OpenChess

Matthias Rauterberg (entcom.eic.r@gmail.com) 2012-07-18 22:31:46.00

--------------------------------------------------------------------------------

Call for paper for a Special Issue of the journal Entertainment Computing (Elsevier)

Topic of the special issue: "Originality and creativity in designing software for games"

Software clones are identical or similar pieces of code, design or other artifacts produced during the development of a software system. Software clones are known to be closely related to various issues in the design of software games, especially with respect to originality and creativity, qualities that have to be evaluated when originality is claimed or an investigation of plagiarism takes place. For instance, in 2011 a computer chess program called Rybka was accused of plagiarism and deprived of several world championship titles it had won along the preceding five years.

This special issue intends to address the emerging problems in evaluating the originality and creativity of the design of software products in the field of entertainment computing. Topics of interest include, but are not limited to:

- Originality and creativity in software design
- Definition of software clones in the field of entertainment computing
- Discovering and measuring the similarity of clones
- Causes and effects of clones in computer games
- Software engineering methods and techniques for clone detection, analysis, and management in the field of entertainment computing
- Tools and systems for detecting and analyzing software clones
- Role of clones in entertainment computing
- Effect of clones in agonistic informatics
- Industrial experiences
- Licensing and plagiarism issues
- Clone-aware software design and development

Deadline for submissions November 30, 2012. You can upload your manuscript here http://ees.elsevier.com/entcom/

Guest editor of the special issue: Paolo Ciancarini, University of Bologna, Italy

There is another side of that coin.

Almost all of the innovation in computer chess comes from the sharing and learning of new, good ideas.
All of the best chess engines share an enormous number of common features like pvs search with zero windows, null move pruning, LMR reductions, Zobrist hashing, etc, etc, etc.

There is a fine line between cheating (stealing something someone else's code and calling it your own) and learning (using someone else's ideas and giving them credit for it).
This first thing, I think we can all agree, is very bad. The second thing, I think we can all agree, is very good. They both have the same result in some sense, with a new chess program that plays strong chess. But the journey's pathway is very important.

Something happened somewhere along the way that caused a large fraction of the chess programming crowd to become petty, selfish, and arbitrary. I think for the most part, it came from real infractions, where the legitimate programmers were being ripped off and they simply got sick and tired of it. But before lashing out, I think it would be a good idea to take in a deep breath and think about the supposed infidelity. Is it really something different than what everyone else is legitimately doing?

Anthony Cozzie quit chess programming because he felt that the winning program was nothing more than the biggest bag of tricks collected from all the existing programs. And there is a sad germ of truth in that. However, chess programming mirrors programming in general. The greatest advancements are made when we share information and ideas. There will always be room for new innovation.

IMO-YMMV

User923005 wrote: Anthony Cozzie quit chess programming because he felt that the winning program was nothing more than the biggest bag of tricks collected from all the existing programs.

True. And the one op top is the person that discovers the next innovation. History is pretty consistent. I had the pleasure to be (perhaps) the first one who implemented reductions successfully in Rebel 8 (1996), it gave me 2 plies extra and put me on top of the SSDF list with a lead of 65 elo points. The next innovation was null-move successfully implemented by Frans Morch in Fritz 5 (1998) and it put him on top for a couple of years. Thereafter came Shredder, the father of LMR and king SMK ruled with an iron fist. Thereafter came a new generation of programmers (Zappa, Fruit, Rybka) and they revolutionized computer chess catching up LMR and especially Vas innovated further and further. But..... history repeated itself, several catching up after a couple of years (Critter, Komodo, Stockfish) and now we are waiting for the next innovation (breakthrough) which will put the next programmer on top for a couple of years.

And there is a sad germ of truth in that. However, chess programming mirrors programming in general. The greatest advancements are made when we share information and ideas. There will always be room for new innovation. IMO-YMMV

Indeed.

Rebel wrote:

User923005 wrote: Anthony Cozzie quit chess programming because he felt that the winning program was nothing more than the biggest bag of tricks collected from all the existing programs.

True. And the one op top is the person that discovers the next innovation. History is pretty consistent. I had the pleasure to be (perhaps) the first one who implemented reductions successfully in Rebel 8 (1996), it gave me 2 plies extra and put me on top of the SSDF list with a lead of 65 elo points. The next innovation was null-move successfully implemented by Frans Morch in Fritz 5 (1998) and it put him on top for a couple of years. Thereafter came Shredder, the father of LMR and king SMK ruled with an iron fist. Thereafter came a new generation of programmers (Zappa, Fruit, Rybka) and they revolutionized computer chess catching up LMR and especially Vas innovated further and further. But..... history repeated itself, several catching up after a couple of years (Critter, Komodo, Stockfish) and now we are waiting for the next innovation (breakthrough) which will put the next programmer on top for a couple of years.

And there is a sad germ of truth in that. However, chess programming mirrors programming in general. The greatest advancements are made when we share information and ideas. There will always be room for new innovation. IMO-YMMV

Indeed.

???

