Topic Rybka Support & Discussion / Rybka Support / My 2 cents on the Rybka - Fruit GPL violation (if any)
![[us]](/mwf/v12/flags/us.png "United States")
How about cases in which ~some~ code for a ~very different~ commercial program was taken? What kinds of cases do we have for this? That is what I was referring to--obviously, my statement itself doesn't hold at all without the context.
I don't follow the 'very different commercial application' part. The GPL cases I have seen, as well as the copyright cases, all involved a direct competitor taking a part of an existing program when developing a program with a similar target audience. So the "very different" doesn't apply as I see it. Perhaps you meant "much better"? Which doesn't pass the stink-test at all.
What case or cases come to your mind that are most analogous to the present case?
Just a few:
1.. lliad France. They included GPL linux code in a DSL box they sold. However, when pressed, they refused to provide the source code as required by the GPL.
2.. Dlink had a similar problem with their network attached storage device that used the linux kernel. Note that this case was held in Germany and they were found guilty of copyright infringement regarding the GPL code. If you read the complete report, it seems that they sounded just like some here (Dlink) in that they said "we do not believe the GPL is legally binding." The court disagreed, however, and found them guilty.
3.. A similar case in Austria dealing with their health-card system. No source released, GPL code used. Found guilty.
4.. One for Targus notebooks, same idea.
Here is a nice quote from the founder of the FSF on this issue:
Mr Welte continues: "The Free Software community is very happy to see that
notebook vendors actually pre-install Linux onto their devices because they're
dissatisfied with the bootup time of the popular Microsoft(R) Windows(TM)
operating system. However, Free Software is copyrighted material, much like
any other software. Redistribution can only happen according to this license."
Note that keyword "copyrighted". As I have repeatedly said, and a "few" repeatedly deny, GPL and copyright go hand-in-hand...
And the courts agree, whether the posters here do or not...
If you go to the GPL violations page, there are a bunch of links you can follow to see what is currently being litigated and what has been litigated in the past...
Some are about the linux kernel. Some are about a media player. A compiler. Etc...
1.. lliad France. They included GPL linux code in a DSL box they sold. However, when pressed, they refused to provide the source code as required by the GPL.
2.. Dlink had a similar problem with their network attached storage device that used the linux kernel. Note that this case was held in Germany and they were found guilty of copyright infringement regarding the GPL code. If you read the complete report, it seems that they sounded just like some here (Dlink) in that they said "we do not believe the GPL is legally binding." The court disagreed, however, and found them guilty.
3.. A similar case in Austria dealing with their health-card system. No source released, GPL code used. Found guilty.
4.. One for Targus notebooks, same idea.
Here is a nice quote from the founder of the FSF on this issue:
Mr Welte continues: "The Free Software community is very happy to see that
notebook vendors actually pre-install Linux onto their devices because they're
dissatisfied with the bootup time of the popular Microsoft(R) Windows(TM)
operating system. However, Free Software is copyrighted material, much like
any other software. Redistribution can only happen according to this license."
Note that keyword "copyrighted". As I have repeatedly said, and a "few" repeatedly deny, GPL and copyright go hand-in-hand...
And the courts agree, whether the posters here do or not...
If you go to the GPL violations page, there are a bunch of links you can follow to see what is currently being litigated and what has been litigated in the past...
Some are about the linux kernel. Some are about a media player. A compiler. Etc...
Thanks; (1) and (2) certainly sound like pretty analogous examples to the present case.
I did not read the entire transcript stuff completely. The main focus is to enforce the GPL. Although there were some mentions about money in a couple of them...
Another question: can you state clearly which pages of which ICGA released document contain what, in your view, are the most serious GPL violations? I particularly note pages 17 and 20, as well as 14 and 15 of the document Rybka_Fruit_Mar11, but these seem to me to respectively show (a) little if any actually copied code, and (b) copied code from a common function that isn't really so much a part of the chess-playing entity. Neither of these seems damning in the least respect, and yet the first of these is often mentioned. Similarly, the board and piece comparisons in the first half of the document don't seem to show any actual copying that I can see. Plagiarism by the strictest of academic standards, perhaps--but copying, no.
