MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

16:04 UTC
ISC 2024
 
  Forum*
 
 
 
 

Username:

Password:

Remember me

 
Forgot your
password?
Click here!
SIGN IN
to create your account if you don't already have one.
CHESS
SOLVING

Tournaments
Rating lists
1-Jan-2024

B P C F





 
 
MatPlus.Net Forum General FIDE Album election- change the rules
 
You can only view this page!
Page: [Previous] [Next] 1 2 3 4
(61) Posted by Harri Hurme [Friday, Aug 25, 2017 17:05]; edited by Harri Hurme [17-08-25]

Hello!

We discuss about Album selection. Actually the whole chess problem judging is the weakest link in our art. This is clear looking at the given points not only in Album selection but also in WCCT. There are so many oddly different points for the same problems that one easily can conclude that something is badly wrong.

There are many sports in which the judgement is based on opinion. The figure skating used to be very subjective and judges gave high placements for own country competitors. But now they have changed the rules. In figure skating there are different judges for artistic effects and for technical merits, who use video pictures to see how many circles happened in the air. The end result is that now they have very reliable rating systems and no one is complaining judging any more. Another sport which is well judged is swim jumping. One judge gives at maximum 10 points, out of which only one point is "artistic", the remaining 3*3 points are based on objective measured qualitees.

I have been thinking how to improve chess problem judging. Vaux Wilson proposal of MOE (Method Of Evaluation) was interesting attempt in the past. He tried to make judging measurable in absolute terms. But his proposal failed in one important point, it did not take one of the problem chess basic principles in count at all. That missing principle was "thematical unity". Including that it should be possible to develope a judging algorithm in future. No easy task but the goal is so important that even a light progression is welcome.

As many has indicated the title requirements make the Album selection so hot. One approach would be that the titles and FA points are separated. WFCC could give composing titles based on achieved prizes in various composing tournaments.

Personally I feel its awfully difficult to get a problem to FA. My main hobby is composing. It took 50 composing years from me to come The Fide master. I have several first prize problems in international tourneys which never went to Album. One example is 1.pr in Sola memorial which shows both Finnish Plachutta and Finnish Nowotny. As far as I know no other single problem has both. It got very good points from two judges, but one gave only 0.5 distroing its way to album. I have many other cases which did not make in FA even if I have some lesser ones included.

I propose that in the title application 10 first prizes not included in FA are counted as one album point. This change would make more justice, no harm.

One proposal could be that a composer gets one problem in with 7 points, one with 7.5 and the rest needs 8 points. This makes justice to lesser composers who often are ignored.

I propose also that in the coming WCCT11 the theme examples should have judging points associated with them. That helps in composing, in selection in countries and in judging giving a consistent scale.

But really: do some (good) changes!

Harri Hurme
 
   
(Read Only)pid=15932
(62) Posted by Dupont Nicolas [Friday, Aug 25, 2017 17:58]

It seems there is another drawback concerning Neal's proposal "Yes Maybe Not" - what would be the final FA mark of a given entry? One can imagine "best problems" to be quoted 3 stars (= 3 yes), but it refers in the current system to marks between 9 (3+3+3) and 12, hence is rather imprecise. In particular there is no more difference in Neal's proposal between a very good 9-marked problem and a truly outstanding and novative 12-marked problem.
 
   
(Read Only)pid=15933
(63) Posted by shankar ram [Saturday, Aug 26, 2017 21:41]; edited by shankar ram [17-08-27]

While I have nothing to add as regards the selection criteria discussed so far, I have 2 suggestions about using current technology:
1. WFCC can set up a server which could be used as a single point window for problem submission, review and forwarding to the judges, entering of points by the judges and final review and forwarding for publication.This would perhaps reduce all the emailing currently being done and speed up the whole process.
2. Album in kindle format: I had a brief exchange of mails with Peter about this. He expressed concern about copy protection. Currently, it's ridiculously easy to break the protection of a supposedly secure DRMed kindle book. But maybe some of the computer experts in our fraternity can devise a stronger method. I dream of the day when I can read the album on my kindle, with Dmitri's py2web animation incorporated!
 
