Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Redacting the ECM suite

Author: Howard Exner

Date: 08:34:38 05/02/98

Go up one level in this thread


On May 01, 1998 at 14:38:14, Don Dailey wrote:

>Hi everyone,
>
>Howard asked me to finish the process of redacting the ECM set.  Since
>I now have plenty  of spare time I am  willing to do this, if everyone
>agree's.

I emailed this request of Don. This is the email:*

*I was going to  post this suggestion on  CCC  but thought it might  be
*best to email you  first. Now that  the  bulk of the broken  positions
*have been tossed  out the next  phase  seems to  be how  to handle the
*removal  of the  easy positions.   I felt   competent in dealing  with
*finding the  deletition candidates but  not  too comfortable  with the
*handling of how to remove the easy ones.  Basically I'm wondering if I
*could pass the baton back to you. I think you would do  a great job in
*co-ordinating this. I was going to post something like, "would someone
*like   to take over    the  next phase  of    removing the easy   ones
*...blah,blah,blah.  Thought I'd run it by you first.

*If  you  decide to do  this  I could then  simply post  in  CCC that I
*requested it or even just  post this email on  the site. I will  still
*remain interested in the suite in  terms of why or  why not a key move
*is chosen.

>
>I suggest that we leave the easy  problems in the set.  The consensous
>seems to be the same.  There is no harm in  having a few easy ones and
>some of us  have expressed some  concern that  the elimination process
>could be subject to problems.  I agree with this viewpoint.

Or there is the possibility of keeping ECM98 and have a subset of this
as well. I still recall the seeds of this project from Bob and Ernst
and there desire to remove the easy ones common to all programs. But
I guess each person can simply keep their own subset of ECM98 rather
than create a sort of "universal" one.
>
>The main thing to address  is the correctness of  the set as it stands
>now.  If anyone finds or suspects errors,  let's post this and open it
>up for discussion.  To   summarize  our criteria for  inclusion,  each
>position should have  the following characteristics (please correct me
>if I'm wrong on what was agreed to):
>
>A) Solution should be correct.  (obviously)
>
>B) Solution  should  not be   an  obvious move   that any program  has
>   some  chance  of  playing even if  it doesn't really understand the
>   idea.

On this point B I recognized some key moves as being played to achieve a
draw
rather than seeing deeper that they are winning. Botvinnik's Ba3 against
Capablanca comes to mind. Would this be the kind of example you are
referring
to?
>
>C) All solutions  should be given.  It's ok to have multiple solutions
>   if  they  all  involve  a winning plan.   Conversely, non-solutions
>   should change the game theoretic outcome with perfect play.
>
>Later on, if anyone is interested, we can sort  the set by difficulty.
>This can be  approximated by having n  programs sort them individually
>and then   combining  the results in some   fashion  such  as average,
>median, worst case ranking or some other method.   We can discuss this
>later if anyone feels it is worth persuing.
>
>- Don



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.