I think that perhaps if this can be addressed and agreed upon by the various sides, there won't be much more in the way of argument.
I think that perhaps if this can be addressed and agreed upon by the various sides, there won't be much more in the way of argument.
First we need to decide what we are looking for. Remember, the ICGA investigation was to answer the single question "Did Rybka violate ICGA tournament rules because it contained parts of other programs?"
Are you talking about that, or the somewhat ancillary issue of the GPL and copyright infringement?
Are you talking about that, or the somewhat ancillary issue of the GPL and copyright infringement?
I guess that I'm wondering about both. While I accept that Vas probably broke the tournament rules in 2006 and 2007 based on the various discussions, I still don't see that directly from the data, so it might be best to start with that. Of course, if you can knock out the GPL and copyright issue in one fell swoop, that would also be great. :-)
OK, next question. Give me a link to what you are looking at, so that I don't look at one thing and you another. I have access to all the ICGA discussions and such and want to make sure that I am referring to what you are actually looking at...
The simplest to understand part is Mark Watkin's analysis of Rybka 1.0 beta vs Fruit evaluations. He takes them piece by piece, and describes what each does, and on rare occasions, points out a difference. This is not code that compares side-by-side because of the bitboard/mailbox differences. But when you take each piece, and notice that both programs evaluate everything identically, with only a small difference here and there, you begin to get a very uneasy feeling. For example, I do mobility in Crafty, but I don't worry about "safe squares". Nor do I do "forward mobility bonuses." But I do "center mobility" where a piece gets a bigger bonus for attacking central squares than squares closer to the edges.
One of those examples doesn't raise a red flag, but so many of them do. Surely rybka and fruit would have lots of differences somewhere in the evaluation? Comparing Crafty shows something completely different, for example. Mark goes on to analyze other things common between fruit and rybka 1.0 beta. By the time you reach the end, you are left shaking your head...
If you don't see Mark's report where you are looking, let me know and I can post a PDF here.
The simplest to understand part is Mark Watkin's analysis of Rybka 1.0 beta vs Fruit evaluations. He takes them piece by piece, and describes what each does, and on rare occasions, points out a difference. This is not code that compares side-by-side because of the bitboard/mailbox differences. But when you take each piece, and notice that both programs evaluate everything identically, with only a small difference here and there, you begin to get a very uneasy feeling. For example, I do mobility in Crafty, but I don't worry about "safe squares". Nor do I do "forward mobility bonuses." But I do "center mobility" where a piece gets a bigger bonus for attacking central squares than squares closer to the edges.
One of those examples doesn't raise a red flag, but so many of them do. Surely rybka and fruit would have lots of differences somewhere in the evaluation? Comparing Crafty shows something completely different, for example. Mark goes on to analyze other things common between fruit and rybka 1.0 beta. By the time you reach the end, you are left shaking your head...
If you don't see Mark's report where you are looking, let me know and I can post a PDF here.
I'm looking at the set of files given in the link at the bottom of the long "announcement" on the ICGA page here.
I think that while one can get "uneasy feelings" in such situations of evaluation similarities, when there isn't direct evidence of code copying in that situation, we don't really have anything on which to proceed if the programs play significantly differently (about 150 elo in the case of Rybka 2.3.2a).
I think that while one can get "uneasy feelings" in such situations of evaluation similarities, when there isn't direct evidence of code copying in that situation, we don't really have anything on which to proceed if the programs play significantly differently (about 150 elo in the case of Rybka 2.3.2a).
The evaluation is but one part of the engine. The search is more important, as it has to traverse the tree in such a way as to give pertinent positions to the evaluation code.
I can not access the link you provided, I'll talk to Mark / David, as they forgot about us linux users. I might be able to find a .rar program, but it is not a standard Fedora utility...
If it has a report by Mark Watkins that goes through the eval, and talks about rooks, knights, bishops, etc, and piece/square tables, and then moves on to parts of the move ordering and such, that is the one I am talking about...