   
(Read Only)pid=15935
(64) Posted by Rajendiran Raju [Sunday, Aug 27, 2017 04:41]

I second the same !

travelling with technology is need of the hour and inevitable.
 
 
(Read Only)pid=15938
(65) Posted by Harry Fougiaxis [Sunday, Aug 27, 2017 10:36]

 QUOTE 
1. WFCC can set up a server which could be used as a single point window for problem submission, review and forwarding to the judges, entering of points by the judges and final review and forwarding for publication.This would perhaps reduce all the emailing currently being done and speed up the whole process.

We have already tried towards this direction, see http://www.wfcc.ch/wp-content/uploads/2013/05/minutes2012-Kobe1.pdf

Only two (!!) problemists used the alternative and submitted their entries to the server, so the project was abandoned in the next year.
 
   
(Read Only)pid=15940
(66) Posted by Rajendiran Raju [Sunday, Aug 27, 2017 11:40]

yes , sadly most of our peoples still not yet migrated to new technologies !
 
   
(Read Only)pid=15942
(67) Posted by Mario Parrinello [Monday, Aug 28, 2017 14:54]; edited by Mario Parrinello [17-08-28]

I really appreciated the post of GM Petko Petkov; thanks, Petko, for you eminent opinion!

Let me add some my personal thoughts.
As all we know, the current evaluation of entries for selecting the problems for FIDE-Album requires that judges express their evaluation in the form of a score.
Therefore it implies that the value of a problem is expressed by numbers (for example total score of 3, 2,5 or 4...); on the other hand this method of evaluation based on scores implies that a particular score can reflect the intrinsic value of a problem.
But how could we fairly evaluate the intrinsic merits of an artwork (yes, a problem can be surely cosidered an artwork) just and solely on the basis of a score? It is possibile of course, but a score does not entirely reflect the value of a problem, I think.
It is something like to evaluate the beauty of a painting on the basis of a score.
This is what I think is the 'original sin': it is impossibile, to some extent, to fairly evaluate an artwork on this basis.
But it is surely possibile to tell whether a relevant problem is worthy to enter the album, just thanks to the current method. The current method is not perfect of course (nothing is perfect on this planet) and it is the 'lesser evil', BUT SIMPLIFYING THE SCORING SYSTEM WOULD BE A BIG MISTAKE!
How can fairly evaluate the construction? A construction could well be heavy, light, adequate for the idea etc.
How can fairly evaluate the originality? The idea could well be more or less original.
How can fairly evaluate the purity?
How can fairly evaluate the armony?
How can fairly evaluate the strategical depth? There are various degree of depth...
How can fairly evaluate the solving difficulty? There are several degrees of solving difficulty of a relevant problem.
And so on.
Simplifying the scoring system would limit our ability to fairly evaluate all the aspects of a problem and would be surely a big mistake.
Therefore, though the current method is not perfect, I think it is really the best one in our hands.
 
   
(Read Only)pid=15951
(68) Posted by Michel Caillaud [Saturday, Sep 2, 2017 01:25]

As a coincidence, on my way to Marianka meeting, I got separate offers to act as a judge for FIDE Album and for WCCI.
I already wrote on this forum some time ago that I will not act as a judge in these competitions under the current system of mixing them (Kobe decisions), and I repeat it here.

As a FIDE Album judge, I don't want choices imposed on me (what if some composition imposed by WCCI I find less good than similar entry in the set of those submitted to FIDE Album?) .
Symetrically, as I don't want to do to others what I don't want others to do to me, I don't want my work of designating a World Champion deviated for imposing problems in FIDE Album.

