: acl2001.dfki.de/style. In case of problems with the submission format, please contact one of the co-chairs. * Submissions to either co-chair (Mike Rosner and Thierry Declerck). All submissions will be acknowledged. * Please provide a list of keywords in the separate header page and indicate the best fitting subtopic(s) from the above list. SUBMISSION REQUIREMENTS FOR DEMONSTRATIONS * Demos may be submitted with or without an accompanying paper. * Please write a 2-page description of the demo and send to either co-chair. Please let us know about special hardware requirements over and above the standard PC + beamer without internet access provided by default CONFIRMED COMMITTEE MEMBERS - Thierry Declerck (DFKI) Co-chair (Repository) declerck@dfki.de - Mike Rosner (Malta) Co-chair (Education) mros@cs.um.edu.mt - Steven Krauwer (Utrecht University) Co-chair s.krauwer@elsnet.org - Steven Bird (U. Penn) sb@unagi.cis.upenn.edu - Bill Black (UMIST) (UMIST, Manchester, UK) wjb@co.umist.ac.uk - Gosse Bouma (University of Groningen) gosse@let.rug.nl - Koenraad de Smedt (University of Bergen) desmedt@uib.no - Claire Gardent (CNRS, Nancy) claire.gardent@loria.fr - Donna Harman (NIST) donna.harman@nist.gov - Julia Hirschberg (ATT, ISCA) julia@research.att.com - Jun'ichi Tsujii (Tokyo) tsujii@is.s.u-tokyo.ac.jp - Andy Way (Dublin City University) away@compapp.dcu.ie - Antonio Zampolli (Univ. of Pisa), pisa@ilc.pi.cnr.it DEADLINES * Submission Deadline: 6th April 2001 * Notification Date: 27th April 2001 * Camera ready copy due: 16th May 2001 WORKSHOP URL http://www.elsnet.org/acl2001-tools.html FOR FURTHER INFORMATION CONTACT Michael Rosner mros@cs.um.edu.mt Thierry Declerck declerck@dfki.de From Steven.Krauwer@let.uu.nl Fri Mar 23 14:39:23 2001 From: Steven.Krauwer@let.uu.nl (Steven Krauwer) Date: Fri, 23 Mar 2001 15:39:23 +0100 (MET) Subject: [MT-List] ACL/EACL2001 Arabic NLP Workshop -- Final CfP Message-ID: <200103231439.PAA07116@salto.let.uu.nl> FINAL CALL FOR PAPERS ACL/EACL 2001 Workshop ARABIC Language Processing: Status and Prospects Toulouse, France, Friday 6 July 2001 Co-organized by: ELSNET/NAPLUS WORKSHOP OBJECTIVES AND DESCRIPTION: The objective of the workshop is threefold. * First of all we want to bring together people who are actively involved in Arabic language and/or speech processing in a mono- or multilingual context, and give them an opportunity to report on completed and ongoing work as well as on the availability of products and core technologies. This should enable the participants to develop a common view on where we stand with respect to Arabic language processing. * Secondly, we want to identify problems of common interest, and possible mechanisms to move towards solutions, such as sharing of tools and resources, moving towards standards, sharing and dissemination of information and expertise, adoption of current best practices, setting up joint projects and technology transfer mechanisms, etc. * Third, we would like to enhance collaboration between the Arabic NLP community and the NLP community at large. The workshop program will include the following components: * Introduction * Overview talks * Scientific papers * Short presentations of projects, core technologies and products * A panel session and/or a round table discussion * Conclusions SUBMISSION REQUIREMENTS FOR SCIENTIFIC PAPERS: Papers are solicited that address all aspects of Arabic language processing in a mono- or multilingual context, including tools, resources and standards. Papers will have to be original and report on completed research. Submissions of scientific papers should not exceed eight (8) pages. Please provide a list of keywords in the separate header page. Further submission details below. SUBMISSION REQUIREMENTS FOR SHORT PRESENTATIONS OF PROJECTS, CORE TECHNOLOGIES AND PRODUCTS: Short presentations serve to give the audience an impression of ongoing activities and projects, and of existing core technologies and products, with a view to possible collaboration and synergies (i.e. NO commercial product presentations). Submissions of short presentations should not exceed two (2) pages. Short presentations will be reviewed on the basis of relevance and clarity of presentation. REQUIREMENTS FOR ALL SUBMISSIONS: Electronic submissions only (PostScript, Word, or PDF), following the appropriate ACL latex style or Microsoft Word style. Submissions should not exceed the length indicated above, including references. You can download the appropriate style or template files using the following link: http://acl2001.dfki.de/style/. Submission and presentation language is English. In case of problems with the submission format, please contact steven.krauwer@elsnet.org. Submissions should be sent to steven.krauwer@elsnet.org. All submissions will be acknowledged. DEADLINES FOR ALL SUBMISSIONS: * Submission deadline: 6th April 2001 * Notification date: 27th April 2001 * Camera-ready papers due: 16th May 2001 * Workshop date: 6th July 2001 CONFIRMED CORE PROGRAMME/ORGANISATION COMMITTEE: * Mustafa Yaseen, Amman University, Jordan (Co-chair, myaseen@cbj.gov.jo) * Joseph Dichy, Universite Lumiere-Lyon 2, France (Co-chair, dichy@univ-lyon2.fr) * Steven Krauwer, Utrecht University / ELSNET, The Netherlands (Contact person, steven.krauwer@elsnet.org) * Adnane Zribi, University of Tunis, Tunisia (adn@gnet.tn) * Salem Ghazali, IRSIT, Tunisia (ghazali@irsit.rnrt.tn) * Humoud Al-Sadoun, Ministry of Education, Kuwait (hbh@moe.edu.kw) * Jean Senellart, SYSTRAN, France (senellart@systran.fr) * Nadia Hegazy, ERI, Egypt (nhegazy@idsc.gov.eg) * Khalid Choukri, ELRA/ELDA, France (choukri@elda.fr) * Malek Boualem, FTRD/DMI/LAN, France (malek.boualem@rd.francetelecom.fr) * Everhard Ditters, University of Nijmegen, The Netherlands, (e.ditters@let.kun.nl) WORKSHOP URL: http://www.elsnet.org/acl2001-arabic.html CONTACT INFO: Steven Krauwer email: steven.krauwer@elsnet.org ELSNET / UiL OTS www: http://www.elsnet.org Trans 10 phone: +31 30 253 6050 3512 JK Utrecht, NL fax: +31 30 253 6000 From WJHutchins@compuserve.com Mon Mar 26 10:42:17 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 26 Mar 2001 05:42:17 -0500 Subject: [MT-List] MT Summit conference -- Invitation Message-ID: <200103260542_MC2-CA2A-41EC@compuserve.com> MT Summit VIII, Santiago de Compostela, Spain, 18-22 September 2001 The European Association for Machine Translation (EAMT) invites you to participate in the Eighth "Machine Translation Summit" which is being held in Santiago de Compostela, = Galicia, Spain from the 18th to the 22nd September 2001. The MT Summit conferences bring together the global MT = community to review the state of the art of machine = translation, computer-aided translation support tools and multilingual tools, both in the commercial = world and in the research sphere. It is a unique opportunity to learn about the latest developments in the field, for researchers, users and vendors to exchange experiences and to get valuable feedback, and for meeting others involved in this technology. There will be a full programme of papers, presentations = by invited speakers, panel sessions, exhibitions, demonstrations, and workshops (on the day preceding the conference and on the final day). There will be = excursions, a banquet and a folk music performance, and also some activities for spouses and partners of delegates. VENUE The conference is taking place in Santiago de Compostela in North West Spain(Galicia), famous as a pilgrimage destination since the early Middle Ages, a city full of ancient buildings, from its eleventh-century cathedral to its seventeenth and eighteenth century houses, convents and palaces. The old town has been listed as a 'world heritage city' by Unesco. The venue for the conference is the Hostal de los Reyes Cat=F3licos (Hostal dos Reis Cat=F3licos), originally founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style. It is now both the oldest and one of the most impressive hotels in the world. The opening session and the banquet will take place in the Comedor Real, other sessions will be in the hotel's auditorium and other rooms, and the exhibition and demonstrations will be housed in the Chapel. The opening reception will take place in the Pazo de Raxoi (City Council), an 18th century building next to the Hostal. HOTELS Santiago de Compostela is a major tourist centre attracting many visitors throughout the year. We have reserved accommodation in the Hostal itself and in a number of hotels close by (many restored 17th and 18th century buildings). All except one are within walking distance of the Hostal venue. (A bus will be provided for the Hotel Puerta del Camino). Because of the high demand for hotel rooms in Santiago and the substantial deposits, we ask for about half the full hotel charge when booking, and we encourage you to book early to be certain of your preference. WEBSITE INFORMATION For details of the hotels, excursions and workshops see the MT Summit website (www.eamt.org/summitVIII) or send email to: summitVIII@eamt.org. REGISTRATION Registration should preferably be made via the MT Summit website. Alternatively, you may ask for a registration form from: Carlos Paz (MT Summit Registration), Service de Traduction, European Commission, Rue de la Loi 200 - JECL 4/200, B-1049 Brussels, Belgium (Fax: +32-2-296.59.68; Telephone: +32-2-296.05.59; Email: Carlos.Paz-Carneiro@cec.eu.int). CONFERENCE PROGRAMME Tuesday, 18 September Workshops, welcome reception, opening of exhibitions Wednesday, 19 September opening session, first day of conference Thursday, 20 September Second day of conference Friday, 21 September Third day of conference, IAMT general assembly, folk music entertainment Saturday, 22 September Workshops, excursion to Coruna, special panel, banquet Sunday, 23 September Excursion to southern Galicia and northern Portugal CONFERENCE FEES The registration fees for the conference (19-22 September) are: IAMT members* Non-members Before 15 May Euro 250 Euro 285 After 15 May, before 15 July Euro 275 Euro 325 After 15 July Euro 325 Euro 375 * members of the Asia-Pacific Association for Machine Translation (AAMT), Association for Machine Translation in the Americas (AMTA), European Association for Machine Translation (EAMT) One Euro is approximately 0.90 US dollar. Note: The conference fees do not include meals, but there are numerous good-value restaurants nearby in the city, which you will have opportunity to enjoy during the conference. CANCELLATIONS Before 15 July: 75% of the registration fee will be refunded After 15 July and before 1 September: 50% of the registration fee will be refunded After 1 September: no refunds. EXHIBITIONS Exhibition space will be provided in the Chapel of the Hostal for up to 25 companies and will be active throughout the conference. Fees for exhibitors will be Euro 1200 (or Euro 1000 for early applicants). Contact Joerg Schuetz (joerg@iai.uni-sb.de) for details, or look at the Summit website (www.eamt.org/summitVIII). WORKSHOPS [for more details see Summit website] Tuesday 18 September fee Towards a Roadmap for MT (half day) Euro 25 Organiser: Steven Krauwer Machine interpretation and machine translation: different evaluation methods for different systems? (half day) Euro 25 Organiser: Susanne Jekat Example-based machine translation (full day) Euro 50 Organiser: Andrew Way Saturday 22 September Teaching machine translation (all day) Euro 50 Organisers: Mikel Forcada and Derek Lewis Interlinguas and interlingual approaches to MT (Fifth Interlingua workshop)(all day) Euro 50 Organisers: David Farwell and Steven Helmreich MT evaluation: who did what to whom (Fourth ISLE workshop) (all day) Euro 50 Organisers: Ed Hovy, Maghi King, Sandra Manzi and Florence Reeder OPEN MEETING on terminology exchange and synchronization (SALT, OSCAR, OLIF), Saturday 22 Sept (all day) Euro 50 [for more information see http://www.ttt.org] FLIGHT DISCOUNTS. Iberia Airlines is offering participants a 25% discount (or more) on many of its flights (for details see: www.eamt.org/summitVIII). On registration you will be sent a password, in order to take advantage of this offer. ORGANISERS John Hutchins (general chair): EAMT/IAMT president Bente Maegaard (programme chair) Joerg Schuetz (exhibitions) Harold Somers (workshops) Carlos Paz (registration and local arrangements) Anthony Clarke (finance): EAMT treasurer Viggo Hansen (publicity): EAMT secretary Colin Brace (website) From 106937@zeus.wmid.amu.edu.pl Thu Mar 29 08:24:42 2001 From: 106937@zeus.wmid.amu.edu.pl (Pinczewski Wojciech) Date: Thu, 29 Mar 2001 10:24:42 +0200 Subject: [MT-List] Can somebody... Message-ID: ...write to me where can I find informations about LOGOS MT system(excluding LOGOS and IBM homepages). From Kerstin.Sinautzki@daimlerchrysler.com Fri Mar 30 07:01:52 2001 From: Kerstin.Sinautzki@daimlerchrysler.com (Kerstin.Sinautzki@daimlerchrysler.com) Date: Fri, 30 Mar 2001 09:01:52 +0200 Subject: [MT-List] test suits Message-ID: <0057440013452500000002L402*@MHS> Dear all,=20 we are currently looking for test suits in German and English in order = to test=20 our MT dictionaries. Do any of you know of interesting (and free) test = suits in=20 the Internet? Your help is highly appreciated. Kerstin Sinautzki= From workshop-taln@systran.org Fri Mar 30 15:56:03 2001 From: workshop-taln@systran.org (TALN) Date: Fri, 30 Mar 2001 17:56:03 +0200 Subject: [MT-List] CALL FOR PARTICIPATION - Workshop on Machine Translation Message-ID: >>Sorry for multiple copies<< CALL FOR PARTICIPATION Workshop on Machine Translation "Machine Translation and Large-Scale Applications" 5 July 2001 http://www.systransoft.com/TALN2001-en.html in conjunction with TALN'2001(2-5 July 2001) TOURS Europe - FRANCE Submission deadline : 15 april 2001 MOTIVATION Against the backdrop of an increasingly multilingual society, the Machine Translation Workshop of TALN'2001 looks at the main challenges to Machine Translation and multilingual NLP. Indeed, MT has seen a remarkable development for the past few years in terms of the number of translation requests (about a million translation requests everyday on the Web) as well as in terms of the different types of formats translated : translation of dynamic resources(FAQ, Web pages, daily papers), emails, requests for search engines, etc. These different types of translations demand different levels of quality, and induce an important associated feedback in the form of additional bilingual dictionaries. At the same time there is a growing need for more specialized translation. For this "specialization" the construction of wide-covering mutlilingual resources is an essential component of this more demanding translation. Because of the increase of the number of contributors (experts or users), the questions of exchange, reusability and incrementability of the translation resources had to be raised. The associated investment (public or private) lead to a general tendency to initiate centralizations (cf. Elra), standards (ISO 12220), transfer formalisms (OLIF), and more generally, to create mutlilingual data exchange formats (UNL). In any case, the choice of a linguistic data support cannot be separated from pragmatic considerations concerning the scale, the volume of additional resources becoming more important than the generic resources. These resources can vary from "simple" terminological resources (aligned lexicons) to more complex resources such as terminological grammars, transfer dictionaries, and graphs/grammars of translation. FOCUS OF THE WORKSHOP The focus of the workshop is not only recent machine translation research and products, but also the latest developments in multilingual language technology. The workshop aims at considering the questions of the construction, the validation and the utilization of large-scale resources applied to machine translation. The papers submitted can be related to the different topics below, but also to some other closely related topics : - use of standardization formalisms on important volumes of data - evaluation and realization of construction tools of such resources (from corpus, monolingual resources...) - integration and conversion of existing resources - validation and "expert" improvement processes - massive parallel acquisition of terminology in the framework of a network of users - management of mutlilingual resources compared to bilingual resources - multilingual extraction and machine translation QUESTIONS AND ISSUES You can either submit an abstract of 5 pages, or a complete version of your paper(up to 10 pages). Submitted papers should be in French or in English (for non French speakers). The abstracts and the papers submitted should conform to the submission format of TALN'2001 (style sheets are available on the TALN'2001 site http://www.li.univ-tours.fr/taln-recital-2001/ ) : Times 12, single-spaced, 10 pages at the most, including figures, examples and references. Authors should send their submission as a file attached to an e-mail (ps, pdf, rtf files, A4 format) at the following e-mail address : mailto:workshop-taln@systran.org The e-mail should contain the following information: submission title, authors' names and affiliation. Submissions will be reviewed by two experts of the Program Comittee. The final version of the accepted papers will be published in the workshop proceedings. The papers will give rise to 20 minute presentations followed by 10 minutes of questions. PROGRAM COMITTEE Christian Boitet, CLIPS-IMAG,equipe GETA, Grenoble http://clips.imag.fr/geta/ Maurice Gross, LADL & UMLV, Marne-la-Vallee Igor Boguslavskij, IPPI PAN, Moscou Georges Carayannis, ELRA & ILSP, Athen Joseph Dichy, Faculte des Langues, Universite Lyon II Pierre-Yves Foucou, SYSTRAN, Paris http://www.systransoft.com/ Daniel Grasmick, SAP, Waldorf Jean Senellart, SYSTRAN, Paris http://www.systransoft.com/ Tamas Varadi, Linguistics Institute, Hungarian Academy of Sciences Remi Zajac, CRL/NMSU, Las Cruces, http://crl.nmsu.edu/~rzajac/ IMPORTANT DATES Submission deadline: 15/04/2001 Notification to authors: 25/04/2001 Final version due (camera-ready) : 15/05/2001 Workshop : 5 July 2001 From ilidju@cal2.vsnl.net.in Sat Mar 31 08:14:38 2001 From: ilidju@cal2.vsnl.net.in (Dr. S. Bandyopadhyay) Date: Sat, 31 Mar 2001 13:44:38 +0530 Subject: [MT-List] WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION ( Dr. S. BANDYOPADHYAY ) Message-ID: <000e01c0b9ba$a2f69ae0$0d0101ca@vsnl.net.in> This is a multi-part message in MIME format. ------=_NextPart_000_000B_01C0B9E8.BA9EE2A0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION =20 Workshop on Example-Based Machine Translation: 2nd Call for Papers Hosted by MT-SUMMIT VIII Santiago de Compostella, Spain, September 18-22, 2001 http://www.eamt.org/summitVIII/index.html http://www.compapp.dcu.ie/~away/EBMT.html Co-chairs: Michael Carl, IAI, Saarbr=FCcken; Andy Way, Computer = Applications, Dublin City University=20 In recent years, corpora of multilingual translated texts have become = widely available for a number of languages. Notwithstanding the seminal = paper by Nagao (84), it is primarily since the early 90's that such = bilingual texts have been exploited in the area of Machine Translation = (MT).=20 The two main paradigmatic approaches which have been proposed are = Statistics-based Machine Translation (SBMT) and Example-Based Machine = Translation (EBMT). A related variant of EBMT that we ignore here, = despite being widely used in the localisation area, is that of = Translation Memories (TM). No new translations are created afresh from = previously existing examples in the system database: rather, the closest = translation matches are proposed to the user for post-editing into the = correct translation.=20 While translation memory systems are used in restricted domains, SBMT = systems require training on huge, good quality bilingual corpora. As a = consequence TMs can hardly be applied as a general purpose solution to = MT and SBMT as yet cannot produce complex translations to the desired = quality, even if such translations are given to the system in the = training phase. EBMT seeks to exploit and integrate a number of = knowledge resources, such as linguistics and statistics, and symbolic = and numerical techniques, for integration into one framework. In this = way, rule-based morphological, syntactic and/or semantic information is = combined with knowledge extracted from bilingual texts which is then = re-used in the translation process.=20 However, it is unclear how one might combine the different knowledge = resources and techniques in an optimal way. In EBMT, therefore, the = question is asked: what can be learned from a bilingual corpus and what = needs to be manually provided? Furthermore, we remain uncertain as to = how far the EBMT methodology can be pushed with respect to translation = quality and/or translation purpose. Finally, one wonders what the = implications and consequences are for size and quality of the reference = translations, (computational) complexity of the system, sizeability and = transportability, if such an approach is taken.=20 Given this background, we propose to organize a workshop in order to = shed some light on these open questions, among others. We are seeking = contributions which go beyond the purely statistical and/or rule-based = approaches to MT. We welcome visionary and technical descriptions, = reports of empirical research as well as feasibility studies and system = demonstrations. We would welcome contributions on any of the following = topics and sub-headings:=20 a.. (semi-)automatic preparation of existing bi/multilingual corpora = for EBMT=20 a.. extraction of bi/multilingual texts from the web=20 b.. preparation of treebanks for EBMT=20 c.. bi/multilingual alignment/bracketing/parsing=20 d.. inference of bi/multilingual grammar and transfer rules=20 b.. description of `pure' EBMT systems=20 a.. knowledge resources used=20 b.. representation of numeric and symbolic knowledge=20 c.. descriptions of `hybrid' systems integrating EBMT with rule-based = or other methodologies=20 d.. evaluation of EBMT results and/or comparison with other MT systems = e.. considerations on domain-(in)dependence of EBMT systems=20 f.. computational and/or system complexity of EBMT systems=20 Submissions Submitted papers must describe original, previously unpublished work. = Submissions must not exceed 12 pages. Contributions should be submitted = to Michael Carl. Conference stylesheets are now available. Deadlines = appear below.=20 There may also be poster sessions, subject to demand. We also strongly = encourage system demonstrations, either in conjunction with contentful = paper presentations or as stand-alone demos during the lunch and coffee = intervals. Please make it clear in your submissions whether you plan to = demonstrate your system, either as part of a paper presentation, or as a = stand-alone demo.=20 Publication There will be a common publication format for all workshops in line with = the main conference proceedings. Please follow the guidelines for the = main conference. However, it is anticipated that relevant publishers = will be approached to see if there would be interest in publishing the = proceedings more widely.=20 Important Dates a.. January 2001 First call for papers/demos=20 b.. 15.4.2001 Deadline for receipt of papers=20 c.. 31.5.2001 Notification of acceptance=20 d.. 15.7.2001 Final Paper due=20 e.. 18.9.2001 Workshop takes place=20 Attendance Fee Details of registration procedures, including registration fees, have = just been announced. The attendance fee for our workshop is Euro 50.=20 Organizing Committee a.. Sivaji Bandyopadhyay, India=20 b.. Ralf Brown, USA=20 c.. Michael Carl, Germany=20 d.. Ilyas Cicekli, Turkey=20 e.. Brona Collins, Belgium=20 f.. Oliver Streiter, Taiwan=20 g.. Stephan Vogel, Germany=20 h.. Andy Way, Ireland=20 -------------------------------------------------------------------------= ------- ------=_NextPart_000_000B_01C0B9E8.BA9EE2A0 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable WORKSHOP ON EXAMPLE BASED MACHINE TRANSLATION
 
 

Workshop on Example-Based Machine Translation: 2nd Call for=20 Papers

Hosted by MT-SUMMIT=20 VIII
Santiago de Compostella, Spain,
September 18-22, 2001
http://www.eamt.org/summitVIII/in= dex.html
http://www.compapp.dcu= .ie/~away/EBMT.html

Co-chairs: Michael Carl, = IAI,=20 Saarbr=FCcken; Andy Way, = Computer=20 Applications, Dublin City University=20

In recent years, corpora of multilingual translated texts have become = widely=20 available for a number of languages. Notwithstanding the seminal paper = by Nagao=20 (84), it is primarily since the early 90's that such bilingual texts = have been=20 exploited in the area of Machine Translation (MT).=20

The two main paradigmatic approaches which have been proposed are=20 Statistics-based Machine Translation (SBMT) and Example-Based Machine=20 Translation (EBMT). A related variant of EBMT that we ignore here, = despite being=20 widely used in the localisation area, is that of Translation Memories = (TM). No=20 new translations are created afresh from previously existing examples in = the=20 system database: rather, the closest translation matches are proposed to = the=20 user for post-editing into the correct translation.=20

While translation memory systems are used in restricted domains, SBMT = systems=20 require training on huge, good quality bilingual corpora. As a = consequence TMs=20 can hardly be applied as a general purpose solution to MT and SBMT as = yet cannot=20 produce complex translations to the desired quality, even if such = translations=20 are given to the system in the training phase. EBMT seeks to exploit and = integrate a number of knowledge resources, such as linguistics and = statistics,=20 and symbolic and numerical techniques, for integration into one = framework. In=20 this way, rule-based morphological, syntactic and/or semantic = information is=20 combined with knowledge extracted from bilingual texts which is then = re-used in=20 the translation process.=20

However, it is unclear how one might combine the different knowledge=20 resources and techniques in an optimal way. In EBMT, therefore, the = question is=20 asked: what can be learned from a bilingual corpus and what needs to be = manually=20 provided? Furthermore, we remain uncertain as to how far the EBMT = methodology=20 can be pushed with respect to translation quality and/or translation = purpose.=20 Finally, one wonders what the implications and consequences are for size = and=20 quality of the reference translations, (computational) complexity of the = system,=20 sizeability and transportability, if such an approach is taken.=20

Given this background, we propose to organize a workshop in order to = shed=20 some light on these open questions, among others. We are seeking = contributions=20 which go beyond the purely statistical and/or rule-based approaches to = MT. We=20 welcome visionary and technical descriptions, reports of empirical = research as=20 well as feasibility studies and system demonstrations. We would welcome=20 contributions on any of the following topics and sub-headings:=20

  • (semi-)automatic preparation of existing bi/multilingual corpora = for EBMT=20
    • extraction of bi/multilingual texts from the web=20
    • preparation of treebanks for EBMT=20
    • bi/multilingual alignment/bracketing/parsing=20
    • inference of bi/multilingual grammar and transfer rules =
  • description of `pure' EBMT systems=20
    • knowledge resources used=20
    • representation of numeric and symbolic knowledge
  • descriptions of `hybrid' systems integrating EBMT with rule-based = or other=20 methodologies=20
  • evaluation of EBMT results and/or comparison with other MT systems =
  • considerations on domain-(in)dependence of EBMT systems=20
  • computational and/or system complexity of EBMT systems

Submissions

Submitted papers must describe original, previously unpublished work. = Submissions must not exceed 12 pages. Contributions should be submitted = to Michael Carl. Conference = stylesheets are=20 now available. Deadlines appear below.=20

There may also be poster sessions, subject to demand. We also = strongly=20 encourage system demonstrations, either in conjunction with contentful = paper=20 presentations or as stand-alone demos during the lunch and coffee = intervals.=20 Please make it clear in your submissions whether you plan to demonstrate = your=20 system, either as part of a paper presentation, or as a stand-alone = demo.=20

Publication

There will be a common publication format for all workshops in line = with the=20 main conference proceedings. Please follow the guidelines for = the main=20 conference. However, it is anticipated that relevant publishers will be=20 approached to see if there would be interest in publishing the = proceedings more=20 widely.=20

Important Dates

  • January 2001 First call for papers/demos=20
  • 15.4.2001 Deadline for receipt of papers=20
  • 31.5.2001 Notification of acceptance=20
  • 15.7.2001 Final Paper due=20
  • 18.9.2001 Workshop takes place

Attendance Fee

Details of registration = procedures,=20 including registration fees, have just been announced. The attendance = fee for=20 our workshop is Euro 50.=20

Organizing Committee

 