I can not access the link you provided, I'll talk to Mark / David, as they forgot about us linux users. I might be able to find a .rar program, but it is not a standard Fedora utility...
If it has a report by Mark Watkins that goes through the eval, and talks about rooks, knights, bishops, etc, and piece/square tables, and then moves on to parts of the move ordering and such, that is the one I am talking about...
Yes, I think that we are talking about the same report.
OK, that is a good place to start. If you want to pick some point in the report to ask questions, fire away and we can discuss whatever is in there. Clearly, the assembly-to-C comparisons show exact copying and are the easiest to understand. The eval is less clear because of the bitboard/mailbox issues. The piece/square tables and scoring values are interesting, and if you want to start there, I can snip and post the same arrays from Crafty to see how they stack up, since I have not done that comparison myself... would be interesting to see.
I have been referring to the 29-page report called RYBKA_FRUIT_Mar11 and begins with "A comparison of Rybka 1.0 Beta and Fruit 2.1". Presumably, this Rybka version would have the most similarities, and yet, I am unable to find anything of obvious copying note except for the UCI parser stuff.
Did you look at all the piece/square data? That's just for starters. We can add Crafty to the comparison (here) if you want, just to see how unlikely it is for two different programs to have a dozen arrays that are almost perfectly identical, differing only where integer truncation might have hit.
I did look at those data, and these clearly show that Vas looked at the Fruit code. He then used his multiplication to make the code itself different, and changed some values slightly as he saw fit. I don't see anything wrong with this in the commercial realm. In the academic realm, obviously this is a no-no, but this seems quite fine commercially.
Here's the simple explanation of copyright law for this kind of data. You can't copyright a number, any more than you can copyright a single chess game, because any game can be produced by a simple enumeration process (might take a _long_ time, but that's irrelevant.). But once you expend effort on producing a collection of games, such as Fischer's 100 greatest games, you can copyright that collection because of the effort expended in producing the thing. This same idea applies to tables of numbers that could be generated by a completely enumerative process. But when someone spends a lot of effort building 12 tables of 64 numbers each, where each and every number contributes to how the program plays chess, you can't copy those numbers. Say an array of 12 x 64 entries for the 6 pieces x 2 sides x 64 squares. And you access v[x][y] where x is the piece, and y is the square. But you like the numbers so you write a function f(x,y) where f(x,y) == v[x][y]. But you didn't copy the tables of numbers. But you compute them so that you get the same values as if you did copy the array. That is _not_ acceptable and is a copyright violation. And it is a quite common methodology used in plagiarism attempts on student programs. The effort is in selecting those 12 x 64 numbers, with a lot of testing and trial and error. It doesn't matter whether another person uses different array names, or even different array sizes, or a function. If you still get 12 x 64 numbers, and they are the same, that's bad. And that is just one piece of evidence. There are other ways to attempt to hide this, such as multiplying everything by some constant, and then dividing by that constant after using the value. Etc..
You can also look beyond the eval code to the search code. The time code is interesting because things are done the same way, including the inexplicable 0.0 comparison that nobody in their right mind would do in the Rybka context. And then the next two pages of code. It is all too similar. One single piece could be passed off as coincidence, perhaps, but not all of that. And then when you take it in light of his frequent claim that "all versions" of Rybka have always been original, and factor in the minor problem that 1.6.1, which he sent to several testers to have it rated, is about a big a copy job as I have ever witnessed...
You can also look beyond the eval code to the search code. The time code is interesting because things are done the same way, including the inexplicable 0.0 comparison that nobody in their right mind would do in the Rybka context. And then the next two pages of code. It is all too similar. One single piece could be passed off as coincidence, perhaps, but not all of that. And then when you take it in light of his frequent claim that "all versions" of Rybka have always been original, and factor in the minor problem that 1.6.1, which he sent to several testers to have it rated, is about a big a copy job as I have ever witnessed...
This sounds good, but it seems to fall into the realm of things that cannot be copyrighted ("systems" and "formulas") as noted by Alan below.
