[Retros] Comments on Messigny tourney results

[pastmaker at aol.com]

Regret that I must disagree with my old friend here.  Imagine a theme tourney in which one of the entries not only satisfies the tourney theme conditions, but displays a new theme not exhibited before. The computer would not "see" that valuable feature, not having been programmed to notice it (because, by hypothesis, it was unknown).
 
The human judge would presumably be more likely to appreciate the innovation.
 
Regards to all,
Tom Volet
 
-----Original Message-----
From: afretro <afretro at yandex.ru>
To: retros at janko.at
Sent: Thu, 26 May 2005 08:40:00 +0400 (MSD)
Subject: [Retros] Comments on Messigny tourney results


Hi everybody!
The results of the tourney once again emphasize the need for strict term 
definitions. ⌠To avoid check■ √ what does it mean? Given the example and the 
traditional use of the term ⌠avoid/avoidance■ in chess composition, it was quite 
reasonable to interpret the thematic requirement as potential check prevention 
(i.e. without any check actually occurring; this is usually referred to as 
shielding) rather than defense against ⌠material■ checks. Furthermore, if ⌠a 
piece A intercepts a piece B,■ does it mean that piece A is active? Again, it 
was quite reasonable to expect that interception implied activity. If a pawn on 
its initial square is the only unit on an otherwise ⌠checking■ line, what did 
that pawn DO to avoid the check, to INTERCEPT the potentially checking unit? 
Nothing whatsoever.
Francois Labelle▓s analysis is quite remarkable. Thematic tourneys of this sort 
make it possible to determine the winner on the basis of objective criteria 
announced IN ADVANCE. For example, the judge might say that he would rank 
problems according to the overall number of theme cases (counting only those 
thematic pairs in which one piece intercepts the other MORE THAN ONCE), or to 
the number of thematic pairs, etc.; he would also have to indicate criteria 
ranking second in importance, to differentiate among problems presenting the 
same number of theme cases/pairs, etc. The big objection to this judgment 
pattern consists in that esthetic aspects play no role in it. On the other hand, 
esthetic criteria are so vague; a huge pool of them is available for the judge 
to extract the one justifying his ranking of the entries.
Can you imagine a SPORTS competition (in FIDE Albums, chess composition acts 
mostly as an art; in tourneys, as a sport) in which, e.g., runners cover the 
same distance one by one and see their result in seconds, and yet no one is sure 
of anything until the judge announces his decision based primarily on how much 
he liked each participant▓s running style rather than quickness?
Several retro tourneys with strictly specified ranking criteria have been 
conducted by Die Schwalbe. When the results were announced, none of the 
participants complained about award-winning problems lacking esthetic 
attraction.
Btw, Francois missed one ⌠semi-thematic■ point in the 2nd Prize: (1,bSg8,wQd1, 
allowing 000).
Andrey
_______________________________________________
Retros mailing list
Retros at janko.at
http://www.pairlist.net/mailman/listinfo/retros
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pairlist.net/pipermail/retros/attachments/20050526/b7e3cece/attachment.htm>