------=_NextPart_000_000B_01C0B9E8.BA9EE2A0-- From mlf@dlsi.ua.es Mon Apr 2 10:47:04 2001 From: mlf@dlsi.ua.es (Mikel L. Forcada) Date: Mon, 02 Apr 2001 11:47:04 +0200 Subject: [MT-List] Reminder: Teaching MT workshop at MT Summit VII Message-ID: <3AC84A98.29BD079C@dlsi.ua.es> [Please apologize if you receive multiple copies of this call for papers] Teaching Machine Translation A workshop at Machine Translation Summit VIII September 18-22, 2001, Santiago de Compostela, Spain CALL FOR PAPERS Introduction In September 2001, the workshop Teaching Machine Translation will be held as part of the VIII Machine Translation Summit in Santiago de Compostela, Galicia, Spain. The workshop will provide an opportunity for machine translation instructors attending MT Summit VIII to exchange their experience by presenting papers or demonstrations describing the tools and techniques they use in the classroom or in the laboratory. Schedule and venue This one-day workshop will take place on September 22, 2001, after the regular conference sessions. Please visit the Workshop website at http://www.dlsi.ua.es/tmt/ for updates. Papers The Teaching Machine Translation workshops seeks original papers in all aspects of the instruction of machine translation. Topics of interest include, but are not limited to: * why and to whom should MT be taught? * teaching the theoretical background of MT: linguistics, computer science, translation theory * addressing preconceptions about MT in the classroom * the use of commercial MT programs in hands-on teaching * teaching machine translation strategies to non-computer science majors * web-based distance learning of MT * teaching MT as a component of multilingual information resources and as a part of the entire documentation process * MT education and industry: bridging the gap between academia and the real world * teaching pre- and post-MT skills to MT users * teaching MT-related techniques: translation memory, alignment tools, terminological databases, use of corpora, etc. * teaching MT evaluation * building modules or `toy' MT systems in the laboratory * experiences on the evaluation of MT instruction The working language of the workshop will be English. Papers should describe research or experiences in any of the topics mentioned, and should not be longer than 3000 words. In view of the main topics of the workhop, presentations may have a substantial demonstrative component. Papers should be submitted to the address below, preferably by e-mail, and must be received by the indicated date. Papers should include a cover page with the following information: * paper title, * author(s)' name(s), affiliation(s), address(es), and e-mail address(es), * 100 word abstract, * up to 5 keywords, * for demonstrative presentations: the hardware, software and network requirements for the demonstration. Please submit electronically to tmt@dlsi.ua.es, following the guidelines for authors provided at the Teaching Machine Translation website http://www.dlsi.ua.es/tmt/. If you cannot submit electronically, please mail 4 hardcopies of the paper to: Mikel L. Forcada Teaching Machine Translation at MT Summit VIII Dept. Llenguatges i Sistemes Informàtics Universitat d'Alacant E-03071 Alacant, Spain Important Dates 15 April 2001 Paper submission deadline 31 May 2001 Notifications 1 July 2001 Final camera-ready copy deadline About Santiago and the Venue Santiago de Compostela in North West Spain (Galicia) was the most visited pilgrimage shrine in the Middle Ages, and to this day the Road to Santiago is a popular route for walkers through France and northern Spain. The goal is the imposing 11th century cathedral containing the relics of St James the Great. But the well preserved historical centre of the city offers much more: medieval streets, excellent restaurants, ancient churches, monasteries, palaces, etc. The venue for the conference is the Hostal de los Reyes Católicos, founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style, and now a hotel (the oldest and one of the most beautiful in the world). Those staying at the Hostal will have a once-in-a-lifetime experience, but there is also accommodation nearby in many other good hotels, all within easy walking distance of the centre and the Hostal. Santiago de Compostela has its own airport with daily direct flights from Barcelona, Madrid, Brussels, and London. Workshop Organization Workshop Co-chairmen Derek R. Lewis, University of Exeter, UK. d.r.lewis@exeter.ac.uk Mikel L. Forcada, Universitat d'Alacant, Spain. mlf@dlsi.ua.es Organizing committee F. Gaspari, Univ. Bologna, Italy M.H. Møller, Syddansk Univ., Denmark J.A. Pérez-Ortiz, Univ. Alacant Further Information For more details, please visit the workshop website: http://www.dlsi.ua.es/tmt/ . You may also send a request for information to tmt@dlsi.ua.es . -- _____________________________________________________________________ Mikel L. Forcada E-mail: mlf@dlsi.ua.es Departament de Llenguatges Phone: +34-96-590-3400 ext. 3384; i Sistemes Informàtics also +34-96-590-3772. UNIVERSITAT D'ALACANT Fax: +34-96-590-9326, -3464 E-03071 ALACANT, Spain. URL: http://www.dlsi.ua.es/~mlf From malek.boualem@rd.francetelecom.fr Thu Apr 5 15:59:44 2001 From: malek.boualem@rd.francetelecom.fr (BOUALEM Malek FTRD/DMI/LAN) Date: Thu, 5 Apr 2001 16:59:44 +0200 Subject: [MT-List] URL's of Meteo, Alps, Titus and Eurotra ? Message-ID: Hi, Does anyone know the URLs of these MT systems ? [METEO] a voir si possible [ALPS] a voir si possible [TITUS] a voir si possible [EUROTRA] a voir si possible Thanks Malek From steven@let.uu.nl Thu Apr 5 12:48:07 2001 From: steven@let.uu.nl (steven@let.uu.nl) Date: Thu, 5 Apr 2001 13:48:07 +0200 (MEST) Subject: [MT-List] MT Summit Workshop MT2010: Final Call for Papers Message-ID: <200104051148.f35Bm7S12889@sfinx.let.uu.nl> Final Call for Workshop Contributions Analytical, critical and visionary papers wanted: Where do we stand, where do we go?! Submission deadline: April 15 MT 2010 -- Towards a Road Map for MT September 18, 2001 Workshop organised by ELSNET and hosted by MT Summit VIII Santiago de Compostela, Spain September 18-22 2001 Aim and context: The aim of the workshop is to contribute to ELSNET's ongoing action to establish a road map for MT for the next decade. A road map comprises an analysis of the present situation, a vision of where we want to be in ten years from now, and a number of intermediate milestones that would help in setting intermediate goals and in measuring our progress towards our goals. The function of the road map is not to impose anything on anyone, but rather to provide a broadly supported definition of a context in which to position the MT community's efforts, which would allow us to identify common priorities for joint activities in e.g. research, resources and training. Scope: Papers are invited that * give critical analyses of the present state of the art in machine translation of written and spoken language, * present visions of the future of machine translation, both from a technological and from an application point of view, or * identify major milestones and challenges on our way towards the future, and/or ways to measure our progress along the road. Format: It will be a half- or full-day workshop, with three sessions: * The first session will aim at giving a critical analysis of the present situation. * The second session will be dedicated to visions of the future. * The third session will aim at establishing intermediate milestones (e.g. timelines, necessary research achievements), methods to evaluate progress along the road, and training requirements to be met in order to reach our goals. Audience: The primary audience consists of people with an analytical or future oriented, programmatic interest, both from research and from industry. Workshop schedule: * 15 April 2001: Submission deadline * 31 May 2001: Notification of acceptance * 15 July: Submission of final versions Submission details: Papers should be sent in electronic form to steven.krauwer@elsnet.org. For the submission format we follow the same guidelines as the main conference: http://www.eamt.org/summitVIII/author.html Core Programme/Organisation Committee: * Steven Krauwer (ELSNET, Netherlands) Chair * Niels Ole Bernsen (NIS, Denmark) * Paul Heisterkamp (Daimler-Chrysler, Germany) * Jose Pardo (UPM, Spain) * Pierre Isabelle (XEROX, France) * Arjan van Hessen (IP Globalnet, Netherlands) * Antonio Zampolli (U. Pisa, Italy) * Hans Uszkoreit (DFKI, Germany) * Oliviero Stock (IRST, Italy) * Susan Armstrong (ISSCO/ETI, Switzerland) * Herman Caeyers (LANT, Belgium) * Nuria Bel (gilcUB, Spain) * Ed Hovy (ISI, USA) Workshp URL: http://www.elsnet.org/mt2010.html Contact details: Steven Krauwer email: steven.krauwer@elsnet.org ELSNET / UiL OTS www: http://www.elsnet.org Trans 10 phone: +31 30 253 6050 3512 JK Utrecht, NL fax: +31 30 253 6000 From WJHutchins@compuserve.com Mon Apr 9 17:21:38 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 9 Apr 2001 12:21:38 -0400 Subject: [MT-List] FIRST ANNOUNCEMENT & CALL FOR PAPERS : 8th INTERNATIONAL CONFERENCE ON TRANSLATION Message-ID: <200104091221_MC2-CBCF-E527@compuserve.com> FIRST ANNOUNCEMENT & CALL FOR PAPERS : 8th INTERNATIONAL CONFERENCE ON TR= ANSLATION = Dear Colleague With great pleasure we would like to inform you that the University Utara= Malaysia and Institute of Language and Literature (DBP), Malaysian Translators Association, and National Institute of Translation Malaysia will hold the 8th International Conference on Translation in September, 2001 at Langkawi, Malaysia. We would be very grateful if you could share= this information with your colleagues. Theme: "Translation in management and information technology: Towards developing a k-economy" Topics: 1. Issues and challeges in translating management and information technology materials. 2. The applications of technology in translation. 3. Translation as field of trade. 4. The management and quality control in translating. 5. Problems in the translation of management and information technology materials. 6. The interpretation and application of translation in management and information technology in the era of globalisation. 7. The roles of translators in the transfer of technology. 8. The importance of translation in the development of a k-economy. 9. The legal implications in translation. 10. The role of publishers as a catalyst of the translation process. Please visit our web site http://web.uum.edu.my/ict/ Thank you. Dato' Prof. Dr. Mohd Yusof Hasan Director of University Utara Malaysia Press c/o Director of The 8th International Conference on Translation From Christian.Boitet@imag.fr Tue Apr 10 08:52:57 2001 From: Christian.Boitet@imag.fr (Christian Boitet) Date: Tue, 10 Apr 2001 09:52:57 +0200 Subject: [MT-List] No info/access! 1st CFP : 8th INTERNATIONAL CONFERENCE ON TRANSLATION In-Reply-To: <200104091221_MC2-CBCF-E527@compuserve.com> References: <200104091221_MC2-CBCF-E527@compuserve.com> Message-ID: Dear John, 10/4/01 >FIRST ANNOUNCEMENT & CALL FOR PAPERS : 8th INTERNATIONAL CONFERENCE >ON TRANSLATION > > >Dear Colleague > >With great pleasure we would like to inform you that the University Utara >Malaysia and Institute of Language and Literature (DBP), Malaysian >Translators Association, and National Institute of Translation Malaysia >will hold the 8th International Conference on Translation in September, >2001 at Langkawi, Malaysia. We would be very grateful if you could share >this information with your colleagues. > >Theme: "Translation in management and information technology: Towards >developing a k-economy" > >Topics: >1. Issues and challeges in translating management and information >technology materials. >2. The applications of technology in translation. >3. Translation as field of trade. >4. The management and quality control in translating. >5. Problems in the translation of management and information technology >materials. >6. The interpretation and application of translation in management and >information technology in the era of globalisation. >7. The roles of translators in the transfer of technology. >8. The importance of translation in the development of a k-economy. >9. The legal implications in translation. >10. The role of publishers as a catalyst of the translation process. > >Please visit our web site http://web.uum.edu.my/ict/ Sorry, but=8A 1) That site is not accessible "421 Sorry, Access Denied by Firewall Access Rules. " 2) An announcement without dates and confernce web site is a kind of joke!!! > > >Thank you. > > >Dato' Prof. Dr. Mohd Yusof Hasan >Director of University Utara Malaysia Press >c/o Director of The 8th International Conference on Translation > >-- > For MT-List info, see http://www.eamt.org/mt-list.html Best regards, CB -- ------------------------------------------------------------------------- Christian Boitet (Pr. Universite' Joseph Fourier) Tel: +33.4-7651-4355/4817 GETA, CLIPS, IMAG-campus, BP53 Fax: +33.4-7651-4405 385, rue de la Bibliothe`que Mel: Christian.Boitet@imag.fr 38041 Grenoble Cedex 9, France Mobile: +33-(0)6-6005-1969 http://www-clips.imag.fr/geta/christian.boitet ------------------------------------------------------------------------- Projet C-STAR (http://www.c-star.org/) et projet europe'en Nespole (http://nespole.itc.it) de traduction de parole Projet UNL de communication et recherche d'information multilingue sur le re'seau http://www.unl.ias.unu.edu ou http://www.unl.org From panissod@systran.fr Fri Apr 13 14:37:01 2001 From: panissod@systran.fr (Christiane PANISSOD) Date: Fri, 13 Apr 2001 15:37:01 +0200 Subject: [MT-List] DEADLINE EXTENSION - Workshop on Machine Translation Message-ID: >>EXTENDED SUBMISSION DEADLINE : 30 April 2001 << CALL FOR PARTICIPATION Workshop on Machine Translation "Machine Translation and Large-Scale Applications" 5 July 2001 http://www.systransoft.com/TALN2001-en.html in conjunction with TALN'2001(2-5 July 2001) TOURS Europe - FRANCE MOTIVATION Against the backdrop of an increasingly multilingual society, the Machine Translation Workshop of TALN'2001 looks at the main challenges to Machine Translation and multilingual NLP. Indeed, MT has seen a remarkable development for the past few years in terms of the number of translation requests (about a million translation requests everyday on the Web) as well as in terms of the different types of formats translated : translation of dynamic resources(FAQ, Web pages, daily papers), emails, requests for search engines, etc. These different types of translations demand different levels of quality, and induce an important associated feedback in the form of additional bilingual dictionaries. At the same time there is a growing need for more specialized translation. For this "specialization" the construction of wide-covering mutlilingual resources is an essential component of this more demanding translation. Because of the increase of the number of contributors (experts or users), the questions of exchange, reusability and incrementability of the translation resources had to be raised. The associated investment (public or private) lead to a general tendency to initiate centralizations (cf. Elra), standards (ISO 12220), transfer formalisms (OLIF), and more generally, to create mutlilingual data exchange formats (UNL). In any case, the choice of a linguistic data support cannot be separated from pragmatic considerations concerning the scale, the volume of additional resources becoming more important than the generic resources. These resources can vary from "simple" terminological resources (aligned lexicons) to more complex resources such as terminological grammars, transfer dictionaries, and graphs/grammars of translation. FOCUS OF THE WORKSHOP The focus of the workshop is not only recent machine translation research and products, but also the latest developments in multilingual language technology. The workshop aims at considering the questions of the construction, the validation and the utilization of large-scale resources applied to machine translation. The papers submitted can be related to the different topics below, but also to some other closely related topics : - use of standardization formalisms on important volumes of data - evaluation and realization of construction tools of such resources (from corpus, monolingual resources...) - integration and conversion of existing resources - validation and "expert" improvement processes - massive parallel acquisition of terminology in the framework of a network of users - management of mutlilingual resources compared to bilingual resources - multilingual extraction and machine translation QUESTIONS AND ISSUES You can either submit an abstract of 5 pages, or a complete version of your paper(up to 10 pages). Submitted papers should be in French or in English (for non French speakers). The abstracts and the papers submitted should conform to the submission format of TALN'2001 (style sheets are available on the TALN'2001 site http://www.li.univ-tours.fr/taln-recital-2001/ ) : Times 12, single-spaced, 10 pages at the most, including figures, examples and references. Authors should send their submission as a file attached to an e-mail (ps, pdf, rtf files, A4 format) at the following e-mail address : mailto:workshop-taln@systran.org The e-mail should contain the following information: submission title, authors' names and affiliation. Submissions will be reviewed by two experts of the Program Comittee. The final version of the accepted papers will be published in the workshop proceedings. The papers will give rise to 20 minute presentations followed by 10 minutes of questions. PROGRAM COMITTEE Christian Boitet, CLIPS-IMAG,equipe GETA, Grenoble http://clips.imag.fr/geta/ Maurice Gross, LADL & UMLV, Marne-la-Vallee Igor Boguslavskij, IPPI PAN, Moscou Georges Carayannis, ELRA & ILSP, Athen Joseph Dichy, Faculte des Langues, Universite Lyon II Pierre-Yves Foucou, SYSTRAN, Paris http://www.systransoft.com/ Daniel Grasmick, SAP, Waldorf Jean Senellart, SYSTRAN, Paris http://www.systransoft.com/ Tamas Varadi, Linguistics Institute, Hungarian Academy of Sciences Remi Zajac, CRL/NMSU, Las Cruces, http://crl.nmsu.edu/~rzajac/ IMPORTANT DATES Submission deadline: 30/04/2001 Notification to authors: 07/04/2001 Final version due (camera-ready) : 20/05/2001 Workshop : 5 July 2001 From nikolay@npp.cit.bg Tue Apr 17 13:52:46 2001 From: nikolay@npp.cit.bg (Nikolay Ivanov) Date: Tue, 17 Apr 2001 14:52:46 +0200 Subject: [MT-List] Bulgarian Machine Translation Software Message-ID: <021801c0c73d$50dd8080$340310ac@zska> This is a multi-part message in MIME format. ------=_NextPart_000_0213_01C0C74E.105730E0 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: quoted-printable Hi, all! Has anyone ever heard about such software, I do not mean simply Word = Translator tools available on the Internet? Thank you in advance, Nikolay from Bulgaria ------=_NextPart_000_0213_01C0C74E.105730E0 Content-Type: text/html; charset="koi8-r" Content-Transfer-Encoding: quoted-printable
Hi, all!
Has anyone ever heard about such software, I do not mean simply = Word=20 Translator tools available on the Internet?
Thank you in advance,
Nikolay from Bulgaria
------=_NextPart_000_0213_01C0C74E.105730E0-- From WJHutchins@compuserve.com Tue Apr 17 16:49:40 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Tue, 17 Apr 2001 11:49:40 -0400 Subject: [MT-List] Bulgarian Machine Translation Software Message-ID: <200104171149_MC2-CCC5-BF03@compuserve.com> Dear Mr Ivanov, as far as I know -- checking my own "Compendium of translation software" [see EAMT website] -- there are no commercial MT systems for Bulgarian. All there are are dictionaries (of perhaps dubious quality) offering some 'translation help'. John Hutchins 17 April From nikolay@npp.cit.bg Wed Apr 18 07:47:54 2001 From: nikolay@npp.cit.bg (Nikolay Ivanov) Date: Wed, 18 Apr 2001 08:47:54 +0200 Subject: [MT-List] Bulgarian Machine Translation Software References: <021801c0c73d$50dd8080$340310ac@zska> <00aa01c0c756$6bfc8500$baa9c818@videotron.ca> Message-ID: <008b01c0c7d3$8282a200$340310ac@zska> This is a multi-part message in MIME format. ------=_NextPart_000_0088_01C0C7E4.420D1B40 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: quoted-printable Original Message -----=20 From: Michael Blekhman=20 To: Nikolay Ivanov=20 Dear Nikolay, We specialize in MT for Slavic languages (www.ling98.com), and we = would be happy to tackle Bulgarian - should such a project be = financed... Sincerely, Michael Blekhman, Lingvistica '98 Inc. Mr. Blackman, The Penthagon has sponsored the creation of automatic translation = software in Serbian-English in 1998 according to a Web source I found = recently. So it is our turn ;-). John Hutchins wrote on 17 April Dear Mr Ivanov, as far as I know -- checking my own "Compendium of translation software" [see EAMT website] -- there are no commercial MT systems for Bulgarian. All there are dictionaries (of perhaps dubious quality) offering some 'translation help'. http://www.tranexp.com:2000/InterTran are offering word by word = translation in Eng-Bg/Bg-Eng but the result is awful. A Bulgarian = company Bultra (www.bultra.com) offers an Eng-Bg package at the = incredible price of 590$.=20 Best regards, Nikolay =20 ------=_NextPart_000_0088_01C0C7E4.420D1B40 Content-Type: text/html; charset="koi8-r" Content-Transfer-Encoding: quoted-printable
 
 Original Message -----
From:=20 Michael=20 Blekhman
 
Dear Nikolay,
We specialize in MT for Slavic = languages (www.ling98.com), and we would be = happy to=20 tackle Bulgarian - should such a project be financed...
Sincerely,
Michael Blekhman,
Lingvistica '98 Inc.
 
Mr. Blackman,
The Penthagon has sponsored the = creation of=20 automatic translation software in Serbian-English in 1998 according to = a Web=20 source I found recently. So it is our turn ;-).
 
John Hutchins wrote on 17 April
 
Dear Mr Ivanov,

as far as I know -- checking my own=20 "Compendium
of translation software" [see EAMT website] -- = there
are no=20 commercial MT systems for Bulgarian. All there
are dictionaries (of = perhaps=20 dubious quality)
offering some 'translation help'.
http://www.tranexp.com:200= 0/InterTran are=20 offering word by word translation in Eng-Bg/Bg-Eng but the result is = awful. A=20 Bulgarian company Bultra (www.bultra.com) offers an = Eng-Bg package=20 at the incredible price of 590$. 
Best regards,
Nikolay
 