You can't copyright a formula, I agree. But Fruit didn't have a formula. It has a group of arrays (piece/square tables), one for each piece type/color, with each having 64 values for the 64 squares. You can't create a formula that reproduces my group of arrays, precisely, and claim that is different. Nor can you re-shape the arrays and other such things that are often tried. Any more than you can change the language. The piece/square tables are, in a single instance, not so significant. But when you reproduce the data in _all_ of them, plus pawn scoring arrays for things like passers, isolated/weak pawns, candidate passers, and such. The problem is compounded and what happened becomes obvious. The other problem is that if you duplicate the arrays in some form, you _must_ duplicate the code that uses them for the data to be useful, further getting into copying of code.
As I said, when you add it _all_ up, it is quite convincing to anyone willing to look. If you take just one piece, not so much. That's why most court cases don't see just a single piece of evidence shown. The more you show, the harder it is to ignore. And there is a _lot_ to see here.
As I said, when you add it _all_ up, it is quite convincing to anyone willing to look. If you take just one piece, not so much. That's why most court cases don't see just a single piece of evidence shown. The more you show, the harder it is to ignore. And there is a _lot_ to see here.
Wouldn't the group of arrays be categorized as a "system", which also cannot be copyrighted?
No. We are not talking just about the structure. We are talking about the structure and contents. It is but one piece of evidence in the report. But when taken along witheverything else, it shows a clear pattern of copying.
If you copy code, and keep the piece/square tables, the rank/file tables, the mobility tables and such, all intact. And all you do is modify the code as necessary to work with bitboards as opposed to mailbox, it tells a compelling story to those that both look at it carefully, _and_ understand the issues involved in writing a chess engine... This is sort of like a counter-balanced scale. You keep throwing stuff on one side, and eventually that side drops lower because of the weight. Any single piece doesn't mean a lot, but all the pieces in whole say a lot together.
If you copy code, and keep the piece/square tables, the rank/file tables, the mobility tables and such, all intact. And all you do is modify the code as necessary to work with bitboards as opposed to mailbox, it tells a compelling story to those that both look at it carefully, _and_ understand the issues involved in writing a chess engine... This is sort of like a counter-balanced scale. You keep throwing stuff on one side, and eventually that side drops lower because of the weight. Any single piece doesn't mean a lot, but all the pieces in whole say a lot together.
I'm not sure how well a "clear pattern" would hold up in a copyright case. I guess that whether or not the items taken can be said to belong to one of the sets of things that aren't subject to copyright might simply have to be determined by a court case. Changing the structure and contents into something else that basically does the same thing is one of the most common ways that companies get around copyright violations with more tangible parts.
Nobody would try to copy a large source program, and just change data structures and claim uniqueness. Surprisingly, when I wrote Crafty (bitboard) I did not copy the evaluation from Cray Blitz (mailbox) and convert. But I did do something very close in quite a few places.
If you start from scratch, it is much harder than if you start from a working code, even if you want to change to bitboard. It is far easier to change something that is already working, because now you have something to compare against to make sure you got it right. When you start from scratch, you debug from scratch. So if you can avoid that, it is always a win. Unless it violates rules you have to abide by one day..
If you start from scratch, it is much harder than if you start from a working code, even if you want to change to bitboard. It is far easier to change something that is already working, because now you have something to compare against to make sure you got it right. When you start from scratch, you debug from scratch. So if you can avoid that, it is always a win. Unless it violates rules you have to abide by one day..
Hey! Bob, Rybka is "... unique and is owned by the representing programmer." It passes muster - why was it banned?
Because it is not original and contains parts of the program fruit.
Now wasn't that simple???
Now wasn't that simple???
So, what program is original? Is there a such thing as a purely original program that dropped from the sky code intact?
Name an author of a program that lived in total isolation.
Name an author of a program that lived in total isolation.