Null-move was around Far longer than 1998. The 1992 version of Cray Blitz is available on the internet and it had fully-recursive null-move, but used R=1 until 1994 when we experimented with R=2 in parts of the tree far from the tips at a suggestion by Stanback. Beal's paper "Selective search without tears" was a 1980's idea. We were using null-move at the WCCC in 1989, as were others. Burt Wendroff pointed me to Beal's paper in fact (author of LaChex). This wasn't a 1998 idea at all, it was 10 years old by then.

hyatt wrote:

Rebel wrote:

User923005 wrote: Anthony Cozzie quit chess programming because he felt that the winning program was nothing more than the biggest bag of tricks collected from all the existing programs.

True. And the one op top is the person that discovers the next innovation. History is pretty consistent. I had the pleasure to be (perhaps) the first one who implemented reductions successfully in Rebel 8 (1996), it gave me 2 plies extra and put me on top of the SSDF list with a lead of 65 elo points. The next innovation was null-move successfully implemented by Frans Morch in Fritz 5 (1998) and it put him on top for a couple of years. Thereafter came Shredder, the father of LMR and king SMK ruled with an iron fist. Thereafter came a new generation of programmers (Zappa, Fruit, Rybka) and they revolutionized computer chess catching up LMR and especially Vas innovated further and further. But..... history repeated itself, several catching up after a couple of years (Critter, Komodo, Stockfish) and now we are waiting for the next innovation (breakthrough) which will put the next programmer on top for a couple of years.

And there is a sad germ of truth in that. However, chess programming mirrors programming in general. The greatest advancements are made when we share information and ideas. There will always be room for new innovation. IMO-YMMV

Indeed.

Null-move was around Far longer than 1998. The 1992 version of Cray Blitz is available on the internet and it had fully-recursive null-move, but used R=1 until 1994 when we experimented with R=2 in parts of the tree far from the tips at a suggestion by Stanback.

Obviously this subject is not about inventors but the ones who made a breakthrough from a dormant idea. I did not invent reductions but surely was the first one who made it working (popular) by adding an initial 100 elo points and more thereafter. And so it was with Nullmove, it was the number one talk in RGCC after Chrilly's article in the ICCA journal. And Frans was the first one who made Nullmove a success, standard today.

IOW, there is a difference between discovery and publication.
And there is also a difference between publication and popularization.

I don't get the distinction in this case (Null move). We were USING null-move in 1989 WCCC as were at least a couple of others (LaChex comes to mind first). The deep Thought guys also reported on their results using it, and was the place where I first saw the TT avoid-null-move trick mentioned as well. This was not a 1995 thing. It was already well-known by then.

User923005 wrote:IOW, there is a difference between discovery and publication.
And there is also a difference between publication and popularization.

A good example would be Crafty. While Bob did not invent bit-boards he published about it and it became popular because many foresaw a 64-bit PC to arrive soon, now that 64-bit is the norm bit-boards are the norm. Another (but small) breakthrough.

hyatt wrote:I don't get the distinction in this case (Null move). We were USING null-move in 1989 WCCC as were at least a couple of others (LaChex comes to mind first). The deep Thought guys also reported on their results using it, and was the place where I first saw the TT avoid-null-move trick mentioned as well. This was not a 1995 thing. It was already well-known by then.

Yes, first published 1975 in English by the Kaissa authors:
Georgy Adelson-Velsky, Vladimir Arlazarov and Mikhail Donskoy (1975). Some Methods of Controlling the Tree Search in Chess Programs. Artificial Ingelligence, Vol. 6, No. 4, pp. 361-371. ISSN 0004-3702. Reprinted (1988) in Computer Chess Compendium

2. The Order of Move Considerations:
A less trivial idea was that sometimes an extension of the game tree by introducing of dummy move can lead to a reduction of the search tree. In positions with material advantage (with respect to limits) it was permitted to try a so-called "blank" move in which ones own pieces are not moved. Certainly in positions with "Zugzwang" (the side to move must weaken his position) this may lead to errors.

On Lachex, it appears in the WCCC 1989 booklet under program descriptions with null move mentioned (pp 15), but not in the table of participants and Lachex did not play in Edmonton (it played 1986 in Cologne and 1992 Madrid).
http://archive.computerhistory.org/proj ... 028.sm.pdf

Didn't Cray Blitz already use null move in 1983, according to your 1996 rgcc post? I mean that is no contradiction CB used null move in 1989 as well, but one then assumes 1989 was the first occurrence using null move.
http://groups.google.com/group/rec.game ... dc1ec60967

Null Move was used in 1983, because I used it in the program that won the world computer chess championship that year. They aren't particularly new ideas at all ...

Rebel wrote:

User923005 wrote:IOW, there is a difference between discovery and publication.
And there is also a difference between publication and popularization.

A good example would be Crafty. While Bob did not invent bit-boards he published about it and it became popular because many foresaw a 64-bit PC to arrive soon, now that 64-bit is the norm bit-boards are the norm. Another (but small) breakthrough.

Credit has to go to Slate, IMHO, although Donskoy apparently developed them at about the same time (Kaissa)...