------=_NextPart_000_0088_01C0C7E4.420D1B40-- From WJHutchins@compuserve.com Wed Apr 18 14:33:54 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Wed, 18 Apr 2001 09:33:54 -0400 Subject: [MT-List] MT Software Message-ID: <200104180934_MC2-CCEC-2A1B@compuserve.com> Dear Mr Gojol, There are no MT systems involving Rumanian in the Compendium. John Hutchins 18 April ---------------- NB (in general to MT-list) It was to answer such questions that I posted the "Compendium" on the EAMT website. Otherwise I could be spending = days answering MT queries. The Compendium will be updated regularly. From steven.krauwer@let.uu.nl Wed Apr 18 16:07:22 2001 From: steven.krauwer@let.uu.nl (Steven Krauwer) Date: Wed, 18 Apr 2001 17:07:22 +0200 (MEST) Subject: [MT-List] MT Roadmap: deadline extension Message-ID: <200104181507.f3IF7Mi16506@sfinx.let.uu.nl> Yet another deadline extension! Final Call for Workshop Contributions Analytical, critical and visionary papers wanted about all aspects of spoken and written Machine Translation: Where do we stand, where do we go?! We need more papers and wider coverage! Submission deadline extended to May 13 MT 2010 -- Towards a Road Map for MT September 18, 2001 Workshop organised by ELSNET and hosted by MT Summit VIII Santiago de Compostela, Spain September 18-22 2001 Aim and context: The aim of the workshop is to contribute to ELSNET's ongoing action to establish a road map for MT for the next decade. A road map comprises an analysis of the present situation, a vision of where we want to be in ten years from now, and a number of intermediate milestones that would help in setting intermediate goals and in measuring our progress towards our goals. The function of the road map is not to impose anything on anyone, but rather to provide a broadly supported definition of a context in which to position the MT community's efforts, which would allow us to identify common priorities for joint activities in e.g. research, resources and training. Scope: Papers are invited that * give critical analyses of the present state of the art in machine translation of written and spoken language, * present visions of the future of machine translation, both from a technological and from an application point of view, or * identify major milestones and challenges on our way towards the future, and/or ways to measure our progress along the road. Format: It will be a half- or full-day workshop, with three sessions: * The first session will aim at giving a critical analysis of the present situation. * The second session will be dedicated to visions of the future. * The third session will aim at establishing intermediate milestones (e.g. timelines, necessary research achievements), methods to evaluate progress along the road, and training requirements to be met in order to reach our goals. Audience: The primary audience consists of people with an analytical or future oriented, programmatic interest, both from research and from industry. Workshop schedule: * 13 May 2001: Submission deadline * 31 May 2001: Notification of acceptance * 15 July: Submission of final versions Submission details: Papers should be sent in electronic form to steven.krauwer@elsnet.org. For the submission format we follow the same guidelines as the main conference: http://www.eamt.org/summitVIII/author.html Core Programme/Organisation Committee: * Steven Krauwer (ELSNET, Netherlands) Chair * Niels Ole Bernsen (NIS, Denmark) * Paul Heisterkamp (Daimler-Chrysler, Germany) * Jose Pardo (UPM, Spain) * Pierre Isabelle (XEROX, France) * Arjan van Hessen (IP Globalnet, Netherlands) * Antonio Zampolli (U. Pisa, Italy) * Hans Uszkoreit (DFKI, Germany) * Oliviero Stock (IRST, Italy) * Susan Armstrong (ISSCO/ETI, Switzerland) * Herman Caeyers (LANT, Belgium) * Nuria Bel (gilcUB, Spain) * Ed Hovy (ISI, USA) Workshp URL: http://www.elsnet.org/mt2010.html Contact details: Steven Krauwer email: steven.krauwer@elsnet.org ELSNET / UiL OTS www: http://www.elsnet.org Trans 10 phone: +31 30 253 6050 3512 JK Utrecht, NL fax: +31 30 253 6000 From jobs@amikai.com Fri Apr 20 00:49:21 2001 From: jobs@amikai.com (Raymond Flournoy) Date: Thu, 19 Apr 2001 16:49:21 -0700 Subject: [MT-List] Job announcement: Computational Linguist/Linguistic Developer in San Francisco, CA, USA Message-ID: Hello, My company, Amikai, Inc. is looking for a Computational Linguist/Linguistic Developer to join our team beginning this summer. Information about the position is attached below, and information about the company can be found at our website: www.amikai.com. Interested applicants are invited to submit a resume to jobs@amikai.com. Thank you! --Raymond S. Flournoy, Ph.D. Director of Computational Linguistics Amikai, Inc. ray.flournoy@amikai.com +1-415-863-8534 ------------------------------------ Computational Linguist/Linguistic Developer JOB DESCRIPTION As Computational Linguist you would be responsible for designing and developing sophisticated international applications for processing and analyzing human languages. Key responsibilities: * Develop new applications and new features for existing applications to improve Amikai's real-time translation products and services * Actively participate in design discussions and work with team members to define and coordinate project goals and requirements * Design and implement high quality code * Learn new technologies in a short timeframe REQUIREMENTS * BS in Computer Science or equivalent field, with special emphasis in Natural Language Processing * Interest in and familiarity with basic concepts in Natural Language Processing, such as parsing, grammars, corpus analysis, information retrieval, and/or HMMs * Strong programming skills are a must * Strong Object Oriented background is a plus, Java preferred * Dedication to elegant, well thought-out solutions and high quality implementations * Excellent written and verbal communication skills * Bi- or multilingualism is a plus From mlf@dlsi.ua.es Fri Apr 20 09:26:06 2001 From: mlf@dlsi.ua.es (Mikel L. Forcada) Date: Fri, 20 Apr 2001 10:26:06 +0200 Subject: [MT-List] DEADLINE EXTENDED: Teaching Machine Translation at MT Summit VIII Message-ID: <3ADFF29E.5E647F34@dlsi.ua.es> [Please accept our apologies if you receive multiple copies of this message] ********************************************************** ********** DEADLINE EXTENDED: April 30, 2001 ************* ********************************************************** Teaching Machine Translation A workshop at Machine Translation Summit VIII September 18-22, 2001, Santiago de Compostela, Spain CALL FOR PAPERS Introduction In September 2001, the workshop Teaching Machine Translation will be held as part of the VIII Machine Translation Summit in Santiago de Compostela, Galicia, Spain. The workshop will provide an opportunity for machine translation instructors attending MT Summit VIII to exchange their experience by presenting papers or demonstrations describing the tools and techniques they use in the classroom or in the laboratory. Schedule and venue This one-day workshop will take place on September 22, 2001, after the regular conference sessions. Please visit the Workshop website at http://www.dlsi.ua.es/tmt/ for updates. Papers The Teaching Machine Translation workshops seeks original papers in all aspects of the instruction of machine translation. Topics of interest include, but are not limited to: * why and to whom should MT be taught? * teaching the theoretical background of MT: linguistics, computer science, translation theory * addressing preconceptions about MT in the classroom * the use of commercial MT programs in hands-on teaching * teaching machine translation strategies to non-computer science majors * web-based distance learning of MT * teaching MT as a component of multilingual information resources and as a part of the entire documentation process * MT education and industry: bridging the gap between academia and the real world * teaching pre- and post-MT skills to MT users * teaching MT-related techniques: translation memory, alignment tools, terminological databases, use of corpora, etc. * teaching MT evaluation * building modules or `toy' MT systems in the laboratory * experiences on the evaluation of MT instruction The working language of the workshop will be English. Papers should describe research or experiences in any of the topics mentioned, and should not be longer than 3000 words. In view of the main topics of the workhop, presentations may have a substantial demonstrative component. Papers should be submitted to the address below, preferably by e-mail, and must be received by the indicated date. Papers should include a cover page with the following information: * paper title, * author(s)' name(s), affiliation(s), address(es), and e-mail address(es), * 100 word abstract, * up to 5 keywords, * for demonstrative presentations: the hardware, software and network requirements for the demonstration. Please submit electronically to tmt@dlsi.ua.es, following the guidelines for authors provided at the Teaching Machine Translation website http://www.dlsi.ua.es/tmt/. If you cannot submit electronically, please mail 4 hardcopies of the paper to: Mikel L. Forcada Teaching Machine Translation at MT Summit VIII Dept. Llenguatges i Sistemes Informàtics Universitat d'Alacant E-03071 Alacant, Spain Important Dates 30 April 2001 **** NEW PAPER SUBMISSION DEADLINE **** (was April 15) 31 May 2001 Notifications 1 July 2001 Final camera-ready copy deadline About Santiago and the Venue Santiago de Compostela in North West Spain (Galicia) was the most visited pilgrimage shrine in the Middle Ages, and to this day the Road to Santiago is a popular route for walkers through France and northern Spain. The goal is the imposing 11th century cathedral containing the relics of St James the Great. But the well preserved historical centre of the city offers much more: medieval streets, excellent restaurants, ancient churches, monasteries, palaces, etc. The venue for the conference is the Hostal de los Reyes Católicos, founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style, and now a hotel (the oldest and one of the most beautiful in the world). Those staying at the Hostal will have a once-in-a-lifetime experience, but there is also accommodation nearby in many other good hotels, all within easy walking distance of the centre and the Hostal. Santiago de Compostela has its own airport with daily direct flights from Barcelona, Madrid, Brussels, and London. Workshop Organization Workshop Co-chairmen Derek R. Lewis, University of Exeter, UK. d.r.lewis@exeter.ac.uk Mikel L. Forcada, Universitat d'Alacant, Spain. mlf@dlsi.ua.es Organizing committee F. Gaspari, Univ. Bologna, Italy M.H. Møller, Syddansk Univ., Denmark J.A. Pérez-Ortiz, Univ. Alacant Further Information For more details, please visit the workshop website: http://www.dlsi.ua.es/tmt/ . You may also send a request for information to tmt@dlsi.ua.es . -- _____________________________________________________________________ Mikel L. Forcada E-mail: mlf@dlsi.ua.es Departament de Llenguatges Phone: +34-96-590-3400 ext. 3384; i Sistemes Informàtics also +34-96-590-3772. UNIVERSITAT D'ALACANT Fax: +34-96-590-9326, -3464 E-03071 ALACANT, Spain. URL: http://www.dlsi.ua.es/~mlf From rjtonge@hotmail.com Wed Apr 25 20:31:52 2001 From: rjtonge@hotmail.com (Richard Tonge) Date: Wed, 25 Apr 2001 19:31:52 -0000 Subject: [MT-List] MT: History- an overview? Message-ID: Hi, I was hoping someone might be able to help with this pretty simple request. I'm in my second year at Durham University and am doing an essay on MT, in particular it's history and present day history. It's more an overview. I presume someone will know a few good books or websites where I can get this info, or just write a few points for me. It'd be greatly appreciated. Reply on or offlist. Many thanks, Richard Tonge _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. From WJHutchins@compuserve.com Thu Apr 26 11:27:19 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Thu, 26 Apr 2001 06:27:19 -0400 Subject: [MT-List] MT: History- an overview? Message-ID: <200104260627_MC2-CDE3-68A0@compuserve.com> I have written a great deal about the history of MT, both books and articles. Most should be in your university library. Please see my personal website: http://ourworld.compuserve.com/homepages/WJHutchins There is a link to this from the EAMT website (www.eamt.org) John Hutchins 26 April =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D You wrote: > = Hi, I was hoping someone might be able to help with this pretty simple reques= t. I'm in my second year at Durham University and am doing an essay on MT, i= n = particular it's history and present day history. It's more an overview. I= = presume someone will know a few good books or websites where I can get th= is info, or just write a few points for me. It'd be greatly appreciated. Rep= ly on or offlist. Many thanks, Richard Tonge< From WJHutchins@compuserve.com Mon Apr 30 10:23:43 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 30 Apr 2001 05:23:43 -0400 Subject: [MT-List] MT Summit, September 2001 Message-ID: <200104300524_MC2-CE5A-BE35@compuserve.com> REMINDER Only two weeks before close of EARLY BOOKING (15 May) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D MT Summit VIII, Santiago de Compostela, Spain, 18-22 September 2001 The European Association for Machine Translation (EAMT) invites you to participate in the Eighth "Machine Translation Summit" which is being held in Santiago de Compostela, = Galicia, Spain from the 18th to the 22nd September 2001. The MT Summit conferences bring together the global MT = community to review the state of the art of machine = translation, computer-aided translation support tools and multilingual tools, both in the commercial = world and in the research sphere. It is a unique opportunity to learn about the latest developments in the field, for researchers, users and vendors to exchange experiences and to get valuable feedback, and for meeting others involved in this technology. There will be a full programme of papers, presentations = by invited speakers, panel sessions, exhibitions, demonstrations, and workshops (on the day preceding the conference and on the final day). There will be = excursions, a banquet and a folk music performance, and also some activities for spouses and partners of delegates. VENUE The conference is taking place in Santiago de Compostela in North West Spain(Galicia), famous as a pilgrimage destination since the early Middle Ages, a city full of ancient buildings, from its eleventh-century cathedral to its seventeenth and eighteenth century houses, convents and palaces. The old town has been listed as a 'world heritage city' by Unesco. The venue for the conference is the Hostal de los Reyes Cat=F3licos (Hostal dos Reis Cat=F3licos), originally founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style. It is now both the oldest and one of the most impressive hotels in the world. The opening session and the banquet will take place in the Comedor Real, other sessions will be in the hotel's auditorium and other rooms, and the exhibition and demonstrations will be housed in the Chapel. The opening reception will take place in the Pazo de Raxoi (City Council), an 18th century building next to the Hostal. HOTELS Santiago de Compostela is a major tourist centre attracting many visitors throughout the year. We have reserved accommodation in the Hostal itself and in a number of hotels close by (many restored 17th and 18th century buildings). All except one are within walking distance of the Hostal venue. (A bus will be provided for the Hotel Puerta del Camino). Because of the high demand for hotel rooms in Santiago and the substantial deposits, we ask for about half the full hotel charge when booking, and we encourage you to book early to be certain of your preference. WEBSITE INFORMATION For details of the hotels, excursions and workshops see the MT Summit website (www.eamt.org/summitVIII) or send email to: summitVIII@eamt.org. REGISTRATION Registration should preferably be made via the MT Summit website. Alternatively, you may ask for a registration form from: Carlos Paz (MT Summit Registration), Service de Traduction, European Commission, Rue de la Loi 200 - JECL 4/200, B-1049 Brussels, Belgium (Fax: +32-2-296.59.68; Telephone: +32-2-296.05.59; Email: Carlos.Paz-Carneiro@cec.eu.int). CONFERENCE PROGRAMME Tuesday, 18 September Workshops, welcome reception, opening of exhibitions Wednesday, 19 September opening session, first day of conference Thursday, 20 September Second day of conference Friday, 21 September Third day of conference, IAMT general assembly, folk music entertainment Saturday, 22 September Workshops, excursion to Coruna, special panel, banquet Sunday, 23 September Excursion to southern Galicia and northern Portugal CONFERENCE FEES The registration fees for the conference (19-22 September) are: IAMT members* Non-members Before 15 May Euro 250 Euro 285 After 15 May, before 15 July Euro 275 Euro 325 After 15 July Euro 325 Euro 375 * members of the Asia-Pacific Association for Machine Translation (AAMT), Association for Machine Translation in the Americas (AMTA), European Association for Machine Translation (EAMT) One Euro is approximately 0.90 US dollar. Note: The conference fees do not include meals, but there are numerous good-value restaurants nearby in the city, which you will have opportunity to enjoy during the conference. CANCELLATIONS Before 15 July: 75% of the registration fee will be refunded After 15 July and before 1 September: 50% of the registration fee will be refunded After 1 September: no refunds. EXHIBITIONS Exhibition space will be provided in the Chapel of the Hostal for up to 25 companies and will be active throughout the conference. Fees for exhibitors will be Euro 1200 (or Euro 1000 for early applicants). Contact Joerg Schuetz (joerg@iai.uni-sb.de) for details, or look at the Summit website (www.eamt.org/summitVIII). WORKSHOPS [for more details see Summit website] Tuesday 18 September fee Towards a Roadmap for MT (half day) Euro 25 Organiser: Steven Krauwer Machine interpretation and machine translation: different evaluation methods for different systems? (half day) Euro 25 Organiser: Susanne Jekat Example-based machine translation (full day) Euro 50 Organiser: Andrew Way Saturday 22 September Teaching machine translation (all day) Euro 50 Organisers: Mikel Forcada and Derek Lewis Interlinguas and interlingual approaches to MT (Fifth Interlingua workshop)(all day) Euro 50 Organisers: David Farwell and Steven Helmreich MT evaluation: who did what to whom (Fourth ISLE workshop) (all day) Euro 50 Organisers: Ed Hovy, Maghi King, Sandra Manzi and Florence Reeder OPEN MEETING on terminology exchange and synchronization (SALT, OSCAR, OLIF), Saturday 22 Sept (all day) Euro 50 [for more information see http://www.ttt.org] FLIGHT DISCOUNTS. Iberia Airlines is offering participants a 25% discount (or more) on many of its flights (for details see: www.eamt.org/summitVIII). On registration you will be sent a password, in order to take advantage of this offer. ORGANISERS John Hutchins (general chair): EAMT/IAMT president Bente Maegaard (programme chair) Joerg Schuetz (exhibitions) Harold Somers (workshops) Carlos Paz (registration and local arrangements) Anthony Clarke (finance): EAMT treasurer Viggo Hansen (publicity): EAMT secretary Colin Brace (website) From LorenW@agritope.com Mon Apr 30 19:44:40 2001 From: LorenW@agritope.com (Loren Walker) Date: Mon, 30 Apr 2001 11:44:40 -0700 Subject: [MT-List] Machine Translation Consultant Wanted -- Portland, Oregon Message-ID: I am looking for individual(s) with experience in the field of Machine Translation who live in or around Portland, Oregon. An educational nonprofit organization based in Portland is exploring the possiblity of using MT to facilitate online education partnerships between English and Spanish speaking teachers and students. We are in need of consultant(s) with MT experience to serve on our directory board. The position will NOT require a large time commitment. For more information please contact Loren Walker at LWW@graffiti.net (503) 872 9666 From Denise.Morgan@co.fairfax.va.us Mon Apr 30 19:57:27 2001 From: Denise.Morgan@co.fairfax.va.us (Denise.Morgan@co.fairfax.va.us) Date: Mon, 30 Apr 2001 14:57:27 -0400 Subject: [MT-List] other languages Message-ID: <0394E5BA21A1D411B7E600D0B774DC8E01D6C024@ffxex12.co.fairfax.va.us> At our library, we are interested in knowing if there are FREE machine translation capabilities for Hindi and Vietnamese. We are trying to conduct programs that display our library system website in languages that will help our residents. Altavista is a great start, but we'd love to find more FREE services. Thanks, Denise Morgan Kingstowne Library Alexandria, Virginia From WJHutchins@compuserve.com Tue May 1 11:30:44 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Tue, 1 May 2001 06:30:44 -0400 Subject: [MT-List] other languages Message-ID: <200105010631_MC2-CE82-88CF@compuserve.com> Dear Ms Morgan, You will find in the "Compendium of translation software" [available on EAMT website], which includes free services as well as software for purchase, that there are no systems for Hindi and Vietnamese -- let alone free ones! Please see Compendium for other free services (listed in index at end.) John Hutchins 1 May =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D = > At our library, we are interested in knowing if there are FREE machine translation capabilities for Hindi and Vietnamese. We are trying to conduct programs that display our library system website= in languages that will help our residents. Altavista is a great start, but we'd love to find more FREE services. Thanks, Denise Morgan Kingstowne Library Alexandria, Virginia < From ref@cs.cmu.edu Tue May 1 17:30:59 2001 From: ref@cs.cmu.edu (Robert Frederking) Date: Tue, 01 May 2001 12:30:59 -0400 Subject: [MT-List] other languages In-Reply-To: Your message of Tue, 01 May 2001 06:30:44 -0400. <200105010631_MC2-CE82-88CF@compuserve.com> Message-ID: <10863.988734659@lti.cs.cmu.edu> There surely must be a Hindi/English system for sale somewhere; probably in India, maybe advertized in Hindi on an Indian website? 1/2 :-) Also, the hardcopy of the compendium that I have (first edition) claims that a system called "Language Teacher" does Vietnamese -> English. Not for free, though. Bob From WJHutchins@compuserve.com Wed May 2 10:47:24 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Wed, 2 May 2001 05:47:24 -0400 Subject: [MT-List] other languages Message-ID: <200105020547_MC2-CEA4-B407@compuserve.com> Dear Bob, If you find one let me know. The Language Teacher is not a MT system, just an electronic dictionary. Any way I base my answers on the updated version of the Compendium (December 2000) -- see EAMT website. [A new update to come shortly!] John 2 May =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D >There surely must be a Hindi/English system for sale somewhere; probably in India, maybe advertized in Hindi on an Indian website? 1/2 :-) Also, the hardcopy of the compendium that I have (first edition) claims that a system called "Language Teacher" does Vietnamese -> English. Not for free, though. Bob< From vms@nde.vsnl.net.in Wed May 2 12:29:24 2001 From: vms@nde.vsnl.net.in (V.M.Sachdeva) Date: Wed, 2 May 2001 16:59:24 +0530 Subject: [MT-List] other languages Message-ID: <001801c0d2fb$da2a3f80$ea6436ca@v.m.sach> In 1999 ‘Anuvadak 1.0’ (Bilingual Translator) was launched for English-Hindi-English translation. It is priced. However, for latest free tools, if any, following URLs could be useful: http://www.tdil.gov.in/ http://www.cdacindia.com HTH V.M.Sachdeva >There surely must be a Hindi/English system for sale somewhere; >probably in India, maybe advertized in Hindi on an Indian website? > 1/2 :-) > >Also, the hardcopy of the compendium that I have (first edition) >claims that a system called "Language Teacher" does >Vietnamese -> English. > >Not for free, though. > > Bob > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > From regina_bauer@rocketmail.com Fri May 4 16:11:17 2001 From: regina_bauer@rocketmail.com (=?iso-8859-1?q?Regina=20Bauer?=) Date: Fri, 4 May 2001 16:11:17 +0100 (BST) Subject: [MT-List] ISO 9126 - evaluation Message-ID: <20010504151117.50152.qmail@web12505.mail.yahoo.com> Dear MT-Listers, Can anybody tell me whether there is a definition for "evaluation" in ISO 9126, and if so, could you please send it to me. I only have the German translation of this standard, and I am not sure whether the term "Beurteilung", which is defined, is the translation of "assessment" or "evaluation". Other general definitions (with source) for "evaluation", "software evaluation" or "MT evaluation" are also welcome. Regina Bauer Universität Leipzig regina_bauer@rocketmail.com ____________________________________________________________ Do You Yahoo!? Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk or your free @yahoo.ie address at http://mail.yahoo.ie From WJHutchins@compuserve.com Fri May 11 09:46:33 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Fri, 11 May 2001 04:46:33 -0400 Subject: [MT-List] EAMT workshop 2000 Message-ID: <200105110446_MC2-D09E-FCC5@compuserve.com> Papers presented at the EAMT's workshop in Ljubljana, May 2000, are now available on the EAMT website (www.eamt.org). Some are HTML files, some PPT, and some PDF. John Hutchins 11 May = From jack@kanji.org Mon May 14 14:18:26 2001 From: jack@kanji.org (Jack Halpern) Date: Mon, 14 May 2001 08:18:26 -0500 Subject: [MT-List] The CJK Dictionary Institute In-Reply-To: References: Message-ID: <200105141318.AA12952@mail.kanji.org> Dear all, On March 1, 2001 the CJK Dictionary Publishing Society Limited has been reorganized and expanded with the establishment of the The CJK Dictionary Institute, Inc. (CJKI). CJKI maintains comprehensive CJK lexical databases, and now has the world's largest Chinese and Japanese lexical databases. For details, please see: http://www.cjk.org/ We are interested in expanding and improving our databases, as well as in helping MT developers improve their products and in collaboration such as data exchanges. Right now we are especially interested in expanding our monolingual and bilingual databases for: 1. Names of famous people in Chinese 2. Chinese neologisms 3. Japanese neologisms Currently our database consists of about 1.5 million SC and TC entries, and more than 2 million Japanese entries. More than half of these are proper nouns, especially personal names, place names, and company names (see http://www.cjk.org/samples/ for details). I am looking forward to possible collaboration. Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48$B!](B473$B!](B3508 From cb@lim.nl Wed May 30 01:27:14 2001 From: cb@lim.nl (Colin Brace) Date: Tue, 29 May 2001 17:27:14 -0700 (PDT) Subject: [MT-List] MT Summit VIII: list of accepted papers now online Message-ID: <20010530002714.86716.qmail@web13801.mail.yahoo.com> Hi everyone, The list of accepted papers for the forthcoming MT Summit VIII is now online. See: http://www.eamt.org/summitVIII/programme.html#papers As you will see, it presents an intriguing mix of user studies, research papers, system presentations, and poster presentations. The list of exhibitors is also growing, and potential exhibitors who have not yet signed up should make haste, as there is limited space available: http://www.eamt.org/summitVIII/programme.html#exhibitors If you haven't visited the Summit website lately, please stop by, as it is being updated frequently: http://www.eamt.org/summitVIII/ Kind regards... ===== -- Colin Brace Amsterdam http://www.lim.nl __________________________________________________ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ From tread@lsi.uned.es Wed Jun 6 11:26:38 2001 From: tread@lsi.uned.es (Tim Read) Date: Wed, 6 Jun 2001 12:26:38 +0200 Subject: [MT-List] Machine Translation summer school in San Sebastian (north of Spain) Message-ID: <006f01c0ee73$2bd9aef0$161a80d4@kimbi> If you are a non Spanish speaker then just hit the kill button on your email client, if not, then perhaps the following will interest you: CURSO DE VERANO SOBRE TRADUCCIÓN AUTOMÁTICA (SAN SEBASTIÁN, 3-5/9/2001) Este verano, entre los días 3 (lunes) y 5 de septiembre, tendrá lugar un Curso de Verano de la Universidad del País Vasco en el Palacio Miramar de San Sebastián, de gran interés para todos aquéllos que estén interesados en la Lingüística Computacional o en la Traducción, y en particular la Traducción Automática. El título del curso es: "Traducción automática en el umbral del siglo XXI: sistemas y herramientas" y sus directoras son las Dras. Elena Bárcena (Facultad de Filología, UNED, Madrid) y Arantza Díaz de Ilarraza (Facultad de Informática, UPV, San Sebastián). El curso ha sido estratégicamente organizado para complementar otras dos grandes reuniones científicas que tendrán lugar en España también en el próximo mes de septiembre: MT Summit (la VIII Cumbre Internacional sobre Traducción Automática, que tendrá lugar en Santiago de Compostela, los días 18-22 septiembre de 2001; www.eamt.org/summitVIII/info.html), sin duda el principal foro mundial para expertos sobre el tema, y SEPLN 2001 (el congreso anual de Sociedad Española para el Procesamiento del Lenguaje Natural, que tendrá lugar en Jaén los días 12-14 de septiembre de 2001; http://sepln2001.ujaen.es), el principal foro estatal para expertos sobre el Procesamiento del Lenguaje Natural en general y la Traducción Automática en particular. Destinatarios: El curso está destinado especialmente para estudiantes, investigadores y profesores, tanto lingüistas como ingenieros, interesados en el área, con diversos niveles de conocimiento sobre la misma. También está pensado para profesionales cuyas empresas guarden relación - como productoras o meras usuarias - con la Traducción Automática, y por tanto se beneficiarían de recibir mayor información precisa y de actualidad sobre el tema. No es necesario en absoluto ser un experto en Traducción Automática para seguir el curso. Objetivos: El objetivo principal del Curso es hacer un balance sincero sobre el conjunto de la Traducción Automática: los objetivos conseguidos y los obstáculos aún por lograr. Este es sin duda un buen momento histórico para pararse a reflexionar tras más de medio siglo de existencia de esta disciplina y encontrándonos en el umbral del siglo XXI, tras haber asistido a notables avances en las dos disciplinas que vertebran la Traducción Automática: la Lingüística Teórica y la Informática. Se presentarán los principales sistemas y herramientas de TA comercializados ahora mismo en el mundo (durante el curso habrá varias demostraciones prácticas de compañías estatales y extranjeras que mostrarán sus productos), haciendo una valoración comparativa de los mismos. También se reflexionará sobre la naturaleza de los problemas que tiene la Traducción Automática para progresar y las posibilidades que se prevén para la solución o mejora de los mismos, explicándose a los alumnos los pasos que se están dando en la actualidad para conseguirlo. Naturalmente, no se puede obviar el impacto que también sobre la TA ha tenido Internet. En efecto, hoy en día Internet y la Web están alterando sustancialmente el acceso de la sociedad a la información tanto monolingüe como multilingüe. Aunque el progreso de la actividad investigadora en materia de TA sigue su camino al margen de este hecho, esta nueva demanda y oferta de información está abriendo nuevos horizontes para esta disciplina. Por ello se dedicará un parte del curso a estudiar el papel concreto que juega la TA en Internet y las prospectivas en este sentido. Profesorado: El curso cuenta con un variado elenco de expertos en TA, que investigan y publican desde hace años activamente en esta área, comenzando por una autoridad mundial como es el Professor Harold L. Somers, Director del Departmento de Ingeniería Lingüística de UMIST (Manchester, Reino Unido). También se cuenta con profesores de la UNED, la UPV y la Universidad de Deusto que trabajan en TA desde sus respectivos departamentos de lingüística o informática. Para completar este rico abanico de bagajes y perspectivas, el curso cuenta también con un ponente de la empresa privada y otro de un organismo público, ambos expertos en TA. En resumen, se trata de un curso con un claro objetivo práctico y sintetizador, en el que: - se reflexionará abierta y sinceramente sobre los resultados y perspectivas de la Traducción Automática, analizando las líneas de investigación presentes y anticipando las que se prevé prosperen durante los próximos años; - se informará sobre los sistemas y herramientas existentes en la actualidad, evaluándolos comparativamente desde una doble perspectiva teórico-práctica; - se presentarán y demostrarán algunos sistemas y herramientas punteros existentes en el mercado; - se dedicará especial atención a aquéllos que guardan relación con Internet. Idioma oficial del curso: castellano PROGRAMA A continuación se incluye el programa del curso: DIA 1 (3 de septiembre): Avances y problemas de la TA 9h. Título: Existencialismo en la Traducción Automática: ¿de dónde venimos? y ¿adónde vamos? (en inglés con traducción) Prof.: Harold Somers (Dep. Linguistic Engineering, UMIST, Reino Unido) 10:15h. Título: Reflexiones del traductor profesional sobre la TA (en euskara con traducción) Prof.: Gotzon Egia Goienetxea (Diputación Foral de Guipuzcoa) 11:30h. Título: La tragedia de la escalabilidad de la TA Prof.: Elena Bárcena (Dpto. Filologías Extranjeras, UNED) 12:45h. Título: La reutilización de recursos de ingeniería lingüística en la TA (en euskara con traducción) Prof.: Arantza Díaz de Ilarraza (Facultad de Informática, UPV) DIA 2 (4 de septiembre): Sistemas y herramientas de la TA 9h. Título: Evaluación y rentabilidad de la TA Prof.: Tim Read (Dpto. Lenguajes y Sistemas Informáticos, UNED) 10:15h. Título: La presencia del euskara en la TA: una propuesta (en euskara con traducción) Prof.: Aingeru Mayor (Facultad de Informática, UPV) Prof.: Kepa Sarasola (Facultad de Informática, UPV) 11:30h. Título: Herramientas y sistemas para la TA en el ámbito periodístico Prof.: Gorka Palazio (Facultad de CC. Sociales y de la Comunicación, UPV y EIZIE, Asociación Vasca de Traducción) 12:45h. Título: Sistemas y herramientas de TA para lenguas minoritarias (en inglés con traducción) Prof.: Harold Somers (Department of Linguistic Engineering, UMIST, Reino Unido) (Tarde: Mesa redonda y Demostraciones prácticas de sistemas y herramientas) Título de la mesa redonda: Prospectivas para herramientas y sistemas avanzados de TA Participantes: Elena Bárcena (moderadora), Kepa Sarasola, Luis Fernández, Joseba Abaitua DIA 3 (5 de septiembre): Internet, nuevo ámbito de la TA 9h. Título: El papel de la TA en la Web Prof.: Tim Read (Dpto. Lenguajes y Sistemas Informáticos, UNED) 10:15h. Título: Localización: punto de encuentro entre traducción, Internet e ingeniería lingüística (en euskara con traducción) Prof.: Luis Fernández (Code & Syntax) 11:30h. Título: Recursos en la Web para la traducción automática y semi-automática Prof.: Elena Bárcena (Dpto. Filologías Extranjeras, UNED) 12:45h. Título: Memorias de traducción en TMX compartidas a través de Internet Prof.: Joseba Abaitua (Dpto. Filología Inglesa, Universidad de Deusto) (Tarde: Demostraciones prácticas de sistemas y herramientas) Procedimiento de matriculación: Dirigirse a la dirección: www.sc.ehu.es/scrwwwsu/matricula.htm, con el código de curso 0.4. Precio de la matrícula: 11.160 ptas. From marie_adji@yahoo.fr Mon Jun 18 12:14:28 2001 From: marie_adji@yahoo.fr (=?iso-8859-1?q?marie=20adji?