I accept of course decisions of our democratic institutions.
I have full respect for the work of judges in these competitions. Marianka's lecture by Juraj Lörinc allowed a glimpse on his titanesque enterprise (also, I did such work in the past...).
I do not wish to impose any view to anyone.
Simply, I will not participate in doing things that go against my personal ethics.
I do not wish to discuss this : I had more than enough of these never ending discussions when I was a delegate.
I just will add two points that seem obvious to me. Maybe someone already mention them. I confess that I didn't read contributions in detail as all this is tiring me considerably.
 
   
(Read Only)pid=15970
(69) Posted by Michel Caillaud [Saturday, Sep 2, 2017 01:28]

Under the current system, a very simple way to lighten the work of FIDE Album judges would be to exclude every WCCI entry from their examination, those making 8,0 that would be imposed as well as those not making 8,0 that would be eliminated. If WCCI's judges have power of selecting problems, why shouldn't they have power of eliminating them?
The current system is unfair to composers whose output is mainly co-productions (for example, Michal Hlinka who mainly composes in collaboration with Lubos Kekely...) and thus have less chances to get their works in FIDE Album than those with two chances to get the selection.
Would the set of judges for WCCI be more « generous » than the one for FIDE Album, the unfairness would still be accentuated..
 
   
(Read Only)pid=15971
(70) Posted by Michel Caillaud [Saturday, Sep 2, 2017 01:37]

From my point of view, FIDE Album is for compositions (as I understand, so it was at the time of its creation), WCCI is for composers, thus the notation attached to a problem has not to be automatically the same in both competitions.
2 obvious examples :
-H36 in Album 2010-12 : I would give 4 for FIDE Album for a historically significant problem, 0 for WCCI as I consider WCCI is for rewarding skilfulness of human composer, not of computer programmer (that is my view ; some may have other views...).
-H9 in Album 2004-06 : I gave 3,5 for a historically significant problem and would have given much less for WCCI as important part of creative work was done by a predecessor (as a fact, this problem was not submitted to WCCI).
(BTW, H9 was submitted (and presumably published) as : Aleksej Troizky, corrected version by Dmitrij Baibikov. As my memory is not perfect, I cannot remember why this was changed in FIDE Album publication, and say for sure if the judges were informed or not of this « alteration »...)
 
   
(Read Only)pid=15972
(71) Posted by Valery Gurov [Tuesday, Sep 5, 2017 14:25]

I absolutely agree with Michel's opinion: WCCI and Album is essentially different competitions!
 
   
(Read Only)pid=15984
(72) Posted by Vlaicu Crisan [Wednesday, Sep 6, 2017 13:26]

The final vote in Kobe on the automatic qualification for the album of WCCI compositions with at least 8 points was: 11 for, 9 against and 4 abstentions.

The main advantage of the proposal seems to be the gain in time for granting titles. I think there are better ways to achieve the same outcome.
Michel's approach seems very reasonable and consistent. It took me quite some time to properly understand the flaws.

WCCT, WCCI and FIDE Album are essentially different competitions with different formats and different purposes.
The marks received in one competition should not be automatically applied in another.

I admit that in 2007-09 I gave some marks both in WCCI and FA without having a second thought of my role.
But that's why Michel is an international judge and can point out some subtleties...

I think WCCI should also provide an annex with its own guidelines of what each mark represents (similar with those from WCCT and FIDE Album).
Such instructions will make it even more clear why the differences in marking the same composition might occur.

Thank you, Michel, for spotting this delicate issue and convincingly articulating it!
 
   
(Read Only)pid=15991
(73) Posted by Frank Richter [Wednesday, Sep 6, 2017 23:24]

Hm, generally I do not like the approach to rate problems different (relatively) according to a context given by a competition.
A chess problem should speak for itself.
A masterpiece is a masterpiece.
An improvement that turns an older problem into a masterpiece is also a masterpiece.
It's the same issue with giving prizes.
Is a "prize" a relative or absolute category?
I'd prefer to interprete "prize" as an absolute evaluation.
 
   
(Read Only)pid=15992
(74) Posted by Nikola Predrag [Thursday, Sep 7, 2017 00:08]

"Absolute evaluation" sounds terrifying when we know how different the evaluations actually are.
It sounds like invoking a spiritless bureaucratic perception of the original creations.
 
   
(Read Only)pid=15993
(75) Posted by Frank Richter [Thursday, Sep 7, 2017 10:51]; edited by Frank Richter [17-09-07]

We may look for a better term together, if you like.

Edit: Something like "context-independent" etc.
 
   
(Read Only)pid=15994
(76) Posted by Vlaicu Crisan [Friday, Sep 8, 2017 11:00]

@Frank Richter: as Michel Caillaud suggests, I also think the context is quite very important when performing the evaluation.

And even within the same context, the evaluations may differ quite at a large extent. To give you just two examples:
1. In a previous FIDE Album, at the retros section, some entries received the following marks from the judges: 4 + 0 + 4.
2. In the WCCT-10, at the fairies section, one entry received 4.0 from one judge and 0.8 from another. There is another composition in the same section receiving the marks 3.8 and 0.6.

So let's simply accept that an "absolute evaluation" is not [always] possible.
 
   
(Read Only)pid=15997
(77) Posted by Frank Richter [Friday, Sep 8, 2017 12:16]

Yes, we will always have different evaluations by different people.
But I'm not sure about different evaluations by the same person.

Regarding the main topic - I like the current 0-4 points system with three judges. Surely it isn't perfect, but it has proved effective.
The auto-selection of WCCI problems >= 8 points should be discussed. My idea: Album judges may veto. If two judges veto, then a problem is rejected.
The 7.5 point rule should be abolished. We should promote quality, but not quantity.
 
   
(Read Only)pid=15998
(78) Posted by Yochanan Afek [Friday, Sep 8, 2017 18:37]

Frank,7.5 points often means majority of the jury wants to see the problem in the album and decreases the risk of minority decision. Quality and justice,then!
 
   
(Read Only)pid=16000
(79) Posted by Frank Richter [Friday, Sep 8, 2017 19:24]; edited by Frank Richter [17-09-08]

Well, and what about problems with 7.5 points in a "normal" section with selection starting from 8 points?

By the way, as posted in #56 by Harry there are only very few problems with the combination 3+3+1.5 points. There is no information about 3.5+2.5+1.5 or similar combinations, but I think, this happens even rarer.
 
   
(Read Only)pid=16001
(80) Posted by Marko Klasinc [Saturday, Sep 9, 2017 02:36]

@Vlaicu Crisan post no.76
 QUOTE 
1. In a previous FIDE Album, at the retros section, some entries received the following marks from the judges: 4 + 0 + 4.

That was completely different situation, and it was about one problem only in AF 1980-82. In the process of judging judges were asked to reconsider our marks. My mark was the lowest one. I was willing to raise it but I found one similar problem in the meantime and decided that it was an anticipation and therefore suggested to exclude this particular problem (giving zero points to it). Unfortunately the re-examination phase stopped at that point for me. Other two judges were obviously informed about it and they also obviously didn't agree with me and they later raised their marks to four. I was not contacted anymore and was very much surprised noticing these 4-4-0 marks in the published Album. This was the most unusual set of marks in the history of Album FIDE. But it was time before internet and e-mails and all correspondence went via letters. I am sure that it couldn't happen today. And believe me - it left a very bad feeling to me since I felt a sort of punished for my opinion about the anticipation instead of being included in a further discussion to find some good compromise.

And just a little comment to the whole debate within this thread. I am sure that no system can prevent bad judging from judges who are not capable (or they do it deliberately) to recognize the quality of problems of genres which they don't prefer.
 
 
(Read Only)pid=16002

No more posts
Page: [Previous] [Next] 1 2 3 4

MatPlus.Net Forum General FIDE Album election- change the rules