Crafty. Gnuchess. Cray Blitz. Chaos. Belle. Deep Thought. Hitech. Chess 4.x. I saw source for most of those at one time or another. Today I feel fairly confident that most programs are original, particularly the ones that have come up thru the ranks over time, such as Shredder, Hiarcs, MchessPro, Junior, Fritz, Chess Genius, just to name a few well-known commercials. There are other non-WC-class programs today, but I am not going to name any for fear of omitting several I should remember. The originals far outnumber the clones...
You don't have to live in total isolation. You keep up with this crap of copying ideas, which is totally fine. It is the source code that can't be copied, and it _really_ is that simple...
You don't have to live in total isolation. You keep up with this crap of copying ideas, which is totally fine. It is the source code that can't be copied, and it _really_ is that simple...
"Original"? Three of the programs that you listed began searching like Rybka and having a playing style similar to the Rybka series (at least, for two of them) after the Rybka code was made available. I'm not making accusations of anything untoward, as I believe that this is all perfectly fine, and it's possible that everyone would agree that everything is fine at the code level. Nonetheless, it goes to show you that sometimes, it depends on what one's definition of "original" is.
From out of the ICGA rules - note the word "original" is not used- but skillfully preferred by Dr. Hyatt.
The word preferred in the rules is Unique- the little bit of Fruit code did not over shadow Rybka's essence!
The purpose here is to ensure that each program is unique and is owned by the representing programmer.
The word preferred in the rules is Unique- the little bit of Fruit code did not over shadow Rybka's essence!
The purpose here is to ensure that each program is unique and is owned by the representing programmer.
Well, I think that anything that violates either GPL or copyright would be said to be not "unique". Whether this was the case, particularly with versions 2008-2010, has yet to be determined. Perhaps the proper procedure would have been to refer Rybka to proceedings with this and get those results before making any determination on whether or not it is "unique". That would be much more official and well-supported.
> Well, I think that anything that violates either GPL or copyright would be said to be not "unique".
You are putting the cart before the horse- no verdict has been handed down determining either case. Perhaps that would have been criteria for an investigation by the ICGA -like you they too put the cart before the horse.
According to their own ICGA rule for entry Rybka did not break any of the ICGA rules as it is a unique program clearly owned by the programmer. You cannot get around that!
The taking of ideas is not new to programming -Vas made some poor judgment calls early on in his career -most people do- not enough to change the fact of Rybka's singular uniqueness as a chess engine remains solid.
The current issues should be food for thought for future programming of chess engines and their entry into tournament competition but not used as an excuse for the perment banning of Rybka. This is all bullshit and they know it.
I am not putting the cart before the horse. I made it clear that it would have made much more sense to get a verdict from the legal system concerning GPL/copyright violation. If this was the case, this would be a clear statement that the program would not be "unique". This is necessary because there isn't really such a definition in their current system, so the next best thing is to go with something similar in the legal system. Of course, none of this has occurred.
Absolutely and completely false. In another post I quoted rules from 3 of the last 4 ICGA WCCC events. "original work" is in _every last one_ of them...
Of course, honesty is not exactly a hallmark in these discussions, so feel free to continue making things up as you go...
Of course, honesty is not exactly a hallmark in these discussions, so feel free to continue making things up as you go...
The definition of original is simple. Code written by yourself, _only_. If you copy code from others, no matter how much or how little, your program is no longer original.
Here's a sample:
o·rig·i·nal work
Originality is the aspect of created or invented works by as being new or novel, and thus can be distinguished from reproductions, clones, forgeries, or derivative works. An original work is one not received from others nor one copied based on the work of others. ....
It really is "that simple".
Here's a sample:
o·rig·i·nal work
Originality is the aspect of created or invented works by as being new or novel, and thus can be distinguished from reproductions, clones, forgeries, or derivative works. An original work is one not received from others nor one copied based on the work of others. ....
It really is "that simple".
Well, as others say, the probability that most of the programs in the competition are truly "original" is quite small, if you use the strictest interpretation of the word. On the other hand, if you have something that is largely unique and plays noticeably different, especially better, moves, then in the commercial realm, this would typically be considered original. Again, it's a disconnect between the academic and commercial realms here.
plays better moves is a statement that can be tested and proven true or false. But it doesn't say a thing about whether the program is "largely unique". An example is to reverse the comparison in the alpha/beta pruning code, so that good is bad and bad is good. Programs are identical except for one character, yet they play "noticably differently, one being especially better." yet they are dead copies. 44,833 lines, 1.87M characters, with 1 character different. And they play _miles_ apart.