=) Date: Mon, 18 Jun 2001 13:14:28 +0200 (CEST) Subject: [MT-List] looking for translation tool Message-ID: <20010618111428.7946.qmail@web14704.mail.yahoo.com> Hello, Our scientific documentation centre is looking for a translation tool (French-English) to translate chemistry summarizes. Could you give us your opinion talking of that tools? Do you think that one of this tools can be used for specific scientific texts or that we be able to work with a professional translator? Thanks a lot Marie ADJI LRCCP 60 rue Auber 94408 VITRY cedex France http://www.lrccp.com mel: marie_adji@yahoo.fr ___________________________________________________________ Do You Yahoo!? -- Pour faire vos courses sur le Net, Yahoo! Shopping : http://fr.shopping.yahoo.com From ÷ÌÁÄÉÍÉÒ òÙËÏ×" Hello ! Sorry - it is off-scope of discussion question maybe. But maybe somebody knows QA (Question Answering)/FAQ discussion list address? Regards -- P bI K O B B. B. MOCKBA PhD in Computational Linguistics rykov.narod.ru Eng version - www.blkbox.com/~gigawatt/rykov.html From postediting@hotmail.com Fri Jun 22 10:10:05 2001 From: postediting@hotmail.com (Jeff Allen) Date: Fri, 22 Jun 2001 09:10:05 Subject: [MT-List] other languages Message-ID: Regarding English/Hindi translation systems, the Indian government is currently setting up various Cross-Lingual Information Retrieval and Machine-Aided Translation projects for this language pair. More information is available at: http://www.tdil.gov.in/newsletter.htm * Anuvadak : Machine Aided Translation System (English to Hindi) The only contact address I could find for this system is: (shuklavnus@yahoo.com) Also check out the following resource sites: * EMILLE project http://www.emille.lancs.ac.uk/ * http://www.hindi-language.com/ * http://www.word2word.com/ Best regards, Jeff Allen postediting@aol.com ----- >From: Denise.Morgan Subject: [MT-List] other languages Date: Mon, 30 Apr >2001 11:58:22 -0700 > >At our library, we are interested in knowing if there >are FREE machine >translation capabilities for Hindi and Vietnamese. > >We are trying to conduct programs that display our >library system website >in >languages that will help our residents. Altavista is a >great start, but >we'd love to find more FREE services. > >Thanks, >Denise Morgan >Kingstowne Library >Alexandria, Virginia _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com From clarice.de.paul@sap.com Fri Jun 22 10:36:07 2001 From: clarice.de.paul@sap.com (De Paul, Clarice) Date: Fri, 22 Jun 2001 11:36:07 +0200 Subject: [MT-List] MT EN-PT(BR) Message-ID: <9DB6EAD55F17D311BDCE0008C75DAD66082E6024@dbwdfx12.wdf.sap-ag.de> Hi there! I am doing some research/preparation for machine translation from english into brasilian portuguese and would like to ask if you have any information about the differences between these two languages that have to be considered by the programm. Thanks for any help! Best regards, Clarice De Paul From WJHutchins@compuserve.com Mon Jun 25 12:28:51 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 25 Jun 2001 07:28:51 -0400 Subject: [MT-List] MT Summit, September 2001 Message-ID: <200106250729_MC3-D705-F2E8@compuserve.com> REMINDER Hotels are getting full. Please register and book as soon as possible. List of papers now on conference website =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D MT Summit VIII, Santiago de Compostela, Spain, 18-22 September 2001 The European Association for Machine Translation (EAMT) invites you to participate in the Eighth "Machine Translation Summit" which is being held in Santiago de Compostela, = Galicia, Spain from the 18th to the 22nd September 2001. The MT Summit conferences bring together the global MT = community to review the state of the art of machine = translation, computer-aided translation support tools and multilingual tools, both in the commercial = world and in the research sphere. It is a unique opportunity to learn about the latest developments in the field, for researchers, users and vendors to exchange experiences and to get valuable feedback, and for meeting others involved in this technology. There will be a full programme of papers, presentations = by invited speakers, panel sessions, exhibitions, demonstrations, and workshops (on the day preceding the conference and on the final day). There will be = excursions, a banquet and a folk music performance, and also some activities for spouses and partners of delegates. VENUE The conference is taking place in Santiago de Compostela in North West Spain(Galicia), famous as a pilgrimage destination since the early Middle Ages, a city full of ancient buildings, from its eleventh-century cathedral to its seventeenth and eighteenth century houses, convents and palaces. The old town has been listed as a 'world heritage city' by Unesco. The venue for the conference is the Hostal de los Reyes Cat=F3licos (Hostal dos Reis Cat=F3licos), originally founded in 1499 as a Royal hospital for pilgrims, a magnificent building in the Spanish Renaissance style. It is now both the oldest and one of the most impressive hotels in the world. The opening session and the banquet will take place in the Comedor Real, other sessions will be in the hotel's auditorium and other rooms, and the exhibition and demonstrations will be housed in the Chapel. The opening reception will take place in the Pazo de Raxoi (City Council), an 18th century building next to the Hostal. HOTELS Santiago de Compostela is a major tourist centre attracting many visitors throughout the year. We have reserved accommodation in the Hostal itself and in a number of hotels close by (many restored 17th and 18th century buildings). All except one are within walking distance of the Hostal venue. (A bus will be provided for the Hotel Puerta del Camino). Because of the high demand for hotel rooms in Santiago and the substantial deposits, we ask for about half the full hotel charge when booking, and we encourage you to book early to be certain of your preference. WEBSITE INFORMATION For details of the hotels, excursions and workshops see the MT Summit website (www.eamt.org/summitVIII) or send email to: summitVIII@eamt.org. REGISTRATION Registration should preferably be made via the MT Summit website. Alternatively, you may ask for a registration form from: Carlos Paz (MT Summit Registration), Service de Traduction, European Commission, Rue de la Loi 200 - JECL 4/200, B-1049 Brussels, Belgium (Fax: +32-2-296.59.68; Telephone: +32-2-296.05.59; Email: Carlos.Paz-Carneiro@cec.eu.int). CONFERENCE PROGRAMME Tuesday, 18 September Workshops, welcome reception, opening of exhibitions Wednesday, 19 September opening session, first day of conference Thursday, 20 September Second day of conference Friday, 21 September Third day of conference, IAMT general assembly, folk music entertainment Saturday, 22 September Workshops, excursion to Coruna, special panel, banquet Sunday, 23 September Excursion to southern Galicia and northern Portugal CONFERENCE FEES The registration fees for the conference (19-22 September) are: IAMT members* Non-members Before 15 July Euro 275 Euro 325 After 15 July Euro 325 Euro 375 * members of the Asia-Pacific Association for Machine Translation (AAMT), Association for Machine Translation in the Americas (AMTA), European Association for Machine Translation (EAMT) One Euro is approximately 0.90 US dollar. Note: The conference fees do not include meals, but there are numerous good-value restaurants nearby in the city, which you will have opportunity to enjoy during the conference. CANCELLATIONS Before 15 July: 75% of the registration fee will be refunded After 15 July and before 1 September: 50% of the registration fee will be refunded After 1 September: no refunds. EXHIBITIONS Exhibition space will be provided in the Chapel of the Hostal for up to 25 companies and will be active throughout the conference. Fees for exhibitors will be Euro 1200. Contact Joerg Schuetz (joerg@iai.uni-sb.de) for details, or look at the Summit website (www.eamt.org/summitVIII). WORKSHOPS [for more details see Summit website] Tuesday 18 September fee Towards a Roadmap for MT (half day) Euro 25 Organiser: Steven Krauwer Example-based machine translation (full day) Euro 50 Organisers: Andrew Way and Michael Carl Saturday 22 September Teaching machine translation (all day) Euro 50 Organisers: Mikel Forcada and Derek Lewis MT evaluation: who did what to whom (Fourth ISLE workshop) (all day) Euro 50 Organisers: Ed Hovy, Maghi King, Sandra Manzi and Florence Reeder OPEN MEETING on terminology exchange and synchronization (SALT, OSCAR, OLIF), Saturday 22 Sept (all day) Euro 50 [for more information see http://www.ttt.org] FLIGHT DISCOUNTS. Iberia Airlines is offering participants a 25% discount (or more) on many of its flights (for details see: www.eamt.org/summitVIII). On registration you will be sent a password, in order to take advantage of this offer. ORGANISERS John Hutchins (general chair): EAMT/IAMT president Bente Maegaard (programme chair) Joerg Schuetz (exhibitions) Harold Somers (workshops) Carlos Paz (registration and local arrangements) Anthony Clarke (finance): EAMT treasurer Viggo Hansen (publicity): EAMT secretary Colin Brace (website) From ling98@videotron.ca Tue Jul 17 15:45:42 2001 From: ling98@videotron.ca (Michael Blekhman) Date: Tue, 17 Jul 2001 10:45:42 -0400 Subject: [MT-List] LogoMedia MT products Message-ID: <00a301c10ecf$277ff180$baa9c818@videotron.ca> Michael S. Blekhman, President, Lingvistica '98 Inc. Machine Translation Products by LogoMedia LogoMedia, a fast growing MT developer, based in Belmont, MA, USA, has released a new product family: a family of machine translation systems for PCs and ASP-based machine translation available through the web. The family embraces the following direct translation language pairs: English to and from French, German, Italian, Spanish, Portuguese, Russian, Chinese, Korean, and Japanese. Some of the translation engines are based on the LogoMedia proprietary English parser licensed from Language Engineering Company (LEC), while some of them are licensed from other developers, such as the English to German engine, based on the Barcelona technology, or EnglishÕRussian, licensed from Lingvistica '98 Inc. (PARS English to and from Russian). Each translation engine includes a basic dictionary and some additional terminological ("technical" as the LogoMerdia developers name them) dictionaries. For example, the EnglishÕGerman pair includes Business, Computers, Legal, and Medical dictionaries. Translation quality is rather high for a modern MT system, i.e. the translations generated are certainly not ideal but quite useful. I know, for example, that LogoMedia people are successfully using the EnglishÕGerman translation engine to communicate with their German-speaking partners in Germany. A very interesting technological feature is the so-called Composite Engine. This makes it possible for the user to have a text translated for those language pairs for which no direct translation engine exists, such as Spanish to French, or Italian to Korean. For example, if the user wants a Portuguese text to be translated into ChineseKorean, LogoMedia successively activates its 2 engines: first Portuguese to English, and then English to Korean. There is no doubt that translating directly provides higher quality, but, as they say, it's better (even much better!) than nothing. LogoMedia direct and composite engines work in several modes, each of which is very easy to use, even with minimum knowledge of MS Windows. 1. The simplest mode is translating words, and phrases, and sentences. (TransIt)This is, in fact, a dictionary look-up program with Copy/Paste options. 1.2. File translation is used for translating text and HTML files, in particular, in the Drag-and-Drop mode: the user may drag a text selected in his/her application and drop it onto the File Translation icon, after which it will be automatically translated, and the resulting text will be displayed in a Notepad window. 1.3. Text translation: a text can be translated and displayed in a typical 2-window editor. 1.4. Besides the above options, LogoMedia provides web-, text-, file, and E-mail translation on the company's site: www.logomedia.net, both in the direct and composite translation modes. Next steps include the following products and features: ž adding EnglishÕUkrainian and EnglishÕPolish direct translation engines; ž using the LogoMedia LEC parser for developing new translation engines, as well as developing similar parsers for languages other than English; ž integrating all the translation engines with MS Word. For skilled professional users, some of the LogoMedia engines include such attractive features as translation memory archives and displaying variants of translations for complete sentences and separate words. Prices vary from $99 for a basic version of a European language pair (such as EnglishÕPortuguese) and $199 for a similar Asian language pair, such as EnglishÕChinese, to $899 for the complete package including all the language pairs and all the dictionaries. From NacolM@phelps.com Tue Jul 17 16:05:44 2001 From: NacolM@phelps.com (Michael Nacol (H.O.)) Date: Tue, 17 Jul 2001 10:05:44 -0500 Subject: [MT-List] free translator list Message-ID: <150B14B2BB6FD411B7CE00508BD88FFD1E629D@PD5EX1> I recently saw your email & a note saying that you had recently compiled a translator list. Could you forward it to me. Thanks for your time. Michael Nacol From cmal45@yahoo.com Tue Jul 17 20:59:43 2001 From: cmal45@yahoo.com (beata maleci) Date: Tue, 17 Jul 2001 12:59:43 -0700 (PDT) Subject: [MT-List] suggestions Message-ID: <20010717195943.79119.qmail@web10305.mail.yahoo.com> I am working on small machine translation system, which translates magazines from English into Polish but unfortunalely, despite my good will and a lot of time devoted, I found the task a little overwhelming. All mails with personal experiences, hints and suggestions will be really appreciated. Beata Maleci __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail http://personal.mail.yahoo.com/ From postediting@hotmail.com Thu Jul 19 09:53:26 2001 From: postediting@hotmail.com (Jeff Allen) Date: Thu, 19 Jul 2001 08:53:26 Subject: [MT-List] free translator list Message-ID: >From: "Michael Nacol (H.O.)" >To: "'mt-list@eamt.org'" >Subject: [MT-List] free translator list >Date: Tue, 17 Jul 2001 10:05:44 -0500 > >I recently saw your email & a note saying that you had recently compiled a >translator list. Could you forward it to me. >Thanks for your time. >Michael Nacol Assuming that you are referring to a list of free online automatic translation service portals, Laurie Gerber compiled an initial list with input from a few people. The July 2000 version is available on the EAMT website at: http://www.eamt.org/resources/index.html A few other pages containing lists of translation systems available online are the following: http://www.sslmit.unibo.it/zanettin/cattools.htm http://www.ccl.umist.ac.uk/ra/dmowatt/mt-companies.html http://www.prospector.cz/Services/Translation/ The Rivendel Resource list page unfortunately disappeared earlier this year. Some links at these various web site pages are no longer valid. Also, more systems have been made available since the creation of these lists. Keeping an updated list is a never-ending task. Best, Jeff Allen _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp From dave@lai.com Thu Jul 19 18:36:14 2001 From: dave@lai.com (David Lakritz) Date: Thu, 19 Jul 01 10:36:14 PDT Subject: [MT-List] free translator list Message-ID: <10107191736.AA29133@ lai > Another list of on-line MT engines is at: http://www.lai.com/temain.html David Lakritz Language Automation, Inc. Begin forwarded message: From: "Jeff Allen" To: mt-list@eamt.org Cc: NacolM@phelps.com Subject: Re: [MT-List] free translator list Date: Thu, 19 Jul 2001 08:53:26 >From: "Michael Nacol (H.O.)" >To: "'mt-list@eamt.org'" >Subject: [MT-List] free translator list >Date: Tue, 17 Jul 2001 10:05:44 -0500 > >I recently saw your email & a note saying that you had recently compiled a >translator list. Could you forward it to me. >Thanks for your time. >Michael Nacol Assuming that you are referring to a list of free online automatic translation service portals, Laurie Gerber compiled an initial list with input from a few people. The July 2000 version is available on the EAMT website at: http://www.eamt.org/resources/index.html A few other pages containing lists of translation systems available online are the following: http://www.sslmit.unibo.it/zanettin/cattools.htm http://www.ccl.umist.ac.uk/ra/dmowatt/mt-companies.html http://www.prospector.cz/Services/Translation/ The Rivendel Resource list page unfortunately disappeared earlier this year. Some links at these various web site pages are no longer valid. Also, more systems have been made available since the creation of these lists. Keeping an updated list is a never-ending task. Best, Jeff Allen From lgerber@gerbersite.com Thu Jul 19 23:09:44 2001 From: lgerber@gerbersite.com (Laurie Gerber) Date: Thu, 19 Jul 2001 15:09:44 -0700 Subject: [MT-List] free translator list References: <10107191736.AA29133@lai> Message-ID: <3B575AA8.62ACB9FE@pacbell.net> Jeff is kind to mention my little collection of links to online MT systems, however, it is now pretty out of date. I largely gave up when I found these two sites who had done a much nicer and more thorough job than I could do: http://mason.gmu.edu/~aross2/mtgrid.htm http://www.foreignword.com/ Laurie Gerber > From: "Jeff Allen" > To: mt-list@eamt.org > Cc: NacolM@phelps.com > Subject: Re: [MT-List] free translator list > Date: Thu, 19 Jul 2001 08:53:26 > > >From: "Michael Nacol (H.O.)" > >To: "'mt-list@eamt.org'" > >Subject: [MT-List] free translator list > >Date: Tue, 17 Jul 2001 10:05:44 -0500 > > > >I recently saw your email & a note saying that you had recently compiled a > >translator list. Could you forward it to me. > >Thanks for your time. > >Michael Nacol > > Assuming that you are referring to a list of free online automatic > > translation service portals, Laurie Gerber compiled an initial list with > > input from a few people. The July 2000 version is available on the EAMT > > website at: http://www.eamt.org/resources/index.html > > A few other pages containing lists of translation systems available online > > are the following: > http://www.sslmit.unibo.it/zanettin/cattools.htm > http://www.ccl.umist.ac.uk/ra/dmowatt/mt-companies.html > http://www.prospector.cz/Services/Translation/ > > The Rivendel Resource list page unfortunately disappeared earlier this year. > > Some links at these various web site pages are no longer valid. Also, more > > systems have been made available since the creation of these lists. Keeping > > an updated list is a never-ending task. > > Best, > > Jeff Allen > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From stopping@rochester.rr.com Fri Jul 20 20:42:15 2001 From: stopping@rochester.rr.com (Suzanne Topping) Date: Fri, 20 Jul 2001 15:42:15 -0400 Subject: [MT-List] REQUEST FOR COMMENTS! Message-ID: <00aa01c11154$14faf280$541d5d18@rochester.rr.com> Hello all, I would be interested in hearing responses ASAP to the following quote. I have my own reaction to it, but would like to hear from translators and agencies to see if my opinion is correct. I'm trying to write a response related to this statement this afternoon, so quick replies would be very much appreciated. "Even if an organization has human translators, they could translate more if they started with a rough but automatically generated translation." (By "automatically generated translation", they are referring to Machine Translation, not pre-translating with a translation memory tool.) Thank you! --++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Suzanne Topping stopping@rochester.rr.com From esteam@otenet.gr Sat Jul 21 08:12:49 2001 From: esteam@otenet.gr (ESTeam) Date: Sat, 21 Jul 2001 10:12:49 +0300 Subject: [MT-List] REQUEST FOR COMMENTS! References: <00aa01c11154$14faf280$541d5d18@rochester.rr.com> Message-ID: <3B592B71.D3E70F6E@otenet.gr> Dear Susanne and all, A list of companies and institutions that have all proved this to be true - and these are only the ones I know personally: SAP, Germany Derwent Ltd, UK Ericsson, SE LangTech A/S, DK EU Commission Fully Automatic Solution (No Human Intervention - Raw MT results go straight to the end client) CompuMark, BE I understand the reluctance to accept this and the arguements against it are easy to find but it doesn't change the world - this is how things are and have been for years - and it is better to accept a tool for what it is than to stupidly argue that it is useless. The first agrument being that it doesn't work for my text - sure - and you find a few examples to support your thoughts and sell it to your organization. The only way to prove for your organization the contrary to the fact that this is already taking place, is to translate a large amount of data using a system - sit down two translators - one pro MT correcting the MT (since if she/he is not for it you will have no mesure at all, since translations can be changed and manipulated in all endlessness and correctness is mostly a matter of opinion) and the other translating from scratch - measure their preformance and the quality of their result (ofcourse the latter as a blind test) and base your arguements on the results. As a developer I am tired of ill founded negative arguments from translators - they should by now have realized that there is no-one threatening their jobs - so what ever you do don't base your arguments on what you "think" is correct, prove it.... we have. Good luck Gudrun Magnusdottir Managing Director ESTeam AB, an Automatic Translation Solution Company Suzanne Topping wrote: > Hello all, > > I would be interested in hearing responses ASAP to the following quote. I > have my own reaction to it, but would like to hear from translators and > agencies to see if my opinion is correct. I'm trying to write a response > related to this statement this afternoon, so quick replies would be very > much appreciated. > > "Even if an organization has human translators, they could translate more if > they started with a rough but automatically generated translation." > > (By "automatically generated translation", they are referring to Machine > Translation, not pre-translating with a translation memory tool.) > > Thank you! > > --++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > Suzanne Topping > stopping@rochester.rr.com > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From WJHutchins@compuserve.com Tue Aug 7 16:21:06 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Tue, 7 Aug 2001 11:21:06 -0400 Subject: [MT-List] TIME Magazine Article Message-ID: <200108071121_MC3-DBA2-452@compuserve.com> This is a MIME-encapsulated message --883d8d06-8b45-11d5-9a14-0002a5349152 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Recently received. I don't know this article. Does anyone know when it appeared (date, issue, page), and whether it is available online? John Hutchins = -------------Forwarded Message----------------- From: INTERNET:RogerDurban@aol.com, INTERNET:RogerDurban@aol.com To: [unknown], INTERNET:info@eamt.org = Date: 04/08/101 00:49 PM RE: TIME Magazine Article = Sirs: In the past few weeks, Tme Magazine published an informative piece on mac= hine = translation. I hae looked diligently for an online source for this artic= le. = Could your organization provide any assistance in helping located a copy = of = this article? Thank you in advance for any help you can provide. Roger A. Durban --883d8d06-8b45-11d5-9a14-0002a5349152 Content-Type: application/octet-stream; name="C:\CSERV-2\DOWNLOAD\UNTITLED.BIN" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="C:\CSERV-2\DOWNLOAD\UNTITLED.BIN" PGh0bWw+IDxoZWFkPiA8dGl0bGU+QW1hem9uLmNvLnVrOiBBbWF6aW5nIFNhdmluZ3M8L3RpdGxl PiA8L2hlYWQ+DQo8Ym9keSBiZ2NvbG9yPSIjRkZGRkZGIiBsaW5rPSIjMDAzMzk5IiBhbGluaz0i I0ZGOTkzMyIgdmxpbms9IiM5OTY2MzMiIHRleHQ9IiMwMDAwMDAiPg0KDQo8dGFibGUgYm9yZGVy PTAgY2VsbHBhZGRpbmc9MCBjZWxsc3BhY2luZz0wIHdpZHRoPTYwMCB2c3BhY2U9MD4NCjx0cj4N Cjx0ZCB3aWR0aD0xMDAlIGJnY29sb3I9I2ZmZmZmZj48SU1HIHNyYz1odHRwOi8vaW1hZ2VzLWV1 LmFtYXpvbi5jb20vaW1hZ2VzL0cvMDIvdWstd2VsY29tZS9zaG9wcy9iYW5uZXItdG9wLWFtYXpp bmctNDE1eDQ1IGJvcmRlcj0wIGFsdD0iQW1hemluZyBTYXZpbmdzIj48L2E+PC90ZD4NCjwvdHI+ DQo8L3RhYmxlPg0KPHRhYmxlIGJvcmRlcj0wIGNlbGxwYWRkaQ== --883d8d06-8b45-11d5-9a14-0002a5349152-- From mirko.plitt@eur.autodesk.com Tue Aug 7 16:30:06 2001 From: mirko.plitt@eur.autodesk.com (mirko.plitt@eur.autodesk.com) Date: Tue, 7 Aug 2001 17:30:06 +0200 Subject: [MT-List] TIME Magazine Article Message-ID: This article appeared in the European edition of Time dated July 16, 2001. An online version is available at http://www.time.com/time/europe/digital/magazine/0,9868,166805,00.html. Mirko Plitt -----Original Message----- From: John Hutchins [mailto:WJHutchins@compuserve.com] Sent: Tuesday, August 07, 2001 5:21 PM To: MT-list Subject: [MT-List] TIME Magazine Article Recently received. I don't know this article. Does anyone know when it appeared (date, issue, page), and whether it is available online? John Hutchins -------------Forwarded Message----------------- From: INTERNET:RogerDurban@aol.com, INTERNET:RogerDurban@aol.com To: [unknown], INTERNET:info@eamt.org Date: 04/08/101 00:49 PM RE: TIME Magazine Article Sirs: In the past few weeks, Tme Magazine published an informative piece on machine translation. I hae looked diligently for an online source for this article. Could your organization provide any assistance in helping located a copy of this article? Thank you in advance for any help you can provide. Roger A. Durban From stopping@rochester.rr.com Tue Aug 7 16:49:47 2001 From: stopping@rochester.rr.com (Suzanne Topping) Date: Tue, 7 Aug 2001 11:49:47 -0400 Subject: [MT-List] TIME Magazine Article References: <200108071121_MC3-DBA2-452@compuserve.com> Message-ID: <00d101c11f58$9748d380$541d5d18@rochester.rr.com> Roger might also be interested in checking out the article titled "Machine Translation: Is It Good Enough?" from the June 2001 issue of E-business Advisor magazine. Unfortunately, only a summary is available online: http://www.advisor.com/Articles.nsf/aid/SULLD27 ----- Original Message ----- From: "John Hutchins" To: "MT-list" Sent: Tuesday, August 07, 2001 11:21 AM Subject: [MT-List] TIME Magazine Article Recently received. I don't know this article. Does anyone know when it appeared (date, issue, page), and whether it is available online? John Hutchins -------------Forwarded Message----------------- From: INTERNET:RogerDurban@aol.com, INTERNET:RogerDurban@aol.com To: [unknown], INTERNET:info@eamt.org Date: 04/08/101 00:49 PM RE: TIME Magazine Article Sirs: In the past few weeks, Tme Magazine published an informative piece on machine translation. I hae looked diligently for an online source for this article. Could your organization provide any assistance in helping located a copy of this article? Thank you in advance for any help you can provide. Roger A. Durban From kgodden@justtalk.com Wed Aug 8 16:33:19 2001 From: kgodden@justtalk.com (Kurt Godden) Date: Wed, 8 Aug 2001 11:33:19 -0400 Subject: [MT-List] Breakthrough in MT Message-ID: "Dog translator barks up the right tree" at http://www.cnn.com/2001/WORLD/asiapcf/east/08/08/japan.translator/index.html Enjoy! Kurt Godden, Director of Linguistics JustTalk 217 Third Street Ann Arbor, MI 48103 www.justtalk.com 734-623-7954, ext. 216 ----------------------------------------------------------- He gives us a hero every once in a while, so that we can know how we'd behave if we had any character. --David E. Davis on God's job performance From macklovi@IRO.UMontreal.CA Wed Aug 8 19:10:07 2001 From: macklovi@IRO.UMontreal.CA (Elliott Macklovitch) Date: Wed, 08 Aug 2001 14:10:07 -0400 Subject: [MT-List] TSrali.com Message-ID: <3B71807F.C14EEE52@IRO.UMontreal.CA> --------------79740619A965C9B1BFF21B0C Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable The RALI Laboratory at the University of Montreal is pleased to announce the launch of TSrali.com, an enhanced version of its popular, Web-based, bilingual concordancer. TSrali.com is a subscription service, affordable by individuals and offering substantial group discounts. A free 5-day trial of the system is also available. For this and other information consult http://www.TSrali.com. Le laboratoire RALI, de l=92Universit=E9 de Montr=E9al, est fier d=92annoncer le lancement de TSrali.com, version am=E9lior=E9e de son tr=E8= s appr=E9ci=E9 concordancier bilingue sur le Web. TSrali.com offre des abonnements =E0 un co=FBt abordable pour les particuliers, des rabais substantiels =E9tant accord=E9s aux groupes d=92utilisateurs. Une p=E9rio= de d=92essai gratuite de 5 jours est =E9galement propos=E9e. Pour plus d=92information, veuillez consulter le site http://www.TSrali.com. ***************************************** Elliott Macklovitch Laboratoire RALI, Dept. d'informatique et de recherche operationnelle Universite de Montreal, C.P. 6128, succursale Centre-ville Montreal, Quebec, Canada H3C 3J7 tel: (514) 343-7535 fax: (514) 343-2496 http://www-rali.iro.umontreal.ca --------------79740619A965C9B1BFF21B0C Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable  
    The RALI Laboratory at the University of Montreal is pleased to announce the launch of TSrali.com, an enhanced versi= on of its popular, Web-based, bilingual concordancer. TSrali.com is a subscription service, affordable by individuals and offering substantia= l group discounts. A free 5-day trial of the system is also available. For this and other information consult  http://www.TSrali.com.

    Le laboratoire RALI, de l=92Université de Mo= ntréal, est fier d=92annoncer le lancement de TSrali.com, version am&eacut= e;liorée de son très apprécié concordancier bilingue sur le Web. TSrali.com offre des abonnements à un coût abord= able pour les particuliers, des rabais substantiels étant accordé= ;s aux groupes d=92utilisateurs. Une période d=92essai gratuite de 5 = jours est également proposée. Pour plus d=92information, veuillez= consulter le site http://www.TSrali.com= .

 
*****************************************
 
Elliott Macklovitch
     Laboratoire RALI, Dept. d'informatique et de recherche operationnelle
     Universite de Montreal, C.P. 6128, succursal= e Centre-ville
     Montreal, Quebec, Canada  H3C 3J7
     tel: (514) 343-7535    &= nbsp;   fax: (514) 343-2496
     http://www-rali.iro.umontreal.ca
  --------------79740619A965C9B1BFF21B0C-- From cb@lim.nl Tue Aug 21 11:30:57 2001 From: cb@lim.nl (Colin Brace) Date: Tue, 21 Aug 2001 12:30:57 +0200 Subject: [MT-List] MT Summit VIII: full programme now online Message-ID: <20010821102804.JXKR18584.amsfep13-int.chello.nl@[62.108.30.75]> Hi everyone, The full programme for the forthcoming MT Summit VIII, which starts a month from today, is now online. See: http://www.eamt.org/summitVIII/programme.html As you will see, it offers an exciting mix of user studies, research papers, discussion panels, and keynote speeches. In addition, there are exhibitions by developers, a poster session, workshops, and several excursions. And, of course, plenty of coffee breaks! We also have online a list of the registrants so far: http://www.eamt.org/summitVIII/delegates.html Already we have nearly three hundred people from around the world planning to join us in the beautiful city of Santiago de Compostela, Galicia, Spain. At least seventeen companies and organisations will be demonstrating ***working*** translation systems, virtually all the major players in the field. See: http://www.eamt.org/summitVIII/exhibitors.html If you haven't yet registered for the MT Summit VIII and are still considering doing so, you would be advised not to delay any further. Nearly all of the hotel rooms reserved for us are full -- only a four-star and a no-star have vacancies still -- and Santiago is a popular destination year-around, so your choice of flights and accommodation may be severely limited. -- Colin Brace Amsterdam http://www.lim.nl From Andy Way CA Mon Sep 3 18:44:27 2001 From: Andy Way CA (Andy Way CA) Date: Mon, 3 Sep 2001 18:44:27 +0100 (IST) Subject: [MT-List] EBMT Workshop Message-ID: <200109031739.SAA14468@mailhost.compapp.dcu.ie> Workshop on Example-Based Machine Translation Hosted by MT-SUMMIT VIII Santiago de Compostela, Spain, September 18, 2001 http://www.compapp.dcu.ie/~away/EBMT.html Co-chairs: Michael Carl, IAI, Saarbr=FCcken;=20 Andy Way, Computer Applications, Dublin City University Programme 10:00 - 10:15 Opening Remarks 10:15 - 10:45 Davide Turcato and Fred Popowich, What is Example-Based Machine Translation? 10:45 - 11:45 Harold Somers, EBMT seen as Case-based Reasoning 11:45 - 12:15 Coffee Break 12:15 - 12:45 Andy Way, Translating with Examples 12:45 - 13:15 Reinhard Sch=E4ler, Beyond Translation Memories 13:15 -- 14:45 Lunch 14:45 - 15:45 Kevin McTait, Linguistic Knowledge and Complexity in an EBMT System Based on Translation Patterns 15:45 - 16:15 Ralf D. Brown, Transfer-Rule Induction for Example-Based Translation 16:15 - 16:45 Coffee Break 16:45 - 17:15 Michael Carl, Inducing Translation Grammars from Bracketed Alignments 17:15 - 17:45 Arul Menezes and Stephen D. Richardson, A best-first alignmen= t algorithm for automatic extraction of transfer mappings from bilingual corpora 17:45 - 18:00 Closing Remarks From WJHutchins@compuserve.com Fri Sep 14 11:14:46 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Fri, 14 Sep 2001 06:14:46 -0400 Subject: [MT-List] MT Summit VIII will be held Message-ID: <200109140615_MC3-DFBF-9A7A@compuserve.com> MT Summit VIII, Santiago de Compostela, 18-22 September 2001 The terrible events on Tuesday are having an obvious impact on the MT Summit conference. Everybody has been profoundly shocked all over the world. We have heard already that a few Americans may well not be able to= come and that others are unsure. However, we think that we should not cancel the whole conference at this late stage. In a way, cancellation would mean submission to terroris= m. We anticipate that some changes will have to be made to the programme of speakers, but we shall run the conference as successfully as we can. We look forward to seeing you all in Santiago. = John Hutchins (conference chair) 14 Sept. From bonnie@umiacs.umd.edu Fri Sep 14 12:49:50 2001 From: bonnie@umiacs.umd.edu (Bonnie J. Dorr) Date: Fri, 14 Sep 2001 07:49:50 -0400 (EDT) Subject: [MT-List] MT Summit VIII will be held Message-ID: <200109141149.HAA11408@carissa.umiacs.umd.edu> I agree. Do NOT cancel the conference. Our thoughts are all with you, even if we can't come. Thanks for all the letters. :-) Bonnie Dorr From shelmrei@crl.nmsu.edu Fri Sep 14 15:35:07 2001 From: shelmrei@crl.nmsu.edu (Stephen Helmreich) Date: Fri, 14 Sep 2001 08:35:07 -0600 Subject: [MT-List] MT Summit VIII will be held References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <3BA2159B.E762D414@crl.nmsu.edu> I echo Bonnie's sentiments. Don't cancel the conference. Steve Helmreich "Bonnie J. Dorr" wrote: > > I agree. Do NOT cancel the conference. Our thoughts are all with you, > even if we can't come. Thanks for all the letters. :-) > > Bonnie Dorr > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 16:47:43 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 10:47:43 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109141149.HAA11408@carissa.umiacs.umd.edu> References: <200109141149.HAA11408@carissa.umiacs.umd.edu> Message-ID: <200109151547.AA14513@mail.kanji.org> Greetings In message "Re: [MT-List] MT Summit VIII will be held", Bonnie J. Dorr wrote... >I agree. Do NOT cancel the conference. Our thoughts are all with you, >even if we can't come. Thanks for all the letters. :-) I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, I and a quite a number of others will definitely be attending. Don't let the terrorists win! >Bonnie Dorr > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From jack@kanji.org Sat Sep 15 19:34:14 2001 From: jack@kanji.org (Jack Halpern) Date: Sat, 15 Sep 2001 13:34:14 -0500 Subject: [MT-List] MT Summit VIII will be held In-Reply-To: <200109151547.AA14513@mail.kanji.org> References: <200109151547.AA14513@mail.kanji.org> Message-ID: <200109151834.AA14514@mail.kanji.org> VET STRANGE -- I sent this message out this morning and I keep getting it. Maybe ten time in the last hour. Is something wrong with your mail server? Jack Halpern wrote... >Greetings > >In message "Re: [MT-List] MT Summit VIII will be held", >Bonnie J. Dorr wrote... > >I agree. Do NOT cancel the conference. Our thoughts are all with you, > >even if we can't come. Thanks for all the letters. :-) > >I agree. Do NOT cancel. Despite of some difficulties with flights from Japan, >I and a quite a number of others will definitely be attending. Don't let the >terrorists win! > > > >Bonnie Dorr > > > >-- > > For MT-List info, see http://www.eamt.org/mt-list.html > > > > >Regards, Jack Halpern > President, The CJK Dictionary Institute, Inc. > http://www.cjk.org Phone: +81-48-473-3508 > > >-- > For MT-List info, see http://www.eamt.org/mt-list.html > Regards, Jack Halpern President, The CJK Dictionary Institute, Inc. http://www.cjk.org Phone: +81-48-473-3508 From teruko+@cs.cmu.edu Fri Sep 14 20:52:19 2001 From: teruko+@cs.cmu.edu (Teruko Mitamura) Date: Fri, 14 Sep 2001 15:52:19 -0400 Subject: [MT-List] TMI 2002 - Call for Papers Message-ID: <25324.1000497139@kyoto.lti.cs.cmu.edu> TMI 2002 - Second Call for Papers The 9th Conference on Theoretical and Methodological Issues in Machine Translation March 13 - 17, 2002 Keihanna, Japan The ninth meeting of the TMI conference will be held March 13-17, 2002 near the historic cities of Nara and Kyoto in Japan. The workshops and tutorials will be held jointly with the Natural Language Processing Society, Japan. Important Dates: ---------------- Paper Submissions: October 15, 2001 (Monday) Acceptance notification: December 10, 2001 (Monday) Camera-ready copies due: January 25, 2002 (Friday) Submission Guidelines: ---------------------- Authors are invited to submit substantial, original, and unpublished research on any issues relevant to machine translation. Papers should be in English, not longer than 10 pages (around 5,000 words), including references. Topics of interest include, but are not limited to: MT for the Web Practical MT (multilingual eCommerce, localization, etc.) Methodologies for MT (statistical, example-based, KBMT, ...) Speech and dialogue translation NLP techniques for MT Controlled language for MT Knowledge acquisition for MT systems MT evaluation techniques and evaluation results MT for cross-lingual retrieval and question answering Format & Style Files: --------------------- Your paper should be prepared according to the following guidelines (for authors using LaTeX, there is a style file tmi02.cls available, which comes with a pair of style files for formatting examples, gb4e.sty and cgloss4e.sty. See http://sevilla.mt.cs.cmu.edu/TMI2002/cfp.html): -The font size should be no smaller than 11pt, and the paper size should be A4. -TMI uses an anonymous review process. Therefore, all papers should be submitted with a separate author ID page (in a separate file) that includes only the title of the paper, the topic area, and the author name(s) and address(es). The paper itself should begin with the title and an abstract, but should not include the names or addresses of the authors. -Papers should be submitted as .pdf files only. All papers will be submitted electronically; authors must first register before uploading papers to the program committee database (details coming soon on the Author Resource page). For bibliographic references, if the author's name(s) is/are part of the text, then only the date should be in brackets. E.g. "Huddleston (1988) introduced the term ...", not "(Huddleston 1988) introduced the term ..." -Make sure your figures are not wider than the text. Don't forget to use italics for cited words, and double quotes for glosses. If you cite non-Roman script please cite as follows: NON-EUROPEAN transliteration "gloss" (the transliteration should be in italics). Program Committee: ------------------ Teruko Mitamura & Eric Nyberg (co-chairs) Carnegie Mellon Timothy Baldwin CSLI Christian Boitet Universit,Ai(B Joseph Fourier Andrew Bredenkamp University of Essex Lynn Carlson U.S. Department of Defense Satoru Ikehara Tottori University Hitoshi Isahara CRL Japan Kevin Knight USC-ISI Satoshi Sato Kyoto University Harold Somers UMIST Koichi Takeda TRL-IBM Hideki Tanaka ATR TMI 2002 Officers: ------------------ Program Committee Chairs: Teruko Mitamura and Eric Nyberg, Carnegie Mellon University, USA Publicity and Local Arrangements: Francis Bond and Hiromi Nakaiwa, NTT Communication Science Laboratories, Kyoto, Japan General Chair: Sergei Nirenburg, Computing Research Lab, NMSU, USA Locations and Times: -------------------- TMI-2002 Papers and Panels (March 13-15 (Wed-Fri), 2002) NTT Communication Science Laboratories, NTT Keihanna building 2-4, Hikaridai, Seika-cho, Soraku-gun, Kyoto, Japan, 619-0237 Workshops/Tutorials (March 16-17 (Sat-Sun), 2002) Keihanna Plaza, 1-7, Hikaridai, Seika-cho, Soraku-gun, Kyoto, Japan, 619-0237 TMI 2002 Home Page: http://www.kecl.ntt.co.jp/events/tmi/ TMI 2002 CFP: http://sevilla.mt.cs.cmu.edu/TMI2002/cfp.html Questions for CFP? Please contact teruko@cs.cmu.edu or ehn@cs.cmu.edu From fcochard@worldbank.org Mon Sep 17 06:02:13 2001 From: fcochard@worldbank.org (fcochard@worldbank.org) Date: Mon, 17 Sep 2001 01:02:13 -0400 Subject: [MT-List] Frederic M. Cochard/Person/World Bank is out of the office. Message-ID: I will be out of the office starting 09/03/2001 and will not return until 09/26/2001. Thanks for your message. I'll respond upon my return. From nikolay@npp.cit.bg Mon Sep 17 09:19:47 2001 From: nikolay@npp.cit.bg (Nikolay Ivanov) Date: Mon, 17 Sep 2001 10:19:47 +0200 Subject: [MT-List] STAND PROUD, AMERICA! References: <10107191736.AA29133@ lai > Message-ID: <01fa01c13f51$865fa300$340310ac@zska> > TRIBUTE TO THE UNITED STATES > > This is from a Canadian newspaper. > > America: The Good Neighbor. > > Widespread but only partial news coverage was given recently to a > remarkable > editorial broadcast from Toronto by Gordon Sinclair, a Canadian television > commentator. What follows is the full text of his > trenchant remarks as printed in the Congressional Record: > > "This Canadian thinks it is time to speak up for the Americans as the most > generous and possibly the least appreciated people on all the earth. > > Germany, Japan and, to a lesser extent, Britain and Italy were lifted out > of > the debris of war by the Americans who poured in billions of dollars and > forgave other billions in debts. None of these countries is today paying > even the interest on its remaining debts to the United States. > > > When France was in danger of collapsing in 1956, it was the Americans who > propped it up, and their reward was to be insulted and swindled on the > streets of Paris. I was there. I saw it. > > When earthquakes hit distant cities, it is the United States that hurries > in > to help. This spring, 59 American communities were flattened by tornadoes. > Nobody helped. > > The Marshall Plan and the Truman Policy pumped billions of dollars into > discouraged countries. Now newspapers in those countries are writing about > the decadent, warmongering Americans. > > I'd like to see just one of those countries that is gloating over the > erosion of the United States dollar build its own airplane. Does any other > country in the world have a plane to equal the Boeing Jumbo Jet, the > Lockheed > > Tri-Star, or the Douglas DC10? If so, why don't they fly them? Why do all > the International lines except Russia fly American Planes? > > Why does no other land on earth even consider putting a man or woman on the > moon? You talk about Japanese technocracy, and you get radios. You talk > about > German technocracy, and you get automobiles. You talk about > American technocracy, and you find men on the moon - not once, but several > times and safely home again. > > You talk about scandals, and the Americans put theirs right in the store > window for everybody to look at. Even their draft-dodgers are not pursued > and > hounded. They are here on our streets, and most of them, unless they > are breaking Canadian laws, are getting American dollars from mom and dad > at > home to spend here. > > When the railways of France, Germany and India were breaking down through > age, it was the Americans who rebuilt them. When the Pennsylvania Railroad > and the New York Central went broke, nobody loaned them an old > caboose. Both are still broke. > > I can name 5000 times when the Americans raced to the help of other people > in trouble. Can you name me even one time when someone else raced to the > Americans in trouble? I don't think there was outside help even > during the San Francisco earthquake. > > Our neighbors have faced it alone, and I'm one Canadian who is damned tired > of hearing them get kicked around. They will come out of this thing with > their flag high. And when they do, they are entitled to thumb their nose > at the lands that are gloating over their present troubles. I hope Canada > is > not one of those." > > Stand proud, America! > +++++++++++++++++++++++++++++++++++++++++++++++++ > > This is one of the best editorials that I have ever read regarding the > United States. It is nice that one man realizes it. I only wish that the > rest > of the world would realize it. We are always blamed for > everything, and never even get a thank you for the things we do. > > Maybe each of you can send this to at least one person and they might send > it to one of their friends until this letter is sent to every person on the > web. I am just a single American that has read this, > > I SURE HOPE THAT A LOT MORE READ IT SOON. From ling98@videotron.ca Sun Sep 16 16:45:22 2001 From: ling98@videotron.ca (Michael Blekhman) Date: Sun, 16 Sep 2001 11:45:22 -0400 Subject: [MT-List] News from Lingvistica Message-ID: <007701c13ec6$989a99c0$baa9c818@videotron.ca> Please see new information on Lingvistica's potential and developments at: www.ling98.com/Lingv.zip Sincerely, Dr. Michael S. Blekhman, Edward Kool, Presidents, Lingvistica b.v. Canada - Holland. From glenn@lec.com Tue Sep 18 09:00:48 2001 From: glenn@lec.com (Glenn A. Akers) Date: Tue, 18 Sep 2001 04:00:48 -0400 Subject: [MT-List] MT Summit VIII will be held Message-ID: <091A6A349AAD8844A473651BFBEEC25208CD33@lec2k.Lec.local> John, I have looked forward very much to attending the MT Summit and to introducing our new company to the community. As much as I would like to attend, my young son's wish for me to be with him, as well as my sorrow over the evil acts of the past week lead me to remain in Boston. This does not reflect our submission to terrorism, but to our need for reflection on an appropriate response to this terrible event. Glenn Akers CEO Language Engineering Company, LLC LogoMedia Corporation -----Original Message----- From: John Hutchins [mailto:WJHutchins@compuserve.com]=20 Sent: Friday, September 14, 2001 6:15 AM To: MT-list Subject: [MT-List] MT Summit VIII will be held MT Summit VIII, Santiago de Compostela, 18-22 September 2001 The terrible events on Tuesday are having an obvious impact on the MT Summit conference. Everybody has been profoundly shocked all over the world. We have heard already that a few Americans may well not be able to come and that others are unsure. However, we think that we should not canl the whole conference at this late stage. In a way, cancellation would mean submission to terrorism. We anticipate that some changes will have to be made to the programme of speakers, but we shall run the conference as successfully as we can. We look forward to seeing you all in Santiago.=20 John Hutchins (conference chair) 14 Sept. --=20 For MT-List info, see http://www.eamt.org/mt-list.html From rezaqaffori@hotmail.com Sat Sep 29 08:03:39 2001 From: rezaqaffori@hotmail.com (reza qaffori) Date: Sat, 29 Sep 2001 16:03:39 +0900 Subject: [MT-List] Persian/Farsi-English Text Translator Message-ID: Hello, I`m Dr. Qaffori , from Tokyo medical university/Japan, thank you for your all correspondings so far. May you please let me know, is there any text Persian/Farsi-Japanese and also Persian/Farsi-English text free translators on line/Over internet? Best regards, Qaffori _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp From WJHutchins@compuserve.com Mon Oct 1 11:30:30 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Mon, 1 Oct 2001 06:30:30 -0400 Subject: [MT-List] Persian/Farsi-English Text Translator Message-ID: <200110010630_MC3-E1A8-4A8A@compuserve.com> Dear Qaffori, You ask about Farsi systems. As you will see from the "Compendium of translation software" on the EAMT website (www.eamt.org) there are to my knowledge only electronic dictionaries available: Language Teacher (Ectaco) Phrasebook (Lernout & Hauspie) Universal Translator (LanguageForce) WebTrans (Lernout & Hauspie) None are free; none are on-line. Universal Translator "claims" to be more than a dictionary. The L&H products are probably now unavailable. Only Phrasebook and Universal Translator cover Farsi-Japanese in theory. = For fuller information see the Compendium. If any more systems are known to anyone I shall be pleased to hear of the= m, With regards, John Hutchins 1 Oct. Best regards, Qaffori < From malek.boualem@rd.francetelecom.com Mon Oct 1 12:07:59 2001 From: malek.boualem@rd.francetelecom.com (BOUALEM Malek FTRD/DMI/LAN) Date: Mon, 1 Oct 2001 13:07:59 +0200 Subject: [MT-List] Persian/Farsi-English Text Translator Message-ID: The Computing Research Lab (CRL) at New Mexico State University has achieved some work on Persian-English machine translation. See : http://crl.nmsu.edu/Research/Projects/shiraz/index.html Contact : R=E9mi Zajac (rzajac@crl.nmsu.edu) Malek Boualem=20 > -----Original Message----- > From: John Hutchins [mailto:WJHutchins@compuserve.com] > Sent: Monday, October 01, 2001 12:31 PM > To: reza qaffori; [unknown] > Subject: Re: [MT-List] Persian/Farsi-English Text Translator >=20 >=20 > Dear Qaffori, >=20 > You ask about Farsi systems. As you will see from the > "Compendium of translation software" on the EAMT website=20 > (www.eamt.org) > there are to my knowledge only electronic dictionaries available: > Language Teacher (Ectaco) > Phrasebook (Lernout & Hauspie) > Universal Translator (LanguageForce) > WebTrans (Lernout & Hauspie) > None are free; none are on-line. > Universal Translator "claims" to be more than a dictionary. > The L&H products are probably now unavailable. > Only Phrasebook and Universal Translator cover Farsi-Japanese > in theory.=20 >=20 > For fuller information see the Compendium. >=20 > If any more systems are known to anyone I shall be pleased to=20 > hear of them, > With regards, > John Hutchins > 1 Oct. > Best regards, > Qaffori > < >=20 >=20 > --=20 > For MT-List info, see http://www.eamt.org/mt-list.html >=20 From Jose.Coch@lingway.com Mon Oct 1 14:35:20 2001 From: Jose.Coch@lingway.com (Coch, Jose) Date: Mon, 1 Oct 2001 15:35:20 +0200 Subject: [MT-List] Persian/Farsi-English Text Translator Message-ID: <41F5DAEE117CD411BD210050BA0A7DF38A991F@BADIANE> Dear all, There is a free online Farsi-English Dictionary at http://www.ectaco.com/online/diction.php3?lang=3D20 Best, Jos=E9 _________________________________________________ Jos=E9 Coch, Lingway Le M=E9li=E8s / 261, rue de Paris F-93556 Montreuil Cedex / FRANCE Tel: (33)-1-49 93 39 38 / Fax: (33)-1-49 93 39 39 E-mail: jose.coch@lexiquest.fr / jose.coch@lingway.com _________________________________________________ -----Message d'origine----- De : John Hutchins [mailto:WJHutchins@compuserve.com] Envoy=E9 : lundi 1 octobre 2001 12:31 =C0 : reza qaffori; [unknown] Objet : Re: [MT-List] Persian/Farsi-English Text Translator Dear Qaffori, You ask about Farsi systems. As you will see from the "Compendium of translation software" on the EAMT website (www.eamt.org) there are to my knowledge only electronic dictionaries available: Language Teacher (Ectaco) Phrasebook (Lernout & Hauspie) Universal Translator (LanguageForce) WebTrans (Lernout & Hauspie) None are free; none are on-line. Universal Translator "claims" to be more than a dictionary. The L&H products are probably now unavailable. Only Phrasebook and Universal Translator cover Farsi-Japanese in theory.=20 For fuller information see the Compendium. If any more systems are known to anyone I shall be pleased to hear of = them, With regards, John Hutchins 1 Oct. Best regards, Qaffori < --=20 For MT-List info, see http://www.eamt.org/mt-list.html From teruko+@cs.cmu.edu Fri Oct 5 20:19:42 2001 From: teruko+@cs.cmu.edu (teruko+@cs.cmu.edu) Date: Fri, 05 Oct 2001 15:19:42 -0400 Subject: [MT-List] TMI 2002 - CFP and Call for Workshops Message-ID: <2697.1002309582@kyoto.lti.cs.cmu.edu> ***We now have a new date for Paper Submissions. See below.*** TMI 2002 - Call for Papers and Call for Workshops The 9th Conference on Theoretical and Methodological Issues in Machine Translation March 13 - 17, 2002 Keihanna, Japan The ninth meeting of the TMI conference will be held March 13-17, 2002 near the historic cities of Nara and Kyoto in Japan. The workshops and tutorials will be held jointly with the Natural Language Processing Society, Japan. Important Dates: ---------------- Paper Submissions: October 29, 2001 (Monday) **New Date** Acceptance notification: December 17, 2001 (Monday) **New Date** Camera-ready copies due: January 25, 2002 (Friday) Workshops: ---------- If you are interested in organizing a workshop, send your 1-2 page proposal by October 26 to: Teruko Mitamura and Eric Nyberg . Submission Guidelines: ---------------------- Authors are invited to submit substantial, original, and unpublished research on any issues relevant to machine translation. Papers should be in English, not longer than 10 pages (around 5,000 words), including references. Topics of interest include, but are not limited to: MT for the Web Practical MT (multilingual eCommerce, localization, etc.) Methodologies for MT (statistical, example-based, KBMT, ...) Speech and dialogue translation NLP techniques for MT Controlled language for MT Knowledge acquisition for MT systems MT evaluation techniques and evaluation results MT for cross-lingual retrieval and question answering Format & Style Files: --------------------- Your paper should be prepared according to the following guidelines (for authors using LaTeX, there is a style file tmi02.cls available, which comes with a pair of style files for formatting examples, gb4e.sty and cgloss4e.sty. See http://sevilla.mt.cs.cmu.edu/TMI2002/cfp.html): -The font size should be no smaller than 11pt, and the paper size should be A4. -TMI uses an anonymous review process. Therefore, all papers should be submitted with a separate author ID page (in a separate file) that includes only the title of the paper, the topic area, and the author name(s) and address(es). The paper itself should begin with the title and an abstract, but should not include the names or addresses of the authors. -Papers should be submitted as .pdf files only. All papers will be submitted electronically; authors must first register before uploading papers to the program committee database (details coming soon on the Author Resource page). For bibliographic references, if the author's name(s) is/are part of the text, then only the date should be in brackets. E.g. "Huddleston (1988) introduced the term ...", not "(Huddleston 1988) introduced the term ..." -Make sure your figures are not wider than the text. Don't forget to use italics for cited words, and double quotes for glosses. If you cite non-Roman script please cite as follows: NON-EUROPEAN transliteration "gloss" (the transliteration should be in italics). Program Committee: ------------------ Teruko Mitamura & Eric Nyberg (co-chairs) Carnegie Mellon Timothy Baldwin CSLI Christian Boitet Universit,Ai(B Joseph Fourier Andrew Bredenkamp University of Essex Lynn Carlson U.S. Department of Defense Satoru Ikehara Tottori University Hitoshi Isahara CRL Japan Kevin Knight USC-ISI Satoshi Sato Kyoto University Harold Somers UMIST Koichi Takeda TRL-IBM Hideki Tanaka ATR TMI 2002 Officers: ------------------ Program Committee Chairs: Teruko Mitamura and Eric Nyberg, Carnegie Mellon University, USA Publicity and Local Arrangements: Francis Bond and Hiromi Nakaiwa, NTT Communication Science Laboratories, Kyoto, Japan General Chair: Sergei Nirenburg, Computing Research Lab, NMSU, USA Locations and Times: -------------------- TMI-2002 Papers and Panels (March 13-15 (Wed-Fri), 2002) NTT Communication Science Laboratories, NTT Keihanna building 2-4, Hikaridai, Seika-cho, Soraku-gun, Kyoto, Japan, 619-0237 Workshops/Tutorials (March 16-17 (Sat-Sun), 2002) Keihanna Plaza, 1-7, Hikaridai, Seika-cho, Soraku-gun, Kyoto, Japan, 619-0237 TMI 2002 Home Page: http://www.kecl.ntt.co.jp/events/tmi/ TMI 2002 CFP: http://sevilla.mt.cs.cmu.edu/TMI2002/cfp.html Questions for CFP? Please contact teruko@cs.cmu.edu or ehn@cs.cmu.edu From sharder@language.sdu.dk Thu Oct 11 13:26:47 2001 From: sharder@language.sdu.dk (=?iso-8859-1?q?S=F8ren=20Harder?=) Date: Thu, 11 Oct 2001 14:26:47 +0200 Subject: [MT-List] Q: Case Grammar in MT Message-ID: <200110111224.OAA00499@gamma.dou.dk> I have just started my PhD, where I intend to take an already existing system for part-of-speech- and sentence-function-tagging built in the Constraint Grammar framework and extend it into a MT-system for Danish to English. (see http://visl.sdu.dk/visl for the existing tagger). Right now, I am trying to delimit and implement a set of case roles in the style of Fillmore(1968) 'Case for Case' (AGENT, BENEFICIARY etc). The set should be 'practical', ie. be applicable to running text and cover all 'participants', whether they be denoted by nominals, adverbials or clauses. On the other hand I would like a 'constrained' set (currently around 25 roles) and, of course, one that will be helpful in MT. The original Fillmore articles (1966-1977) have a very constrained set of case-roles, that will not be able to cover all participants in running text. The work that Fillmore is doing now has separate roles for each 'frame' (each verb), a bit like HPSG. I have not found any suggestions for a constrained, yet full-coverage-, set by him or any other. I'd like to ask if anyone out there has any experience with case-systems like this, either in MT, CL or 'plain linguistics'. I'm interested in references as well as your thoughts and suggestions on the matter. Has this approach never been used in MT? I'm looking forward to your reply, Yours sincerely Søren Harder, PhD-student, VISL-project, University of Southern Denmark - Odense. e-mail: sharder@language.sdu.dk From ram@srv.net Thu Oct 11 19:04:19 2001 From: ram@srv.net (Rick Morneau) Date: Thu, 11 Oct 2001 12:04:19 -0600 Subject: [MT-List] Q: Case Grammar in MT Message-ID: <200110111804.f9BI4JX00952@localhost.localdomain> Søren Harder wrote: > > I'd like to ask if anyone out there has any experience with > case-systems like this, either in MT, CL or 'plain linguistics'. I'm > interested in references as well as your thoughts and suggestions on > the matter. > For a somewhat unconventional but very powerful approach to case and argument structure in MTs (using an interlingual approach), you may want to check out my monograph at: http://www.srv.net/~ram/lexical_semantics.txt Hope this helps. Regards, Rick Morneau ram@srv.net ram@axxess.net From bond@cslab.kecl.ntt.co.jp Fri Oct 12 02:16:32 2001 From: bond@cslab.kecl.ntt.co.jp (Francis Bond) Date: Fri, 12 Oct 2001 10:16:32 +0900 (JST) Subject: [MT-List] Q: Case Grammar in MT In-Reply-To: <200110111224.OAA00499@gamma.dou.dk> (message from =?iso-8859-1?q?S=F8ren=20Harder?= on Thu, 11 Oct 2001 14:26:47 +0200) References: <200110111224.OAA00499@gamma.dou.dk> Message-ID: <200110120116.KAA21199@fornost.icl.kecl.ntt.co.jp> G'day, I discuss case in MT in my paper "Practical and efficient organization of a large valency dictionary" available from . The paper also includes links to several interesting references. -- Francis Bond NTT Communication Science Laboratories | Machine Translation Research Group Come to TMI-2002 in Kyoto, Japan: ! From bscott22@verizon.net Fri Oct 12 20:37:21 2001 From: bscott22@verizon.net (Bud Scott) Date: Fri, 12 Oct 2001 15:37:21 -0400 Subject: [MT-List] Q: Case Grammar in MT Message-ID: <00a501c15355$4fb6f340$8b136018@vaio5321> This is a multi-part message in MIME format. ------=_NextPart_000_00A2_01C15333.C85166E0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable A full discussion of a case/valency-like view of language that motivated = a commercial MT system (The Logos System) can be found at an IAI = (University of Saarland) URL, to wit: = http://www.iai.uni-sb.de/~carl/iaiwp/p11/index.html. Bud Scott ------=_NextPart_000_00A2_01C15333.C85166E0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

 
A full discussion of a case/valency-like view of language that = motivated a=20 commercial MT system (The Logos System) can be found at an IAI = (University of=20 Saarland) URL, to wit:  http://www.i= ai.uni-sb.de/~carl/iaiwp/p11/index.html.
 
Bud Scott
 
------=_NextPart_000_00A2_01C15333.C85166E0-- From kgodden@justtalk.com Fri Oct 19 14:47:12 2001 From: kgodden@justtalk.com (Kurt Godden) Date: Fri, 19 Oct 2001 09:47:12 -0400 Subject: [MT-List] Swedish-English MT System? Message-ID: I'm looking for a Swedish-English MT system for gisting purposes. Primarily, we just need it to produce a draft that is good enough quality to allow an English speaker to identify documents of interest, which could then be investigated more thoroughly. Does anyone have experience with such a product that they can recommend? URLs would be helpful. Thanks. Kurt Godden, Director of Linguistics JustTalk 217 Third Street Ann Arbor, MI 48103 www.justtalk.com 734-623-7954, ext. 216 ----------------------------------------------------------- He gives us a hero every once in a while, so that we can know how we'd behave if we had any character. --David E. Davis on God's job performance From Juan-Ramon.Del-Pozo@cec.eu.int Mon Oct 22 10:15:16 2001 From: Juan-Ramon.Del-Pozo@cec.eu.int (Juan-Ramon.Del-Pozo@cec.eu.int) Date: Mon, 22 Oct 2001 11:15:16 +0200 Subject: [MT-List] Portuguese-Spanish MT System Message-ID: <91A2F218314CD4119566009027CA36EA0495505F@ex2beimcombx04> Does anybody know if there exists a Portuguese-Spanish MT System, = either a commercial one or a system used in an international organisation or multinational company? Thanks for your help. Juan Ram=F3n del Pozo Translation Service European Commission From farnaby@freezone.co.uk Sun Oct 28 19:45:41 2001 From: farnaby@freezone.co.uk (Miriam Siemmond) Date: Sun, 28 Oct 2001 19:45:41 +0000 Subject: [MT-List] xml and MT Message-ID: <3BDC6065.E99CF1A7@freezone.co.uk> Hello everybody, Just a curiosity questions: Will XML enable to produce MT's job easier? or the other way around? + How will effect Semantic web on MT? + Is imperative language better than declarative language on MT programs regards, Miriam From tree@basistech.com Sun Oct 28 19:58:12 2001 From: tree@basistech.com (Tom Emerson) Date: Sun, 28 Oct 2001 14:58:12 -0500 Subject: [MT-List] xml and MT In-Reply-To: <3BDC6065.E99CF1A7@freezone.co.uk> References: <3BDC6065.E99CF1A7@freezone.co.uk> Message-ID: <15324.25428.125903.942328@magrathea.basistech.com> Miriam Siemmond writes: > Will XML enable to produce MT's job easier? or the other way around? Take a look 'http://www.olif.net/' for a description of an XML-based terminology management system. -tree -- Tom Emerson Basis Technology Corp. Sr. Computational Linguist http://www.basistech.com "Beware the lollipop of mediocrity: lick it once and you suck forever" From Christian.Boitet@imag.fr Mon Oct 29 09:00:02 2001 From: Christian.Boitet@imag.fr (Christian Boitet) Date: Mon, 29 Oct 2001 11:00:02 +0200 Subject: [MT-List] xml and MT In-Reply-To: <3BDC6065.E99CF1A7@freezone.co.uk> References: <3BDC6065.E99CF1A7@freezone.co.uk> Message-ID: Dear all, 30/10/01 These are interesting questions! I would like to have more time to answer them, but here is a 10 mn contribution. At 19:45 +0000 28/10/01, Miriam Siemmond wrote: >Hello everybody, > >Just a curiosity questions: >Will XML enable to produce MT's job easier? or the other way around? >+ Yes because language-oriented tags can be used for easy disambiguation. Simple example: Click on save as to produce a backup copy. Click on save as to produce a backup copy. >How will effect Semantic web on MT? If we can get the UNL content representation to be included with documents as a kind of annotation in the semantic web, MT will split in 3 parts: -1- usual analysis (enconversion) into UNL, with or without interaction. Basically all kinds of techniques may be used for that purpose, from totally symbolic, rule-oriented systems to essentially statistical ones, through direct programming approaches. -2- automatic deconversion into many languages, with insertion of results in the document, as annotations using UNL or other tags. -3- interactive improvement "on demand" through interaction from any language (modifications having a NL side and a UNL side), followed by a new deconversion into all target languages, such achieving a kind of shared and "lazy" revision of only the parts felt as most important by readers. A recurring question is: OK, but we will never reach high quality using a "linguistic semantic" pivot like UNL. Answer: ATLAS-II was built by H.Uchida and is still the best quality MT system in Japan. It has 480000 dictionary entries in English and Japanese. It uses the pivot architecture, and its pivot is as it were the grandfather of the UNL language. UNL is more powerful and more practical for developers scattered all around the world. >+ >Is imperative language better than declarative language on MT programs All programming languages have both sides, incuding the claimed purely declarative languages, where control may be hidden deep but is there (examples: ANY feature in UFG, cut primitive in Prolog). To answer that question would need a quite exhaustive review and classification of programming tools used for various parts of many MT systems! >regards, >Miriam > > >-- > For MT-List info, see http://www.eamt.org/mt-list.html Regards, Ch.Boitet -- ------------------------------------------------------------------------- Christian Boitet (Pr. Universite' Joseph Fourier) Tel: +33.4-7651-4355/4817 GETA, CLIPS, IMAG-campus, BP53 Fax: +33.4-7651-4405 385, rue de la Bibliothe`que Mel: Christian.Boitet@imag.fr 38041 Grenoble Cedex 9, France Mobile: +33-(0)6-6005-1969 http://www-clips.imag.fr/geta/christian.boitet ------------------------------------------------------------------------- Serveurs de dictionnaires: projet SILFIDE (http://silfide.imag.fr) et plus particuli=E8rement fran=E7ais-malais (http://www-clips.imag.fr/geta/services/fem/) Projet C-STAR (http://www.c-star.org/) et projet europe'en Nespole (http://nespole.itc.it) de traduction de parole Projet UNL de communication et recherche d'information multilingue sur le re'seau http://www.unl.ias.unu.edu ou http://www.unl.org, essais: http://www.unl.fi.upm.es/repository/ Projet PAPILLON de construction coop=E9rative d'une base lexicale multilingue et de construction de dictionnaires http://vulab.ias.unu.edu/papillon/ From Ruben@enligne.net Tue Oct 16 15:32:23 2001 From: Ruben@enligne.net (Ruben and Brothers) Date: Tue, 16 Oct 2001 16:32:23 +0200 Subject: [MT-List] translating software Message-ID: <200110161632230570.06B97572@smtp.LaPoste.net> This is a request for advise. We are a french company and we are correponding a lot with english speaking= people, exchanging all kinds of files. We wish to find a software to use with windows that could help us= translating those files. We've heard about different translating softwares like Systran, L&H, ....= but we don't know really the performance of each, or which one is the= best. Which software would you advise us to buy ? We know that such a software would never replace a "human translator" but= we need it to quicken our communication even if the result is not so= perfect... We greatly appreciate your input for that matter. Thank you ! Sarl Ruben and Brothers 64190 Sus, Navarrenx France Ruben@enligne.net Tel: +33 5 59 66 14 28 Fax: +33 5 59 66 20 34 From Vladimir Rykov" Message-ID: <007d01c16075$5127d690$1501a8c0@Dvr04> I would like to say thanx Miriam Siemmond and Christian Boitet for a good Question and a good Answer. A shame to say - but I do not know the UNL abbreviation. P bI K O B B. B. MOCKBA PhD in Computational Linguistics rykov.narod.ru Eng version - www.blkbox.com/~gigawatt/rykov.html Mobile tel: 8-903-749-19-99 (+7-903--749-19-99 from abroad) My add address: rykov2000@hotmail.com ----- Original Message ----- From: "Christian Boitet" To: "Miriam Siemmond" ; Sent: Monday, October 29, 2001 12:00 PM Subject: Re: [MT-List] xml and MT Dear all, 30/10/01 These are interesting questions! I would like to have more time to answer them, but here is a 10 mn contribution. At 19:45 +0000 28/10/01, Miriam Siemmond wrote: >Hello everybody, > >Just a curiosity questions: >Will XML enable to produce MT's job easier? or the other way around? >+ Yes because language-oriented tags can be used for easy disambiguation. Simple example: Click on save as to produce a backup copy. Click on save as to produce a backup copy. >How will effect Semantic web on MT? If we can get the UNL content representation to be included with documents as a kind of annotation in the semantic web, MT will split in 3 parts: -1- usual analysis (enconversion) into UNL, with or without interaction. Basically all kinds of techniques may be used for that purpose, from totally symbolic, rule-oriented systems to essentially statistical ones, through direct programming approaches. -2- automatic deconversion into many languages, with insertion of results in the document, as annotations using UNL or other tags. -3- interactive improvement "on demand" through interaction from any language (modifications having a NL side and a UNL side), followed by a new deconversion into all target languages, such achieving a kind of shared and "lazy" revision of only the parts felt as most important by readers. A recurring question is: OK, but we will never reach high quality using a "linguistic semantic" pivot like UNL. Answer: ATLAS-II was built by H.Uchida and is still the best quality MT system in Japan. It has 480000 dictionary entries in English and Japanese. It uses the pivot architecture, and its pivot is as it were the grandfather of the UNL language. UNL is more powerful and more practical for developers scattered all around the world. >+ >Is imperative language better than declarative language on MT programs All programming languages have both sides, incuding the claimed purely declarative languages, where control may be hidden deep but is there (examples: ANY feature in UFG, cut primitive in Prolog). To answer that question would need a quite exhaustive review and classification of programming tools used for various parts of many MT systems! >regards, >Miriam > > >-- > For MT-List info, see http://www.eamt.org/mt-list.html Regards, Ch.Boitet -- ------------------------------------------------------------------------- Christian Boitet (Pr. Universite' Joseph Fourier) Tel: +33.4-7651-4355/4817 GETA, CLIPS, IMAG-campus, BP53 Fax: +33.4-7651-4405 385, rue de la Bibliothe`que Mel: Christian.Boitet@imag.fr 38041 Grenoble Cedex 9, France Mobile: +33-(0)6-6005-1969 http://www-clips.imag.fr/geta/christian.boitet ------------------------------------------------------------------------- Serveurs de dictionnaires: projet SILFIDE (http://silfide.imag.fr) et plus particulièrement français-malais (http://www-clips.imag.fr/geta/services/fem/) Projet C-STAR (http://www.c-star.org/) et projet europe'en Nespole (http://nespole.itc.it) de traduction de parole Projet UNL de communication et recherche d'information multilingue sur le re'seau http://www.unl.ias.unu.edu ou http://www.unl.org, essais: http://www.unl.fi.upm.es/repository/ Projet PAPILLON de construction coopérative d'une base lexicale multilingue et de construction de dictionnaires http://vulab.ias.unu.edu/papillon/ -- For MT-List info, see http://www.eamt.org/mt-list.html From ling98@videotron.ca Mon Oct 29 13:43:24 2001 From: ling98@videotron.ca (Michael Blekhman) Date: Mon, 29 Oct 2001 08:43:24 -0500 Subject: [MT-List] Comparative analysis of MT systems References: <200110161632230570.06B97572@smtp.LaPoste.net> Message-ID: <001e01c1607f$b3181ea0$baa9c818@videotron.ca> Dear colleagues, For comparative analysis of SYSTRAN, LogoMedia and PROMT, please see: www.ling98.com/test.rtf We will follow up shortly with more analysis results. Sincerely, Dr. Michael S. Blekhman, President, Lingvistica '98 Inc. & Lingvistica b.v. Montreal, Canada; Dongen, Holland. ----- Original Message ----- From: Ruben and Brothers To: Sent: Tuesday, October 16, 2001 9:32 AM Subject: [MT-List] translating software This is a request for advise. We are a french company and we are correponding a lot with english speaking people, exchanging all kinds of files. We wish to find a software to use with windows that could help us translating those files. We've heard about different translating softwares like Systran, L&H, .... but we don't know really the performance of each, or which one is the best. Which software would you advise us to buy ? We know that such a software would never replace a "human translator" but we need it to quicken our communication even if the result is not so perfect... We greatly appreciate your input for that matter. Thank you ! Sarl Ruben and Brothers 64190 Sus, Navarrenx France Ruben@enligne.net Tel: +33 5 59 66 14 28 Fax: +33 5 59 66 20 34 -- For MT-List info, see http://www.eamt.org/mt-list.html From nicole.adamides@tinyworld.co.uk Mon Nov 5 09:39:13 2001 From: nicole.adamides@tinyworld.co.uk (nicole.adamides) Date: Mon, 5 Nov 2001 09:39:13 -0000 Subject: [MT-List] Translating and the Computer 23 Conference Message-ID: <001a01c165dd$bc2677e0$964b3c3e@tiny> This is a multi-part message in MIME format. ------=_NextPart_000_0017_01C165DD.BACF2540 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Colleague,=20 You may have received the following notice already, please accept my = apologies if this is the case. The Translating and the Computer = Conference has been supported by EAMT and IAMT for a number of years. = Members are offered the membership fee. 23rd Annual Conference and Exhibition:=20 TRANSLATING AND THE COMPUTER=20 on 29-30 November 2001 at One Great George Street, London SW1 Presented by Aslib/IMI and supported by: IAMT, EAMT, BCS, IoL and ITI This conference is one of the few international events which focuses on = the user aspects of translation software and as such has been = particularly beneficial to a very wide audience including translators, = business managers, researchers and language experts. Once again, this = year the conference will address the latest developments in translation = (and translation-related) software. The keynote address will be given = by Roger Jeanty, recently retired President of Lionbridge Technologies. = He will overview the current economic, political, organization and = technological "headwinds" facing globalization, at the macro and = enterprise level. You will find the full programme (and booking = details) on the Aslib/IMI website: = http://www.aslib.com/conferences/tc23.html To support those who are full time students or lecturers, Aslib/IMI = offers a special conference fee. If you would prefer a copy of the brochure mailed to you, please = contact: Barbara Hobbs, Aslib/IMI, Staple Hall, Stone House Court, = London, EC3A 7PB or call her on: +44 (0)20 7903 0000.=20 ------=_NextPart_000_0017_01C165DD.BACF2540 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Dear = Colleague, 
You may have received the following notice already, = please=20 accept my apologies if this is the case.  The Translating and the = Computer=20 Conference has been supported by EAMT and IAMT for a number of = years. =20 Members are offered the membership fee.
 
23rd Annual = Conference and=20 Exhibition:
TRANSLATING AND THE = COMPUTER=20
on 29-30 November = 2001 at One=20 Great George Street, London SW1

Presented by Aslib/IMI=20 and supported by: IAMT, EAMT, BCS, IoL and ITI

This conference = is one of the=20 few international events which focuses on the user aspects of = translation=20 software and as such has been particularly beneficial to a very wide = audience=20 including translators, business managers, researchers and language = experts. Once=20 again, this year the conference will address the latest developments in=20 translation (and translation-related) software.  The keynote = address will=20 be given by Roger Jeanty, recently retired President of Lionbridge=20 Technologies.    He will overview the = current=20 economic, political, organization and technological = “headwinds” facing=20 globalization, at the macro and enterprise level.    = You=20 will find the full programme (and booking details) on the = Aslib/IMI=20 website:  http://www.aslib.com/= conferences/tc23.html

To support those who are full time students or lecturers,=20 Aslib/IMI offers a special conference fee.

If you would prefer a copy of the brochure mailed to you, = please=20 contact: Barbara Hobbs, Aslib/IMI, Staple Hall, Stone House Court, = London, EC3A=20 7PB or call her on: +44 (0)20 7903=20 0000. 

------=_NextPart_000_0017_01C165DD.BACF2540-- From kgodden@justtalk.com Mon Nov 5 14:26:50 2001 From: kgodden@justtalk.com (Kurt Godden) Date: Mon, 5 Nov 2001 09:26:50 -0500 Subject: [MT-List] Comparative analysis of MT systems Message-ID: Michael, In your report you claim that your analysis provides "reliable and unbiased conclusions" and unbiased conclusions, but I'm afraid I disagree. You list a 7 point scale for quality evaluation and also a "detailed, purely linguistic analysis" with 5 criteria. However, both systems rely on a 100% subjective decision by an evaluator (you), who is admittedly not fluent in the source languages. Just to take the first criterion in each system, we have: "Translation quality is unacceptable", and: "the number of perfectly translated sentences". There is nothing scientific or objective about these, or the other criteria. 100% human judgment. While the report may be useful to some, you absolutely cannot claim (well, you did, but you shouldn't have) that it is "reliable and unbiased conclusions". I would say that it is totally unreliable, and biased. Unavoidably so. There are several several quality evaluation metrics that have been published in the States and in Europe that would be much better at achieving reliability and low bias. Kurt Godden, Ph.D. Director of Linguistics JustTalk 217 Third Street Ann Arbor, MI 48103 www.justtalk.com 734-623-7954, ext. 216 ----------------------------------------------------------- The true test of a first-rate mind is the ability to hold two contradictory ideas at the same time. --F. Scott Fitzgerald > -----Original Message----- > From: Michael Blekhman [mailto:ling98@videotron.ca] > Sent: Monday, October 29, 2001 8:43 AM > To: Ruben and Brothers > Cc: mt-list@eamt.org > Subject: [MT-List] Comparative analysis of MT systems > > > Dear colleagues, > > For comparative analysis of SYSTRAN, LogoMedia and PROMT, please see: > > www.ling98.com/test.rtf > > We will follow up shortly with more analysis results. > > Sincerely, > Dr. Michael S. Blekhman, > President, > Lingvistica '98 Inc. & Lingvistica b.v. > Montreal, Canada; > Dongen, Holland. > > ----- Original Message ----- > From: Ruben and Brothers > To: > Sent: Tuesday, October 16, 2001 9:32 AM > Subject: [MT-List] translating software > > > This is a request for advise. > > We are a french company and we are correponding a lot with > english speaking > people, exchanging all kinds of files. > We wish to find a software to use with windows that could help us > translating those files. > We've heard about different translating softwares like > Systran, L&H, .... > but we don't know really the performance of each, or which > one is the best. > Which software would you advise us to buy ? > > We know that such a software would never replace a "human > translator" but we > need it to quicken our communication even if the result is > not so perfect... > > We greatly appreciate your input for that matter. > Thank you ! > Sarl Ruben and Brothers > 64190 Sus, Navarrenx > France > Ruben@enligne.net > Tel: +33 5 59 66 14 28 > Fax: +33 5 59 66 20 34 > > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > From kgodden@justtalk.com Mon Nov 5 17:07:24 2001 From: kgodden@justtalk.com (Kurt Godden) Date: Mon, 5 Nov 2001 12:07:24 -0500 Subject: [MT-List] Comparative analysis of MT systems Message-ID: All: I think Jose raises some good points, which are valid for evaluating human translations as well. I would also add that training of the evaluators themselves is also VERY important, as is following up that training with a validation of the evaluation process itself. That is, obtaining a test corpus and a translation of that corpus with deliberate errors that are known in advance; having the evaluators then perform their evaluations (hopefully with a precise quality metric, per my first note); and then evaluating the evaluations. This will reveal several things, among which are (a) deficiencies in the metric, (b) deficiencies in the training, (c) sources of variation, and others. I have participated in just such a validation of a translation quality evaluation process in the past, and I know from that experience that quality evaluation is extremely difficult. To the extent that subjectivity can be reduced, the process will be improved. But ultimately, subjectivity cannot ever be eliminated. This is equivalent to saying that there is more than one correct translation for a given text. Kurt Godden, Ph.D. Director of Linguistics JustTalk 217 Third Street Ann Arbor, MI 48103 www.justtalk.com 734-623-7954, ext. 216 ----------------------------------------------------------- Wise men are instructed by reason; men of less understanding, by experience; the most ignorant by necessity; and beasts by nature. --Cicero > -----Original Message----- > From: Coch, Jose [mailto:Jose.Coch@lingway.com] > Sent: Monday, November 05, 2001 12:00 PM > To: Kurt Godden; 'Michael Blekhman' > Cc: mt-list@eamt.org > Subject: RE: [MT-List] Comparative analysis of MT systems > > > Hello Michael, Kurt and all, > > I think that it could be useful to recall some very important > principles in MT Evaluation: > - you don't evaluate a system, you only evaluate the quality > of a system > with respect to a concrete need, in a specific domain, etc. As a > consequence, > if you think you need MT software, you must start to study > your own needs, > probably to select you own reference corpus, etc. > - you can conduct subjective quality test, but in this case, to avoid > (or to reduce...) subjective bias, you need to perform your tests > in a _blind_ situation, with _several_ evaluators. For example, I have > recently > conducted a test with 15 evaluators (by the way, the results > obtained were > different from the ones described by Michael, it is true that > the texts > came from a different domain). > - it can obviously be useful to use available "professional > dictionaries" if > they are relevant for our need. The same for the > customization capability. > > A very short list of references follows (if sb has one more > complete it > could > be a good idea to send it to the list). [references deleted] From Jose.Coch@lingway.com Mon Nov 5 17:00:20 2001 From: Jose.Coch@lingway.com (Coch, Jose) Date: Mon, 5 Nov 2001 18:00:20 +0100 Subject: [MT-List] Comparative analysis of MT systems Message-ID: <41F5DAEE117CD411BD210050BA0A7DF39A8DE7@BADIANE> Hello Michael, Kurt and all, I think that it could be useful to recall some very important=20 principles in MT Evaluation: - you don't evaluate a system, you only evaluate the quality of a = system=20 with respect to a concrete need, in a specific domain, etc. As a consequence, if you think you need MT software, you must start to study your own = needs, probably to select you own reference corpus, etc. - you can conduct subjective quality test, but in this case, to avoid=20 (or to reduce...) subjective bias, you need to perform your tests=20 in a _blind_ situation, with _several_ evaluators. For example, I have recently conducted a test with 15 evaluators (by the way, the results obtained = were different from the ones described by Michael, it is true that the texts came from a different domain). - it can obviously be useful to use available "professional = dictionaries" if they are relevant for our need. The same for the customization = capability. A very short list of references follows (if sb has one more complete it could be a good idea to send it to the list). Best regards, Jos=E9 ------------------- Some references on MT evaluation - Language and Machines. Computers in translation and linguistics. A report by the Automatic Language Processing Advisory Committee (ALPAC). National Research Council. Washington, D.C.: National Academy of = Sciences, 1966. http://books.nap.edu/books/ARC000005/html/ - Evaluating Natural Language Processing Systems. An Analysis and Review. Sparck Jones, Karen; Galliers, Julia R. 1996. Lecture Notes in Artificial Intelligence 1083, Springer, Berlin, Germany. - The EAGLES 7-step recipe. EAGLES Evaluation Working Group, 1999.=20 http://issco-www.unige.ch/projects/eagles/ewg99/7steps.html - Evaluation of Machine Translation and Translation Tools. John Hutchins. = http://ourworld.compuserve.com/homepages/WJHutchins/Eval-HLT.htm - Lehrberger, J. and Bourbeau, L. (1988). Machine translation: linguistic characteristics of MT systems and general methodology of evaluation. Amsterdam/Philadelphia, John Benjamins. - Derek Lewis. MT Evaluation: Science or Art? in Machine Translation Review, No.6, October 1997 - pages 25-36.=20 http://www.bcs.org.uk/siggroup/nalatran/nalamt68.htm Workshops=20 - MTEval Workshop. Geneva, 19-24 April 2001 http://issco-www.unige.ch/projects/isle/mt-eval-workshop.html - MT Evaluation: Who Did What To Whom September 22nd, 2001, Santiago de Compostela, Spain http://www.issco.unige.ch/projects/isle/MT-Summit-wsp.html _________________________________________________ Jos=E9 Coch, >Lingway Le M=E9li=E8s / 261, rue de Paris F-93556 Montreuil Cedex / FRANCE Tel: (33)-1-49 93 39 38 / Fax: (33)-1-49 93 39 39 E-mail: jose.coch@lingway.com www.lingway.com _________________________________________________ -----Message d'origine----- De : Kurt Godden [mailto:kgodden@justtalk.com] Envoy=E9 : lundi 5 novembre 2001 15:27 =C0 : 'Michael Blekhman' Cc : mt-list@eamt.org Objet : RE: [MT-List] Comparative analysis of MT systems Michael, In your report you claim that your analysis provides "reliable and = unbiased conclusions" and unbiased conclusions, but I'm afraid I disagree. You = list a 7 point scale for quality evaluation and also a "detailed, purely linguistic analysis" with 5 criteria. However, both systems rely on a 100% subjective decision by an = evaluator (you), who is admittedly not fluent in the source languages. Just to = take the first criterion in each system, we have: "Translation quality is unacceptable", and: "the number of perfectly translated sentences". There is nothing scientific or objective about these, or the other = criteria. 100% human judgment. While the report may be useful to some, you absolutely cannot claim = (well, you did, but you shouldn't have) that it is "reliable and unbiased conclusions". I would say that it is totally unreliable, and biased. Unavoidably so. There are several several quality evaluation metrics that have been published in the States and in Europe that would be much better at = achieving reliability and low bias. Kurt Godden, Ph.D. Director of Linguistics JustTalk 217 Third Street Ann Arbor, MI 48103 www.justtalk.com 734-623-7954, ext. 216 ----------------------------------------------------------- The true test of a first-rate mind is the ability to hold two=20 contradictory ideas at the same time. --F. Scott Fitzgerald > -----Original Message----- > From: Michael Blekhman [mailto:ling98@videotron.ca] > Sent: Monday, October 29, 2001 8:43 AM > To: Ruben and Brothers > Cc: mt-list@eamt.org > Subject: [MT-List] Comparative analysis of MT systems >=20 >=20 > Dear colleagues, >=20 > For comparative analysis of SYSTRAN, LogoMedia and PROMT, please see: >=20 > www.ling98.com/test.rtf >=20 > We will follow up shortly with more analysis results. >=20 > Sincerely, > Dr. Michael S. Blekhman, > President, > Lingvistica '98 Inc. & Lingvistica b.v. > Montreal, Canada; > Dongen, Holland. >=20 > ----- Original Message ----- > From: Ruben and Brothers > To: > Sent: Tuesday, October 16, 2001 9:32 AM > Subject: [MT-List] translating software >=20 >=20 > This is a request for advise. >=20 > We are a french company and we are correponding a lot with=20 > english speaking > people, exchanging all kinds of files. > We wish to find a software to use with windows that could help us > translating those files. > We've heard about different translating softwares like=20 > Systran, L&H, .... > but we don't know really the performance of each, or which=20 > one is the best. > Which software would you advise us to buy ? >=20 > We know that such a software would never replace a "human=20 > translator" but we > need it to quicken our communication even if the result is=20 > not so perfect... >=20 > We greatly appreciate your input for that matter. > Thank you ! > Sarl Ruben and Brothers > 64190 Sus, Navarrenx > France > Ruben@enligne.net > Tel: +33 5 59 66 14 28 > Fax: +33 5 59 66 20 34 >=20 >=20 > -- > For MT-List info, see http://www.eamt.org/mt-list.html >=20 >=20 > --=20 > For MT-List info, see http://www.eamt.org/mt-list.html >=20 --=20 For MT-List info, see http://www.eamt.org/mt-list.html From postediting@hotmail.com Mon Nov 5 17:27:00 2001 From: postediting@hotmail.com (Jeff Allen) Date: Mon, 05 Nov 2001 17:27:00 Subject: [MT-List] Comparative analysis of MT systems Message-ID: Below is an example of a set of tests of 4 commercial MT products. The publication of the magazine article dates back 2 years. Test procedure: http://www.zdnet.fr/prod/cgi-bin/a_prod.pl?ID=11310 Texts translated: http://www.zdnet.fr/prod/cgi-bin/a_prod.pl?ID=11267 Test results: http://www.zdnet.fr/produits/bure/a0011300.html These tests of 4 MT systems were conducted in late 1999. It is all written in French. Non-francophones can use their favorite on-line translation system to translate the procedure and results pages into English. List of online MT systems at: http://mason.gmu.edu/~aross2/mtgrid.htm The results of these tests give the perspective of software product testers who are outside of the MT and linguistics circles. It would be good to see an updated version of such tests. A request was made to Consumer Reports magazine a couple of years ago to conduct a similar type of comparative test of commercial MT systems, but to my knowledge it was never undertaken. Best, Jeff Allen > > -----Original Message----- > > From: Michael Blekhman [mailto:ling98@videotron.ca] > > Sent: Monday, October 29, 2001 8:43 AM > > To: Ruben and Brothers > > Cc: mt-list@eamt.org > > Subject: [MT-List] Comparative analysis of MT systems > > > > > > Dear colleagues, > > > > For comparative analysis of SYSTRAN, LogoMedia and PROMT, please see: > > > > www.ling98.com/test.rtf > > > > We will follow up shortly with more analysis results. > > > > Sincerely, > > Dr. Michael S. Blekhman, > > President, > > Lingvistica '98 Inc. & Lingvistica b.v. > > Montreal, Canada; > > Dongen, Holland. > > > > ----- Original Message ----- > > From: Ruben and Brothers > > To: > > Sent: Tuesday, October 16, 2001 9:32 AM > > Subject: [MT-List] translating software > > > > > > This is a request for advise. > > > > We are a french company and we are correponding a lot with > > english speaking > > people, exchanging all kinds of files. > > We wish to find a software to use with windows that could help us > > translating those files. > > We've heard about different translating softwares like > > Systran, L&H, .... > > but we don't know really the performance of each, or which > > one is the best. > > Which software would you advise us to buy ? > > > > We know that such a software would never replace a "human > > translator" but we > > need it to quicken our communication even if the result is > > not so perfect... > > > > We greatly appreciate your input for that matter. > > Thank you ! > > Sarl Ruben and Brothers > > 64190 Sus, Navarrenx > > France > > Ruben@enligne.net > > Tel: +33 5 59 66 14 28 > > Fax: +33 5 59 66 20 34 _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp From bscott22@verizon.net Tue Nov 6 22:09:40 2001 From: bscott22@verizon.net (Bud Scott) Date: Tue, 6 Nov 2001 17:09:40 -0500 Subject: [MT-List] Logos Message-ID: <001a01c1670f$bba4e1c0$8b136018@vaio5321> This is a multi-part message in MIME format. ------=_NextPart_000_0017_01C166E5.D2163820 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To members of the MT community: It will be of interest to the MT community to know that one of the = oldest, pioneering companies in this field, Logos Corporation, the = company that developed the Logos System, has recently ceased to exist as = an operational entity. In the early 1970's, Logos Corporation built an = English-Vietnamese production MT system under contract to the U.S. = Defense Department. The system had translated thousands of pages of = military manuals by the time the conflict ended. The then director of = the Defense Department Research and Engineering (DDR&E), Dr. John = Forster, stated that the Logos E-V System had established for the first = time the feasibility of large-scale machine translation (DDR&E Annual = Report, 1973). This was the first official assessment of a positive = nature since ALPAC quite literally trashed the notion of MT in 1966. =20 Logos Corporation was founded in 1969 by members of a Catholic kibbutz = living on a large farm in the foothills of the Catskills in New York = State. Most of the staff in the company's earlier years were drawn from = that community. Anyone interested in knowing more about the company's = founding and early history can obtain a twenty-page account of it = entitled The Logos Story, by emailing the undersigned. In its heyday, Logos Corporation had over 100 linguists and computer = people working on the system, with development operations at various = times in New York and New Jersey, Boston, Santa Clara, Saigon, Teheran, = Salerno, Montreal, Frankfurt and Saarbruecken. The system is considered = by some experts as uniquely principled, with output quality in certain = domains ranking among the very best. The system's theoretical = motivations are described in a paper available at = http://www.iai.uni-sb.de/~carl/iaiwp/p11/index.html It is expected that the Logos System, which has specialized in corporate = document translation (from English and German), will continue to be = marketed and developed by a successor organization. =20 Bernard (Bud) Scott, =20 Founder, Logos Corporation=20 Chief Architect, Logos System bscott22@verizon.net=20 =20 =20 ------=_NextPart_000_0017_01C166E5.D2163820 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

To members of the MT community:

It will be of interest to the MT community to know = that one=20 of the oldest, pioneering companies in this field, Logos Corporation, = the=20 company that developed the Logos System, has recently ceased to exist as = an=20 operational entity.  In = the early=20 1970’s, Logos Corporation built an English-Vietnamese production = MT system under=20 contract to the U.S. Defense Department. =20 The system had translated thousands of pages of military manuals = by the=20 time the conflict ended.  = The then=20 director of the Defense Department Research and Engineering (DDR&E), = Dr.=20 John Forster, stated that the Logos E-V System had established for the = first=20 time the feasibility of large-scale machine translation (DDR&E = Annual=20 Report, 1973).  This was = the first=20 official assessment of a positive nature since ALPAC quite literally = trashed the=20 notion of MT in 1966.   =

Logos Corporation was founded in 1969 by members of = a=20 Catholic kibbutz living on a large farm in the foothills of the = Catskills in New=20 York State.  Most of the = staff in=20 the company’s earlier years were drawn from that community. Anyone = interested in=20 knowing more about the company’s founding and early history can = obtain a=20 twenty-page account of it entitled The Logos Story, by emailing = the=20 undersigned.

In its heyday, Logos Corporation had over 100 = linguists and=20 computer people working on the system, with development operations at = various=20 times in New York and New Jersey, Boston, Santa Clara, Saigon, Teheran, = Salerno,=20 Montreal, Frankfurt and Saarbruecken. =20 The system is considered by some experts as uniquely principled, = with=20 output quality in certain domains ranking among the very best.  The system’s theoretical = motivations are=20 described in a paper available at =20 http://www.iai.uni-sb.de/~carl/iaiwp/p11/index.html

It is expected that the Logos System, which has = specialized=20 in corporate document translation (from English and German), will = continue to be=20 marketed and developed by a successor organization.   

Bernard (Bud)=20 Scott,  
Founder,  Logos = Corporation 
Chief = Architect, Logos=20 System
bscott22@verizon.net

 

 

------=_NextPart_000_0017_01C166E5.D2163820-- From freeder@mitre.org Wed Nov 7 16:11:27 2001 From: freeder@mitre.org (Reeder,Florence M.) Date: Wed, 07 Nov 2001 11:11:27 -0500 Subject: [MT-List] Comparative analysis of MT systems References: Message-ID: <3BE95D2F.9EEA7BB9@mitre.org> All, > A request was made to Consumer Reports magazine a couple of years > ago to conduct a similar type of comparative test of commercial MT > systems, but to my knowledge it was never undertaken. We are working this, through ISLE and NSF. The first step is to develop a set of features on which evaluations can be performed. Then, the features have to be organized to allow for very detailed evaluations of some features and very top-level scores from other features. It is an area of active research as evidenced by the taxonomy at ISI and from the Geneva workshop. flo From ptoma2000@yahoo.com Thu Nov 8 20:17:18 2001 From: ptoma2000@yahoo.com (Paul Toma) Date: Thu, 8 Nov 2001 12:17:18 -0800 (PST) Subject: [MT-List] Comparative analysis of MT systems In-Reply-To: <3BE95D2F.9EEA7BB9@mitre.org> Message-ID: <20011108201718.19023.qmail@web9304.mail.yahoo.com> what are your common needs for an MT system? what would your needs be if you had a Speech to Speech MT system? thanks --- "Reeder,Florence M." wrote: > All, > > > A request was made to Consumer Reports magazine a > couple of years > > ago to conduct a similar type of comparative test > of commercial MT > > systems, but to my knowledge it was never > undertaken. > > We are working this, through ISLE and NSF. The > first step is to develop > a set of features on which evaluations can be > performed. Then, the > features have to be organized to allow for very > detailed evaluations of > some features and very top-level scores from other > features. It is an > area of active research as evidenced by the taxonomy > at ISI and from > the Geneva workshop. > > flo > > > > -- > For MT-List info, see http://www.eamt.org/mt-list.html __________________________________________________ Do You Yahoo!? Find a job, post your resume. http://careers.yahoo.com From cb@lim.nl Tue Nov 13 13:03:34 2001 From: cb@lim.nl (Colin Brace) Date: Tue, 13 Nov 2001 14:03:34 +0100 Subject: [MT-List] MT Summit proceedings Message-ID: <20011113130340.SOO1172.amsfep11-int.chello.nl@[62.108.30.75]> Hi all, A limited number of copies of the proceedings from the recent MT Summit are now available, either in CDROM or print format (attendees received both). The CDROM contains papers in Adobe Acrobat (PDF) or Word format. Prices (including shipping) are: Print Non-members: Euro 25 EAMT/IAMT members: Euro 20 CDROM Non-members: Euro 10 EAMT/IAMT members: Euro 8 Payment may be made by credit card or bank transfer. To order a copy, please fill out the order form on the EAMT website: http://www.eamt.org/summitVIII/proceedingsorderform.html You can view an alphabetic listing of papers by author at: http://www.eamt.org/summitVIII/papers.html There are also some copies left of the proceedings from the workshops; please contact us if you are interested: summitviii@eamt.org -- Colin Brace Amsterdam http://www.lim.nl From vidura_s@yahoo.com Wed Nov 21 03:42:07 2001 From: vidura_s@yahoo.com (Vidura Seneviratne) Date: Tue, 20 Nov 2001 19:42:07 -0800 (PST) Subject: [MT-List] confirmation Message-ID: <20011121034207.75940.qmail@web20704.mail.yahoo.com> Please send the list __________________________________________________ Do You Yahoo!? Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month. http://geocities.yahoo.com/ps/info1 From wiggjd@sbu.ac.uk Mon Nov 26 11:37:55 2001 From: wiggjd@sbu.ac.uk (David Wigg) Date: Mon, 26 Nov 2001 11:37:55 +0000 Subject: [MT-List] BCS NLTSG AGM and meeting Message-ID: <3C022992.C41FC93B@sbu.ac.uk> THE BRITISH COMPUTER SOCIETY NATURAL LANGUAGE TRANSLATION SPECIALIST GROUP http://www.bcs.org.uk/siggroup/nalatran Seminar on Machine Translation Derek Lewis (School of Modern Languages, Exeter) will demonstrate Machine Translation on a PC with a view to showing the extent of satisfactory MT rather than its shortcomings (French, German, Italian, Portuguese and Spanish will be available) Derek will be pleased to try translating simple sentences suggested by the audience. If you would like to suggest some sentences in advance please send them to wiggjd@bcs.org.uk and it may be possible to get them translated and printed for you at the meeting. Venue: Room G80 King's College London Franklin Wilkins Building 150 Stamford Street London, SE1 9NN Please note that this is not our usual place in the Strand. Stamford Street is opposite the main entrance to Waterloo Station. Take underpass in the direction of The National Theatre and exit to Stamford Street where indicated. Tuesday, 11 December 2001, 6.00 - 8.00 p.m. Entrance free - all welcome. Come when you can. This demonstration will be preceded by a brief AGM of the Group Please note that the committee would welcome new members, particularly for Treasurer or Auditor. If you might be interested in helping the committee contact any committee member for further information. As usual, the committee will entertain the speaker afterwards in a local restaurant and participants are welcome to come along at their own expense to continue the discussion in more convivial surroundings. As this meeting is quite near Christmas we have had to book a limited number of places so if you would like to join us after the meeting please let the Secretary know as soon as possible. For further details, please contact the Group's Secretary, Monique L'Huillier, 01784 443243 or 01276 20488 or m.l'huillier@rhul.ac.uk. Please could you post and/or circulate this notice From malek.boualem@rd.francetelecom.com Thu Nov 29 13:32:55 2001 From: malek.boualem@rd.francetelecom.com (BOUALEM Malek FTRD/DMI/LAN) Date: Thu, 29 Nov 2001 14:32:55 +0100 Subject: [MT-List] BCS NLTSG AGM and meeting Message-ID: Dear David, Thank you for your invitation to this interesting seminar. Unfortunately I will not able to attend it but following your suggestion, I would like to suggest some sentences to be translated. Of course I would be glad to receive them translated in French. 1) The bandage was wound around the wound. 2) The farm was used to produce produce. 3) The dump was so full that it had to refuse more refuse. 4) We must polish the Polish furniture. 5 He could lead if he would get the lead out. 6) The soldier decided to desert his dessert in the desert. 7) Since there is no time like the present, he thought it was time to present the present. 8) A bass was painted on the head of the bass drum. 9) When shot at, the dove dove into the bushes. 10) I did not object to the object. 11) The insurance was invalid for the invalid. 12) There was a row among the oarsmen about how to row. 13) They were too close to the door to close it. 14) The buck does funny things when the does are present. 15) A seamstress and a sewer fell down into a sewer line. 16) To help with planting, the farmer taught his sow to sow. 17) The wind was too strong to wind the sail. 18) After a number of injections my jaw got number. 19) Upon seeing the tear in the painting I shed a tear. 20) I had to subject the subject to a series of tests. 21) How can I intimate this to my most intimate friend? Thanks PS. I do not own the Copyright of these sentences. Malek Boualem France Telecom R&D > -----Original Message----- > From: David Wigg [mailto:wiggjd@sbu.ac.uk] > Sent: Monday, November 26, 2001 12:38 PM > To: mt-list@eamt.org > Subject: [MT-List] BCS NLTSG AGM and meeting > > > THE BRITISH COMPUTER SOCIETY > NATURAL LANGUAGE TRANSLATION SPECIALIST GROUP > http://www.bcs.org.uk/siggroup/nalatran > > Seminar on Machine Translation > > Derek Lewis (School of Modern Languages, Exeter) > > will demonstrate Machine Translation on a PC with a view to showing > the extent of satisfactory MT rather than its shortcomings > > (French, German, Italian, Portuguese and Spanish will be available) > > Derek will be pleased to try translating simple sentences suggested > by the audience. If you would like to suggest some sentences > in advance > please send them to wiggjd@bcs.org.uk and it may be possible > to get them > translated and printed for you at the meeting. > > Venue: Room G80 > King's College London > Franklin Wilkins Building > 150 Stamford Street > London, SE1 9NN > > Please note that this is not our usual place in the Strand. > Stamford Street is opposite the main entrance to Waterloo Station. > Take underpass in the direction of The National Theatre and > exit to Stamford Street where indicated. > > Tuesday, 11 December 2001, 6.00 - 8.00 p.m. > > Entrance free - all welcome. Come when you can. > > This demonstration will be preceded by a brief AGM of the Group > > Please note that the committee would welcome new members, particularly > for Treasurer or Auditor. If you might be interested in helping the > committee contact any committee member for further information. > > As usual, the committee will entertain the speaker afterwards > in a local > restaurant and participants are welcome to come along at their own > expense to continue the discussion in more convivial surroundings. As > this meeting is quite near Christmas we have had to book a limited > number of places so if you would like to join us after the meeting > please let the Secretary know as soon as possible. > > For further details, please contact the Group's Secretary, > Monique L'Huillier, 01784 443243 or 01276 20488 or > m.l'huillier@rhul.ac.uk. > > Please could you post and/or circulate this notice > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > From steveri@microsoft.com Fri Nov 30 07:06:47 2001 From: steveri@microsoft.com (Steve Richardson) Date: Thu, 29 Nov 2001 23:06:47 -0800 Subject: [MT-List] AMTA-2002 Preliminary Call for Participation Message-ID: <0FDD2891FCDF6E42891AEDE2198E5F7C03B9D46B@red-msg-04.redmond.corp.microsoft.com> This is a multi-part message in MIME format. --------------InterScan_NT_MIME_Boundary Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C1796D.93B3C8D1" ------_=_NextPart_001_01C1796D.93B3C8D1 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable --- PRELIMINARY CALL FOR PARTICIPATION --- The Association for Machine Translation in the Americas =20 AMTA-2002 Conference Location: Tiburon, California Dates: October 8-12, 2002 =20 The Association for Machine Translation in the Americas (AMTA) is = pleased to announce its fifth biennial conference, planned for October = 8-12, 2002, in Tiburon (near San Francisco), California. =20 Conference theme: From Research to Real Users =20 Ever since the showdown between Empiricists and Rationalists a decade = ago at TMI-92, MT researchers have hotly pursued promising paradigms for = MT, including data-driven approaches (e.g., statistical, example-based) = and hybrids that integrate these with more traditional rule-based = components. =20 During the same period, commercial MT systems with standard transfer = architectures have evolved along a parallel and almost unrelated track, = increasing their coverage (primarily through manual update of their = lexicons, we assume) and achieving much broader acceptance and usage, = principally through the medium of the Internet. Web page translators = have become commonplace; a number of online translation services have = appeared, including in their offerings both raw and post-edited MT; and = large corporations have been turning increasingly to MT to address the = exigencies of global communication. Still, the output of the = transfer-based systems employed in this expansion represents but a small = drop in the ever-growing translation marketplace bucket. =20 Now, 10 years later, we wonder if this mounting variety of MT users is = any better off, and if the promise of the research technologies is being = realized to any measurable degree. In this regard, we pose the = following questions: =20 Why aren't any current commercially available MT systems primarily = data-driven? =20 Do any commercially available systems integrate (or plan to integrate) = data-driven components? =20 Do data-driven systems have significant performance or quality issues? =20 Can such systems really provide better quality to users, or is their = main advantage one of fast, facilitated customization? =20 If any new MT technology could provide such benefits (somewhat higher = quality, or facilitated customization), would that be the key to more = widespread use of MT, or are there yet other more relevant unresolved = issues, such as system integration? =20 If better quality, customization, or system integration aren't the = answer, then what is it that users really need from MT in order for it = to be more useful to them? =20 We solicit participation on these and other topics related to the = research, development, and use of MT in the form of original papers, = demonstrations, workshops, tutorials, and panels. We invite all who are = interested in MT to participate, including developers, researchers, end = users, professional translators, managers, and marketing experts. We = especially invite users to share their experiences, developers to = describe their novel systems, managers and marketers to talk about what = is happening in the marketplace, researchers to detail new capabilities = or methods, and visionaries to describe the future as they see it. We = also welcome and encourage participation by members of AMTA's sister = organizations, AAMT in Asia and EAMT in Europe.=20 =20 For planning purposes, preliminary dates for submissions are as follows: Submissions due: April 15, 2002 (Monday) Notification of acceptance: May 31, 2002 (Friday) Final versions due: July 15, 2002 (Monday) =20 Details regarding the conference, including submission guidelines, will = be provided shortly on the AMTA Web site: http://www.amtaweb.org =20 =20 Elliott Macklovitch, General Chair Stephen D. Richardson, Program Chair =20 =20 ------_=_NextPart_001_01C1796D.93B3C8D1 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

       &nbs= p;    --- PRELIMINARY CALL FOR PARTICIPATION ---

=A0=A0=A0=A0=A0 The = Association for Machine Translation in the = Americas

 

AMTA-2002 = Conference

Location:=A0 = Tiburon, California

Dates:=A0 = October 8-12, 2002

 

The Association for = Machine Translation in the Americas = (AMTA) is pleased to announce its fifth biennial conference, planned for October = 8-12, 2002, in Tiburon (near San Francisco), California.

 

Conference theme: = From Research to Real Users

 

Ever since the = showdown between Empiricists and Rationalists a decade ago at TMI-92, MT = researchers have hotly pursued promising paradigms for MT, including data-driven = approaches (e.g., statistical, example-based) and hybrids that integrate these with = more traditional rule-based components.

 

During the same = period, commercial MT systems with standard transfer architectures have evolved = along a parallel and almost unrelated track, increasing their coverage = (primarily through manual update of their lexicons, we assume) and achieving much = broader acceptance and usage, principally through the medium of the Internet. = Web page translators have become commonplace; a number of online translation = services have appeared, including in their offerings both raw and post-edited MT; = and large corporations have been turning increasingly to MT to address the exigencies of global = communication.  Still, the output of the transfer-based systems employed in this expansion represents but a small drop in the ever-growing translation marketplace bucket.

 

Now, 10 years = later, we wonder if this mounting variety of MT users is any better off, and if = the promise of the research technologies is being realized to any measurable degree.  In this regard, we pose the following = questions:

 

Why aren't any = current commercially available MT systems primarily data-driven?

 

Do any commercially available systems integrate (or plan to integrate) data-driven = components?

 

Do data-driven = systems have significant performance or quality issues?

 

Can such systems = really provide better quality to users, or is their main advantage one of fast, facilitated customization?

 

If any new MT = technology could provide such benefits (somewhat higher quality, or facilitated customization), would that be the key to more widespread use of MT, or = are there yet other more relevant unresolved issues, such as system = integration?

 

If better quality, customization, or system integration aren't the answer, then what is it = that users really need from MT in order for it to be more useful to = them?

 

We solicit = participation on these and other topics related to the research, development, and use of = MT in the form of original papers, demonstrations, workshops, tutorials, and = panels. We invite all who are interested in MT to participate, including = developers, researchers, end users, professional translators, managers, and marketing experts. We especially invite users to share their experiences, developers to = describe their novel systems, managers and marketers to talk about what is = happening in the marketplace, researchers to detail new capabilities or methods, and visionaries to describe the future as they see it.  We also welcome = and encourage participation by members of AMTA's sister organizations, AAMT = in Asia and EAMT in Europe.

 

For planning = purposes, preliminary dates for submissions are as follows:

Submissions = due:=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 = April 15, 2002 (Monday)

Notification of = acceptance:=A0=A0=A0=A0=A0=A0=A0=A0 May 31, 2002 (Friday)

Final versions = due:=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 July 15, 2002 (Monday)

 

Details regarding = the conference, including submission guidelines, will be provided shortly on = the AMTA Web site:

http://www.amtaweb.org

 <= /font>

Elliott = Macklovitch, General Chair

Stephen D. = Richardson, Program Chair

 

 

------_=_NextPart_001_01C1796D.93B3C8D1-- --------------InterScan_NT_MIME_Boundary-- From flanders1@libero.it Fri Nov 30 10:42:42 2001 From: flanders1@libero.it (sergio) Date: Fri, 30 Nov 2001 11:42:42 +0100 Subject: [MT-List] questions on MT usage Message-ID: <3C0762A1.919B21EB@libero.it> --------------D5DBE35A1EAEE8D9EA485BE8 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Dear Sir, I joined this mailing list since I am currently working on a dissertation on MT. As premise, I assumed that MT usage will grow in the next years, thus raising new consequences. I have a few questions over this last point: * What can possibily change in a professional translator's approach? And what about his required background knowledge? * Translating will become mainly a post-editing activity? * Will translators have to use MT software (Dèjà Vu, Trados, etc.) like everybody use word-processors today? Thanks Yours sincerely Sergio Russo flanders1@libero.it www.chez.com/lesergio --------------D5DBE35A1EAEE8D9EA485BE8 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Dear Sir,
I joined this mailing list since I am currently working on a dissertation on MT. As premise, I assumed that MT usage will grow in the next years, thus raising new consequences. I have a few questions over this last point:
  • What can possibily change in a professional translator's approach? And what about his required background knowledge?
  • Translating will become mainly a post-editing activity?
  • Will translators have to use MT software (Dèjà Vu, Trados, etc.) like everybody use word-processors today?


Thanks

Yours sincerely

Sergio Russo
flanders1@libero.it
www.chez.com/lesergio --------------D5DBE35A1EAEE8D9EA485BE8-- From wjassem@math.amu.edu.pl Fri Nov 30 21:20:10 2001 From: wjassem@math.amu.edu.pl (Wiktor Jassem) Date: Fri, 30 Nov 2001 22:20:10 +0100 Subject: [MT-List] Re: MT-List digest, Vol 1 #35 - 7 msgs References: <20011130073203.4F2E553AFA@pairlist.net> Message-ID: <3C07F80A.D575B080@math.amu.edu.pl> > them translated in French. > > 1) The bandage was wound around the wound. > 2) The farm was used to produce produce. > 3) The dump was so full that it had to refuse more refuse. > 4) We must polish the Polish furniture. > 5 He could lead if he would get the lead out. > 6) The soldier decided to desert his dessert in the desert. > 7) Since there is no time like the present, he thought it was time to > present the present. > 8) A bass was painted on the head of the bass drum. > 9) When shot at, the dove dove into the bushes. > 10) I did not object to the object. > 11) The insurance was invalid for the invalid. > 12) There was a row among the oarsmen about how to row. > 13) They were too close to the door to close it. > 14) The buck does funny things when the does are present. > 15) A seamstress and a sewer fell down into a sewer line. > 16) To help with planting, the farmer taught his sow to sow. > 17) The wind was too strong to wind the sail. > 18) After a number of injections my jaw got number. > 19) Upon seeing the tear in the painting I shed a tear. > 20) I had to subject the subject to a series of tests. > 21) How can I intimate this to my most intimate friend? > Here is the translation I have downloaded from www.SYSTRANsoft.com Le bandage a été enroulé autour de la blessure. La ferme a été employée pour produire le produit. Le vidage mémoire était si plein qu'il ait dû refuser plus d'ordures. Nous devons polir les meubles polonais. Il pourrait mener s'il obtiendrait le fil de sortie dehors. Le soldat a décidé d'abandonner son dessert dans le désert. Puisqu'il n'y a plus de temps comme le présent, il a pensé qu'il était temps de présenter le présent. Une basse a été peinte sur la tête du tambour bas. Quand le projectile à, la colombe a plongé dans les buissons. Je ne me suis pas opposé à l'objet. L'assurance était incorrecte pour l'incorrect. Il y avait une ligne parmi oarsmen au sujet de la façon ramer. Ils étaient trop près de la porte pour le clôturer. Le mâle fait des choses drôles quand sont présents. Un ouvrière couturier et un égout sont tombés vers le bas dans une ligne d'égout. Pour aider avec la plantation, le fermier a enseigné sa truie à semer. Le vent était trop fort pour enrouler la voile. Après un certain nombre d'injections ma mâchoire a obtenu le nombre. En voyant la larme dans la peinture j'ai jeté une larme. J'ai dû soumettre sujet à une série d'essais. Comment ose-t-il I intime ceci à mon ami plus intime? . > Wiktor Jassem > From wjassem@math.amu.edu.pl Fri Nov 30 21:26:02 2001 From: wjassem@math.amu.edu.pl (Wiktor Jassem) Date: Fri, 30 Nov 2001 22:26:02 +0100 Subject: [MT-List] Re: MT-List digest, Vol 1 #35 - 7 msgs References: <20011130073203.4F2E553AFA@pairlist.net> Message-ID: <3C07F96A.7ED5839E@math.amu.edu.pl> > > 1) The bandage was wound around the wound. > 2) The farm was used to produce produce. > 3) The dump was so full that it had to refuse more refuse. > 4) We must polish the Polish furniture. > 5 He could lead if he would get the lead out. > 6) The soldier decided to desert his dessert in the desert. > 7) Since there is no time like the present, he thought it was time to > present the present. > 8) A bass was painted on the head of the bass drum. > 9) When shot at, the dove dove into the bushes. > 10) I did not object to the object. > 11) The insurance was invalid for the invalid. > 12) There was a row among the oarsmen about how to row. > 13) They were too close to the door to close it. > 14) The buck does funny things when the does are present. > 15) A seamstress and a sewer fell down into a sewer line. > 16) To help with planting, the farmer taught his sow to sow. > 17) The wind was too strong to wind the sail. > 18) After a number of injections my jaw got number. > 19) Upon seeing the tear in the painting I shed a tear. > 20) I had to subject the subject to a series of tests. > 21) How can I intimate this to my most intimate friend? > > Thanks > > > > > THE BRITISH COMPUTER SOCIETY > > NATURAL LANGUAGE TRANSLATION SPECIALIST GROUP > > http://www.bcs.org.uk/siggroup/nalatran > > Here is the translatation I have downloaded from www.SYSTRANsoft.com: Le bandage a été enroulé autour de la blessure. La ferme a été employée pour produire le produit. Le vidage mémoire était si plein qu'il ait dû refuser plus d'ordures. Nous devons polir les meubles polonais. Il pourrait mener s'il obtiendrait le fil de sortie dehors. Le soldat a décidé d'abandonner son dessert dans le désert. Puisqu'il n'y a plus de temps comme le présent, il a pensé qu'il était temps de présenter le présent. Une basse a été peinte sur la tête du tambour bas. Quand le projectile à, la colombe a plongé dans les buissons. Je ne me suis pas opposé à l'objet. L'assurance était incorrecte pour l'incorrect. Il y avait une ligne parmi oarsmen au sujet de la façon ramer. Ils étaient trop près de la porte pour le clôturer. Le mâle fait des choses drôles quand sont présents. Un ouvrière couturier et un égout sont tombés vers le bas dans une ligne d'égout. Pour aider avec la plantation, le fermier a enseigné sa truie à semer. Le vent était trop fort pour enrouler la voile. Après un certain nombre d'injections ma mâchoire a obtenu le nombre. En voyant la larme dans la peinture j'ai jeté une larme. J'ai dû soumettre sujet à une série d'essais. Comment ose-t-il I intime ceci à mon ami plus intime? . Best Wiktor Jassem From wiggjd@sbu.ac.uk Sat Dec 1 16:59:17 2001 From: wiggjd@sbu.ac.uk (David Wigg) Date: Sat, 01 Dec 2001 16:59:17 +0000 Subject: [MT-List] Re: MT-List digest, Vol 1 #35 - 7 msgs References: <20011130073203.4F2E553AFA@pairlist.net> <3C07F96A.7ED5839E@math.amu.edu.pl> Message-ID: <3C090C65.F799443@sbu.ac.uk> Hello Wiktor, Thanks very much for translating these sentences for us. The thing that interests me initially is the question of receiving the accented letters correctly. Did anyone receive the French sentences showing the accented letters correctly? If so, I would like to know how it was done. Thanks. David. Wiktor Jassem wrote: > > > > > 1) The bandage was wound around the wound. > > 2) The farm was used to produce produce. > > 3) The dump was so full that it had to refuse more refuse. > > 4) We must polish the Polish furniture. > > 5 He could lead if he would get the lead out. > > 6) The soldier decided to desert his dessert in the desert. > > 7) Since there is no time like the present, he thought it was time to > > present the present. > > 8) A bass was painted on the head of the bass drum. > > 9) When shot at, the dove dove into the bushes. > > 10) I did not object to the object. > > 11) The insurance was invalid for the invalid. > > 12) There was a row among the oarsmen about how to row. > > 13) They were too close to the door to close it. > > 14) The buck does funny things when the does are present. > > 15) A seamstress and a sewer fell down into a sewer line. > > 16) To help with planting, the farmer taught his sow to sow. > > 17) The wind was too strong to wind the sail. > > 18) After a number of injections my jaw got number. > > 19) Upon seeing the tear in the painting I shed a tear. > > 20) I had to subject the subject to a series of tests. > > 21) How can I intimate this to my most intimate friend? > > > > Thanks > > > > > > > > THE BRITISH COMPUTER SOCIETY > > > NATURAL LANGUAGE TRANSLATION SPECIALIST GROUP > > > http://www.bcs.org.uk/siggroup/nalatran > > > > > Here is the translatation I have downloaded from www.SYSTRANsoft.com: > > Le bandage a iti enrouli autour de la blessure. La ferme a iti employie pour > produire le produit. Le vidage mimoire itait si plein qu'il ait d{ refuser > plus d'ordures. Nous devons polir les meubles polonais. Il pourrait mener > s'il obtiendrait le fil de sortie dehors. Le soldat a dicidi d'abandonner son > dessert dans le disert. Puisqu'il n'y a plus de temps comme le prisent, il a > pensi qu'il itait temps de prisenter le prisent. Une basse a iti peinte sur > la tjte du tambour bas. Quand le projectile `, la colombe a plongi dans les > buissons. Je ne me suis pas opposi ` l'objet. L'assurance itait incorrecte > pour l'incorrect. Il y avait une ligne parmi oarsmen au sujet de la fagon > ramer. Ils itaient trop prhs de la porte pour le cltturer. Le mble fait des > choses drtles quand sont prisents. Un ouvrihre couturier et un igout sont > tombis vers le bas dans une ligne d'igout. Pour aider avec la plantation, le > fermier a enseigni sa truie ` semer. Le vent itait trop fort pour enrouler la > voile. Aprhs un certain nombre d'injections ma mbchoire a obtenu le nombre. > En voyant la larme dans la peinture j'ai jeti une larme. J'ai d{ soumettre > sujet ` une sirie d'essais. Comment ose-t-il I intime ceci ` mon ami plus > intime? > > . > > Best > > Wiktor Jassem > > -- > For MT-List info, see http://www.eamt.org/mt-list.html From guessoum@sharjah.ac.ae Sun Dec 2 05:27:26 2001 From: guessoum@sharjah.ac.ae (Ahmed Guessoum) Date: Sun, 2 Dec 2001 09:27:26 +0400 Subject: [MT-List] Re: MT-List digest, Vol 1 #35 - 7 msgs References: <20011130073203.4F2E553AFA@pairlist.net> <3C07F96A.7ED5839E@math.amu.edu.pl> <3C090C65.F799443@sbu.ac.uk> Message-ID: <001101c17af2$07633120$b12d10ac@sharjah.ac.ae> Hi David, > The thing that interests me initially is the question of receiving the > accented letters correctly. > > Did anyone receive the French sentences showing the accented letters > correctly? > > If so, I would like to know how it was done. > Yes; I did. In fact, it's only after I received your message that I realised that the accented e's were turned into i's (in your message). I use Outlook Express 5 (in a Windows 98 environment). Ahmed Guessoum. From postediting@hotmail.com Mon Dec 3 17:00:35 2001 From: postediting@hotmail.com (Jeff Allen) Date: Mon, 03 Dec 2001 17:00:35 Subject: [MT-List] postediting answers (was: questions on MT usage) Message-ID: >Sergio Russo flanders1@libero.it >To: MT list >Subject: [MT-List] questions on MT usage >I assumed that MT usage will grow in the >next years, thus raising new consequences. I have a few questions over >this last point: > * What can possibily change in a professional translator's approach? > And what about his required background knowledge? This is partly described in: "Repairing Texts: Empirical Investigations of Machine Translation Post-Editing Processes" by Hans Krings. A short abstract about the book and ordering information is available at: http://bookmasters.com/ksu-press/ksu071.htm My long review of the book is expected to appear in upcoming issue 46 of Multilingual Computing and Technology magazine (http://www.multilingual.com). Given that the book was recently published several years after having been written, there are many points that I address in the review which give up-to-date information on the status of postediting. > * Translating will become mainly a post-editing activity? There has been a considerable increase in the number of contracted posteditors and postediting agencies for the European Commission. See a current list of contractors at: http://europe.eu.int/comm/translation/free-lance/en/ao00-en.html In North America, a few localization and translation agencies are known to offer MT postediting: Able International, Detroit Translation Bureau, Linguistics Systems Inc, Omnilingua, etc. Note however that in all of these cases, the type of postediting conducted by the external posteditors (agencies and individuals) is sometimes referred to as passive postediting. They receive pre-translated files and then must conduct the postediting in their own desktop publishing environment (such as in MS Word). Passive postediting is less productive compared to interactive postediting (cf: Postediting: an integrated part of a translation software program. In Language International magazine, Vol. 13, No. 2, April 2001. Pp. 26-29. Available at: http://www.language-international.com/) The US Air Force, the Pan American Health Organization, and a few large corporations have integrated MT and postediting into their translation processes. Some use the passive postediting approach and others use the interactive approach. It does not appear that MT post-editing is replacing other forms of translation. It is rather a type of translation service that is offered in parallel to human and translation-memory aided translation, and which is increasing over time for certain types of translated documents, and for certain organizations. > * Will translators have to use MT software (Dèjà Vu, Trados, etc.) > like everybody use word-processors today? Trados Workbench, Atril Déjà Vu, Star Transit, SDL's SDLX, Wordfast, and others are examples of TM (Translation Memory) software, not MT (machine translation) software. Yes, experience on TM systems is becoming more important than ever. Many help request messages on translation lists (LANTRA, traducteurs, etc) indicate how to optimize the use of TM software. Also, experience on TM software is often indicated as an entry level requirement for jobs in the translation field. Will MT software become a necessary skill? Probably not a necessary skill, but certainly a unique selling point for one's CV/resume in order to get a foot in the door within the market of postediting texts. Best, Jeff ---- Jeff Allen postediting@hotmail.com _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp From twangsquelch@yahoo.com Tue Dec 4 16:47:02 2001 From: twangsquelch@yahoo.com (Heather Bateman) Date: Tue, 4 Dec 2001 08:47:02 -0800 (PST) Subject: [MT-List] Eurotra Message-ID: <20011204164702.24000.qmail@web13303.mail.yahoo.com> Hello everybody, I was just wondering if anybody could tell me what happened to Eurotra? I'm doing some research into MT for my dissertation and I keep seeing it mentioned in the books from the late 70's - early 80's. It seemed like everyone was really excited and hopeful about it... and now, the EU is using SYSTRAN and everyone's gone strangely quiet on the subject. Was it just a question of withdrawn funding? Or were there serious problems with the program that couldn't be overcome? Does it still live on in any form? If anyone could help me out with information, or where to find information, I would be really grateful. Heather (University of Durham, England) __________________________________________________ Do You Yahoo!? Buy the perfect holiday gifts at Yahoo! Shopping. http://shopping.yahoo.com From cb@lim.nl Tue Dec 4 16:54:22 2001 From: cb@lim.nl (Colin Brace) Date: Tue, 04 Dec 2001 17:54:22 +0100 Subject: [MT-List] Eurotra In-Reply-To: <20011204164702.24000.qmail@web13303.mail.yahoo.com> Message-ID: <20011204170338.HNVP1261.amsfep15-int.chello.nl@[62.108.30.75]> In <20011204164702.24000.qmail@web13303.mail.yahoo.com>, on 12/04/01 at 08:47 AM, Heather Bateman said: > I was just wondering if anybody could tell me what > happened to Eurotra? There are numerous (non-technical) articles on Eurotra (and various spinoffs) in the archives of Language Industry Monitor, the newsletter I published from 1991 through 1996. The are indexed by Google, and if you perform the following query: http://www.google.com/search?q=site:lim.nl+eurotra&num=50&hl=en&filter=0 you will get a whole list of articles on Eurotra or which mentioned it. They will give you a general idea of what happened to it. To browse the Monitor archives by category and article name, go here: http://www.lim.nl/monitor/ You will find a wealth of articles on other translation systems as well. HTH -- Colin Brace Amsterdam http://www.lim.nl From WJHutchins@compuserve.com Wed Dec 5 13:48:13 2001 From: WJHutchins@compuserve.com (John Hutchins) Date: Wed, 5 Dec 2001 08:48:13 -0500 Subject: [MT-List] Eurotra Message-ID: <200112050848_MC3-E968-BC54@compuserve.com> Dear Heather, >I was just wondering if anybody could tell me what >happened to Eurotra? As well as the non-technical articles mentioned by Colin Brace, you will find substantial information about Eurotra of a more technical nature (with numerous references) in the book I wrote with Harold Somers: "An introduction to machine translation" (Academic Press, 1992) pages 239-258, The book is available in most UK academic = libraries. The Eurotra project ended very soon afterwards. >Was it just a question of withdrawn funding? Or were >there serious problems with the program that couldn't >be overcome? Both are true, but the picture is complex. The chapter in the book above indicates what the technical problems were. The last funded phase of the project ended in 1992. The final review took place in 1993; and the final report from the EC, which = summed up the project, appeared in 1994 ("Final evaluation of the = results of Eurotra" COM (94) 69). From the EC viewpoint the research of Eurotra was continued in the subsequent Language Engineering projects. = >Does it still live on in any form? There have been a number of Eurotra-inspired MT systems, e.g.: PaTrans (Denmark) for translating patents CAT2 (Saarbruecken, etc.) a research system I can send references about these to you if you are interested. John Hutchins 5 Dec =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D Email: WJHutchins@compuserve.com Web: http//ourworld.compuserve.com/homepages/WJHutchins From al-asali@un.org Thu Dec 6 10:45:10 2001 From: al-asali@un.org (Saad Al-Asali) Date: Thu, 6 Dec 2001 13:45:10 +0300 Subject: [MT-List] Arabic/English MT Systems Message-ID: Dear all, I am preparing an updated survey on the above. So far I managed to collect a few names of companies but I couldn't find a single interested academic institute. I would be very grateful if you could help me providing me with names of companies, academic institutes, scholars and other interested individuals. Regards, From Cameron.Ross@cec.eu.int Thu Dec 6 16:35:19 2001 From: Cameron.Ross@cec.eu.int (Cameron.Ross@cec.eu.int) Date: Thu, 6 Dec 2001 17:35:19 +0100 Subject: [MT-List] postediting answers (was: questions on MT usage) Message-ID: >There has been a considerable increase in the number of contracted >posteditors and postediting agencies for the European Commission. See a >current list of contractors at: >http://europe.eu.int/comm/translation/free-lance/en/ao00-en.html Sorry, but this list refers to providers of conventional translation services (for target languages other than EN, see: http://europe.eu.int/comm/translation/free-lance/en/ao00-lc.html). The list of Commission post-editing contractors is much shorter and has not been posted on the Web. Regards, Cameron Ross Commission MT team From cb@lim.nl Thu Dec 6 17:24:39 2001 From: cb@lim.nl (Colin Brace) Date: Thu, 06 Dec 2001 18:24:39 +0100 Subject: [MT-List] MT Summit papers now online Message-ID: <20011206172438.XDOJ1203.amsfep14-int.chello.nl@[62.108.30.75]> Hi everyone, All workshop and conference papers from the recent MT Summit are now online. For the conference papers, see http://www.eamt.org/summitVIII/papers.html For the workshop papers, see http://www.eamt.org/summitVIII/workshop-papers.html Nearly all are in Adobe Acrobat (PDF) format. If you don't have a Acrobat viewer installed, you should download one first from Adobe: http://www.adobe.com/prodindex/acrobat/ Most of the files are between 100 and 500 Kb in size, so if you have a slow and/or expensive connection, the CDROM version of the proceedings maybe a better choice. This, as well as the print version, is still available, and can be ordered at: http://www.eamt.org/summitVIII/proceedingsorderform.html If you encounter any problems viewing the files, please let me know. -- Colin Brace Amsterdam http://www.lim.nl From malek.boualem@rd.francetelecom.com Thu Dec 6 16:06:23 2001 From: malek.boualem@rd.francetelecom.com (BOUALEM Malek FTRD/DMI/LAN) Date: Thu, 6 Dec 2001 17:06:23 +0100 Subject: [MT-List] Arabic/English MT Systems Message-ID: Sakhr, Egypt : http://www.sakhrsoft.com/ http://tarjim.ajeeb.com/ajeeb/ CIMOS, France : http://www.cimos.com/ http://www.cimos.com/TradNet.htm CRL, NMSU, US (research prototype) : http://crl.nmsu.edu http://crl.nmsu.edu/ATS/ I would be interested to get a copy of your survey. Malek Boualem France Telecom R&D > -----Original Message----- > From: Saad Al-Asali [mailto:al-asali@un.org] > Sent: Thursday, December 06, 2001 11:45 AM > To: mt-list@eamt.org > Subject: [MT-List] Arabic/English MT Systems > > > > Dear all, > I am preparing an updated survey on the above. So far I > managed to collect > a few names of companies but I couldn't find a single > interested academic > institute. > I would be very grateful if you could help me providing me > with names of > companies, academic institutes, scholars and other interested > individuals. > Regards, > > > > -- > For MT-List info, see http://www.eamt.org/mt-list.html > From malek.boualem@rd.francetelecom.com Fri Dec 7 12:57:56 2001 From: malek.boualem@rd.francetelecom.com (BOUALEM Malek FTRD/DMI/LAN) Date: Fri, 7 Dec 2001 13:57:56 +0100 Subject: [MT-List] English corpus to identify a translation system category Message-ID: Hi, I am looking for an English corpus that allows to identify whether a translation system is a machine translation system (i.e. integrating classic linguistic mechanisms) or if it is a translation memory based one. Typically, if some word groups, in different sentences, are always translated in the same way, while they should not, in this case, the system would probably be a translation memory one. Thanks in advance. Malek Boualem France Telecom R&D From cb@lim.nl Tue Dec 11 18:49:48 2001 From: cb@lim.nl (Colin Brace) Date: Tue, 11 Dec 2001 19:49:48 +0100 Subject: [MT-List] MT Summit: additional materials Message-ID: <20011211185927.NBAD1261.amsfep15-int.chello.nl@[62.108.30.75]> Hi everyone, A number of panel discussion presentations from the recent MT Summit are now online. See: http://www.eamt.org/summitVIII/papers.html or go directly to: http://www.eamt.org/summitVIII/panels.html They are in PowerPoint format. If you don't have a copy of PowerPoint, you can also view them using Sun's StarOffice, which you can download here: http://www.sun.com/staroffice/ If you encounter any problems viewing the files, please let me know. -- Colin Brace Amsterdam http://www.lim.nl From cb@lim.nl Fri Dec 14 12:23:37 2001 From: cb@lim.nl (Colin Brace) Date: Fri, 14 Dec 2001 13:23:37 +0100 Subject: [MT-List] Fwd: The Italian government using MT Message-ID: <20011214122504.MTRT1203.amsfep13-int.chello.nl@[62.108.30.75]> ------------------------------------------------------------------------- The following message is forwarded to you by Colin Brace ------------------------------------------------------------------------- >Message-Id: >Date: Fri, 14 Dec 2001 09:37:26 -0000 >From: "Harold Somers" >Subject: The Italian government using MT. My student Federico Gaspari passed this on to me: There is a hot debate (?) now in Italy because last week the government put on their official website the English version of the biographies of the ministers. Only the one of the Prime Minister (well-known Silvio Berlusconi) was in excellent English. All the others clearly were raw unedited MT output. Believe it or not, some people accuse Berlusconi that he did it on purpose to stand out and show how good he is... Some English-speaking web surfer happened to find the URL of the web page with these unedited MT texts into English (which according to the government was meant not to allow public access, but apparently many people saw it), and informed the newspaper "La Repubblica" (slightly left-wing oriented), which the following day published the news to mock the government (definitely right- wing). They immediately moved the URL off-line, but the damage had been done, since the web page had already been mirrored all over the place! The story has also been reported upon today by the on-line edition of The Guardian. Here's the URL: http://www.guardian.co.uk/elsewhere/journalist/story/0,7792,617612 ,00.html On Monday the Daily Mail (I think... Daily Mirror, maybe?) had a news flash on this on its front page, which I saw... Interesting material for next essay, maybe? Anyway, certainly good for a laugh, and after all instructive in a class to show typical howlers of unedited MT, perhaps! ----------------------------------------------------- -- End of forwarded message ----------------------------------------------------- -- Colin Brace Amsterdam http://www.lim.nl From jrl_ms@ga2.so-net.ne.jp Wed Dec 19 09:30:57 2001 From: jrl_ms@ga2.so-net.ne.jp (Jonathan Lewis) Date: Wed, 19 Dec 2001 09:30:57 +0000 Subject: [MT-List] How to pronounce Yngve? In-Reply-To: <20011214122504.MTRT1203.amsfep13-int.chello.nl@[62.108.30.75]> Message-ID: I wonder if anyone would be so kind as to tell me how to pronounce the name of the famous MT researcher V.H. Yngve? I am writing a chapter about MT in a Japanese book and have to render all Western names in katakana. Please respond off-list. Jonathan Lewis Tokyo Denki University From neilcoupland@btinternet.com Sat Dec 22 22:54:16 2001 From: neilcoupland@btinternet.com (Neil) Date: Sat, 22 Dec 2001 22:54:16 -0000 Subject: [MT-List] english to turkish Message-ID: <000001c18b3f$80c9e4c0$b0d17ad5@bob> This is a multi-part message in MIME format. ------=_NextPart_000_0007_01C18B3B.952BEDE0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable hello im looking for a site that will let me translate from english to = turkish, as i have a turkish girlfriend and she cannot read english so i = can only write to her and get others to read it for her, so er intimate = letters are out of the question. id appreciate any help you can give me thanks ------=_NextPart_000_0007_01C18B3B.952BEDE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

hello
im looking for a site that will let me = translate=20 from english to turkish, as i have a turkish girlfriend and she cannot = read=20 english so i can only write to her and get others to read it for her, so = er=20 intimate letters are out of the question.
id appreciate any help you can give = me
thanks
------=_NextPart_000_0007_01C18B3B.952BEDE0-- From hannouna@uruklink.net Mon Jan 15 21:17:11 2001 From: hannouna@uruklink.net (hannouna) Date: Tue, 16 Jan 2001 00:17:11 +0300 Subject: [MT-List] Enquiry Message-ID: <000101c07f39$52314d00$7128000a@brainsco> This is a multi-part message in MIME format. ------=_NextPart_000_0009_01C07F51.AB919FE0 Content-Type: text/plain; charset="windows-1256" Content-Transfer-Encoding: quoted-printable Dear Sirs ,=20 You have kindly accepted me as a member of the MT - List sending me a = password " gufucu " to be able to get to the MT-List . But , = unfortunately , when I went to the Internet Center today , I couldn't = contact you even though I had written my e-mail address and the password = you had sent me . So, I wonder here if there is any mistake ? Or could = you please send me other instructions and details regarding the way of = entering your Web and how can I receive e-mails from other participants = in the MT-List . I do appreciate .=20 Best regards .=20 Yours sincerely ,=20 Yasmin Hannouna .=20 ------=_NextPart_000_0009_01C07F51.AB919FE0 Content-Type: text/html; charset="windows-1256" Content-Transfer-Encoding: quoted-printable
Dear Sirs ,=20
You have kindly accepted me = as a member=20 of the MT - List sending me a password " gufucu " to be able to get to = the=20 MT-List . But , unfortunately , when I went to the Internet Center today = , I=20 couldn't contact you even though I had written my e-mail address and the = password you had sent me . So, I wonder here if there is any mistake ? = Or could=20 you please send me other instructions and details regarding the way of = entering=20 your Web and how can I receive e-mails from other participants in the = MT-List .=20 I do appreciate .
 
Best regards . =
Yours sincerely ,=20
Yasmin Hannouna .=20
------=_NextPart_000_0009_01C07F51.AB919FE0-- From hannouna@uruklink.net Thu Jan 18 14:19:45 2001 From: hannouna@uruklink.net (hannouna) Date: Thu, 18 Jan 2001 17:19:45 +0300 Subject: [MT-List] MT Evaluation History Message-ID: <000001c0815c$1654bb80$5a28000a@brainsco> This is a multi-part message in MIME format. ------=_NextPart_000_0009_01C08172.DA46C5E0 Content-Type: text/plain; charset="windows-1256" Content-Transfer-Encoding: quoted-printable Dear Friends of MT - List ,=20 I am a Ph.D student carrying out a research regarding the "Evaluation = of MT Systems that go from English into Arabic ". I am in need of = certain information relevant to the : History of MT Evaluation , = Theoretical background of MT Evaluation metrics , scales , and = measurements and other issues . If you are interested to give me = information on these aspects , please kindly send it to the following = e-mail address : hannouna@uruklink.net .You can also contact me on this = postal address : Miss.Yasmin H.Hannouna=20 AL-Karkh Post Office=20 P.O.Box . 28322 Baghdad - Iraq . =20 I highly appreciate your interest and cooperation , with respect .=20 Awaiting your kind reply .=20 Best regards .=20 Sincerely , Yasmin .=20 ------=_NextPart_000_0009_01C08172.DA46C5E0 Content-Type: text/html; charset="windows-1256" Content-Transfer-Encoding: quoted-printable
Dear Friends of MT - List , =
I am a Ph.D student carrying out a = research=20 regarding the  "Evaluation of MT Systems that go from English into = Arabic=20 ". I am in need of certain information relevant to the : History of MT=20 Evaluation , Theoretical background of MT Evaluation metrics , scales , = and=20 measurements and other issues . If you are interested to give me = information on=20 these aspects , please kindly send it to the following e-mail address : = hannouna@uruklink.net .You = can also=20 contact me on this postal address :
 
Miss.Yasmin H.Hannouna
AL-Karkh Post Office
P.O.Box . 28322
Baghdad - Iraq .  
 
I highly appreciate your interest and = cooperation ,=20 with respect .
 
Awaiting your kind reply . =
 
Best regards .
Sincerely ,
Yasmin .
------=_NextPart_000_0009_01C08172.DA46C5E0--