Skill level has nothing to do with "uniqueness" or "originality".
Skill level has nothing to do with "uniqueness" or "originality".
It has plenty to do with uniqueness if the programs are, by and large, completely different programs. The only matching code that we have seen thus far is in a non-playing aspect of the program and which could easily be dismissed as accidental, with the consequence being either to change the offending code or to release the source.
How can you "accidentally" copy someone else's code and then deny doing so. The same way you can "accidentally" copy someone else's paper and turn it in as your own? Or the same way you can "accidentally" copy answers from your neighbor's test in a classroom?
One doesn't "accidentally" copy something from someone else. That is as deliberate an action as one can perform.
One doesn't "accidentally" copy something from someone else. That is as deliberate an action as one can perform.
I deal with copyright code that has been used unknowingly in products all the time. It usually happens when the code is placed in a publication without permission. People then assume the code is public domain and use it, only to find out later, it's not. Apparently that couldn't be the case here though, because you guys don't publish math papers...
I don't see any way that would apply to chess engines...
I can. All the robo code is now in the public domain. If you think that none of the other engine developers are looking carefully at it, you are not thinking straight. If you think there are no engine developers who will not take pieces of it that work well but which they can't understand well enough to recode, or that they are too lazy to recode, you're kidding yourself.
Again, the topic was "accidentally copying." That is simply not possible in a chess program development. If you copy something, it is a deliberate act. Source code from another program doesn't sneak into your code in the middle of the night while you are asleep. It it shows up, it was a deliberate and intentional act to put it there. I don't see any reasonable explanation for how a single author can "accidentally copy" someone else's code.
It is perfectly legitimate in the real world to copy public domain code. Code in robo is purportedly public domain. Therefor any author can perfectly legally copy code from this source. But things get sticky if the code is not really public domain, i.e. if robo contains copyright code from other sources. Saying this can't happen is very naive.
Please spare me the response that this is not compliant with ICGA rule 2. This is obvious and irrelevant (because nobody really cares about the ICGA).
Please spare me the response that this is not compliant with ICGA rule 2. This is obvious and irrelevant (because nobody really cares about the ICGA).
Seems that quite a few care, else why are these discussions continuing?
methinks you protesteth too much...
And nobody claims copying public domain code is illegal. But copying copyrighted code certainly _is_.
And no matter what you copy, it can't play in ICGA or other events that use the same rules, such as CCT and ACCA...
methinks you protesteth too much...
And nobody claims copying public domain code is illegal. But copying copyrighted code certainly _is_.
And no matter what you copy, it can't play in ICGA or other events that use the same rules, such as CCT and ACCA...
Hi Uncle Bob, If an eidetic could visualize say the phone book of New York and could do it without a single mistake then he might be able to do this in a smaller chesscode too. Let's not debate such basic things. It's not about what you believe or not.
Look at the definition of " Original". You have taken only one aspect of that definition and called it the whole of the definition. You are always twisting things around. It can even mean prototype. Who enters a tournament with a Prototype and continues to use a Prototype?
Idea's evolve through the study of other's ideas- We are talking a "Body of Work" not isolated parts of that body of work. It is the whole of a body of work that determines 'originality'. It is how ideas are gathered and used taken to form a whole work in expression say like that of Rybka as opposed to that of Hiarcs. You don't dissect a work and determine that this is not Picasso, " I'm not paying for that idea- he painted into this composition. Hell, he took it from another artist-and he didn't even give credit to the other art for that conceptual fragment. Oh, my God! Rip that right off of the canvas."
original (e-rîj´e-nel) adjective
Abbr. orig.
1. Preceding all others in time; first.
2. a. Not derived from something else; fresh and unusual: an original play, not an adaptation. b. Showing a marked departure from previous practice; new: a truly original approach. See synonyms at new.
3. Productive of new things or new ideas; inventive: an original mind.
4. Being the source from which a copy, reproduction, or translation is made.
noun
1. A first form from which other forms are made or developed: Later models of the car retained many features of the original.
2. a. An authentic work of art: bought an original, not a print. b. Work that has been composed firsthand: kept the original but sent a photocopy to his publisher.
3. A person who is appealingly odd or curious; a character.
The American Heritage® Dictionary of the English Language, Third Edition copyright © 1992 by Houghton Mifflin Company. Electronic version licensed from INSO Corporation. All rights reserved.
Idea's evolve through the study of other's ideas- We are talking a "Body of Work" not isolated parts of that body of work. It is the whole of a body of work that determines 'originality'. It is how ideas are gathered and used taken to form a whole work in expression say like that of Rybka as opposed to that of Hiarcs. You don't dissect a work and determine that this is not Picasso, " I'm not paying for that idea- he painted into this composition. Hell, he took it from another artist-and he didn't even give credit to the other art for that conceptual fragment. Oh, my God! Rip that right off of the canvas."
original (e-rîj´e-nel) adjective
Abbr. orig.
1. Preceding all others in time; first.
2. a. Not derived from something else; fresh and unusual: an original play, not an adaptation. b. Showing a marked departure from previous practice; new: a truly original approach. See synonyms at new.
3. Productive of new things or new ideas; inventive: an original mind.
4. Being the source from which a copy, reproduction, or translation is made.
noun
1. A first form from which other forms are made or developed: Later models of the car retained many features of the original.
2. a. An authentic work of art: bought an original, not a print. b. Work that has been composed firsthand: kept the original but sent a photocopy to his publisher.
3. A person who is appealingly odd or curious; a character.
The American Heritage® Dictionary of the English Language, Third Edition copyright © 1992 by Houghton Mifflin Company. Electronic version licensed from INSO Corporation. All rights reserved.
There is nothing in the definition you gave that contradicts a single thing any of us have written about ICGA rule 2.
I suppose you have a point buried in there somewhere? It was not obvious.
we are not talking about ideas. Not one time has the ICGA rule mentioned ideas. Not one time did the ICGA report claim ideas were copied. Ideas are acceptable. Source code, on the other hand, can not be copied...
I suppose you have a point buried in there somewhere? It was not obvious.
we are not talking about ideas. Not one time has the ICGA rule mentioned ideas. Not one time did the ICGA report claim ideas were copied. Ideas are acceptable. Source code, on the other hand, can not be copied...
It contradicts the why you are applying the term "original" when you are placing it up against the foundation of your rule 2. governing the determination of,
"The purpose here is to ensure that each program is unique and is owned by the representing programmer."
The percentage of another programmers "originality" would have to be so clearly present that the chess engine being entered into tournament was really someone else's program A fraud-charlatan. That was the implications of that rule, Bob.
P.S.
You want to RE on your witch hunt every chess engine to see if it is clocked -go ahead with your magnifying glass approach to see what is original - what bogus bunch of bullshit you and your buddies are handing out. But it is good that others see you for the sleaze you are.
"The purpose here is to ensure that each program is unique and is owned by the representing programmer."
The percentage of another programmers "originality" would have to be so clearly present that the chess engine being entered into tournament was really someone else's program A fraud-charlatan. That was the implications of that rule, Bob.
P.S.
You want to RE on your witch hunt every chess engine to see if it is clocked -go ahead with your magnifying glass approach to see what is original - what bogus bunch of bullshit you and your buddies are handing out. But it is good that others see you for the sleaze you are.
You are using an utterly twisted definition of "unique". Unique means different. Not sharing common parts. That is not inconsistent with rule two which says "original". If you take the broad definitions of both, rather than trying to find some slim intersection between the two, you reach the intent of ICGA competitions.
Topic Rybka Support & Discussion / Rybka Support / My 2 cents on the Rybka - Fruit GPL violation (if any)
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill