Talk:Rybka

From Wikipedia, the free encyclopedia

This article is within the scope of WikiProject Chess, a collaborative effort to improve Wikipedia's coverage of chess. For more information, visit the project page, where you can join the project and/or contribute to the discussion.
Chess Portal
B This article has been rated as B-Class on the quality scale.
Mid This article has been rated as mid-Importance on the importance scale.

Contents

[edit] old talk

Name Rybka is a little joke, since one of the world champion's name is Fischer(close to fisher)

Are you sure? "Fischer" really means fisher in German - but I cannot imagine that somebody would make such a strange pun when naming his chess program... Would it mean that this small rybka will escape even Fischer? --Ioannes Pragensis 08:01, 8 September 2006 (UTC)
It would be great to see some information on how rybka differs from other engines. Is any such information available?
It is not a marketing tool, it is an encyclopedia article whose purpose is to provide readers with knowledge about the current top rated chess engine. Dionyseus 07:03, 15 February 2007 (UTC)
The second line of the article states that Rybka leads all computer chess ratinglists. It has won many tournaments that are indistinguishable with the "world championship" in format or number of rounds. As for the world championship, with so few rounds the winner is basically the engine that manages to draw less than the others against the lesser opposition, which was the case that night. Dionyseus 07:30, 15 February 2007 (UTC)
That might be so, and your point about how such tourneys tend to be very short, hence the high luck factor is well taken. But the point is people generally do want to know who is world champion, it's kind of like people wanting to know who is the current Olympic champion (or sometimes world champion)in sprinting, even though it's just one silly race among dozens of races they will run in a season. So I pretty much added a part in, stating that it is top ranked in rating lists and has won many tourneys, though it has yet to win a world championship. I hope this criticism is mild enough for you, particularly since I'm just stating a fact. Aarontay 21:35, 26 April 2007 (UTC)


Name "Rybka" is rather polish word. It doesn't sound like czech word and this word exists in polish language(Vasik wife is from Poland)

Czech is my mother language and I know that this word is a correct Czech word and has the same meaning as in other Slavic languages where it exists - a little fish. Rajlich must have known the word from Czech a long time before he met his wife.--Ioannes Pragensis 22:13, 16 March 2007 (UTC)
"Rybka" is a Polish word (and a Czech world etc.) and means a little fish. Vasik's wife is Polish so it makes sense that he knows the word as a Polish word.
Nope - it doesn't make any sense: Vasik's parents are Czech, Czech is Vasik's mother language, so he knows the word as a Czech world. What you put above is same nonsense as if an Englishman married a German girl and as a result knew "fish" as a German word. —Preceding unsigned comment added by 212.90.231.188 (talk) 18:03, 29 November 2007 (UTC)

[edit] Critisms

Of course everyone kind of knows that Rybka is dominating the comp versus comp engine matches. But I think it might be a good idea to include some criticisms. The problem with pages like this on Wikipedia is that it ends up looking like an advertising/marketing page. Just a thought. Aarontay 21:25, 26 April 2007 (UTC)

A "criticism" section would have to contain stuff from reliable, reputable sources. This stuff is notoriously hard to find, so it is likely that such a section would end up being a repository for random editor's original venting about the product. I think this is a bad idea. --Ryan Delaney talk 19:52, 11 May 2007 (UTC)
Adding criticism for the sake of criticizing something is not a NPOV. There has to be a compelling reason to add it in,. and as Ryan pointed out it has to be from a reputable source. 75.85.171.230 21:35, 14 May 2007 (UTC)

The worlds strongest commercially available engine sounds more definitive than merely describing it as one of the best. Any engine that does not place last, could conceivably be said to be one of the best. I suppose if we wanted to get real specific we could write that no known engine performs better on 1,2, and 4 processor machines, which is what chess engines are now tested on. The purpose being because Hydra is said to be around 3000+ in strength, on 64+ processors, though there is no way to objectively determine this as with other privately owned and operated chess "machines" (another example being Deep Blue)

Also, as stated, Rybka gains roughly 50 Elo with each doubling of processors. Rybka does not gain 50 with every extra processor nor has the author of Rybka ever made this claim. Vasik's estimates give an increase from 4 to 8 processors roughly 45 Elo. Uavle 02:58, 15 June 2007 (UTC)

[edit] Relevance of sample game

What is the relevance of the sample game section? To the casual reader, it imparts zero information. To the interested chess enthusiast, it's just one game; how does it in any way impart information about the subject of the article? Oli Filth 17:19, 12 August 2007 (UTC)

I tend to agree with Oli Filth here. As no one else has commented this matter, I've boldly removed the section now. —ZeroOne (talk / @) 17:02, 15 August 2007 (UTC)
See discussion at Talk:Gothic chess. Apparently this is a more widespread problem of sample games that need to be dealt with in multiple articles.--Isotope23 talk 17:19, 15 August 2007 (UTC)
Strictly taken the sample game in this article is not comparable to those games presented in the chess variant articles. The sample game in this article was just meant to demonstrate the Rybka playing style, cunning moves or something similar. That's not a how-to do anything and neither I nor Oli Filth are implying it is. In chess variant articles, however, sample games are used to illustrate their rules, possibly making them removable how-to content. —ZeroOne (talk / @) 18:42, 15 August 2007 (UTC)
As I have stated on the Gothic Chess discussion area, if it is "human nature" to include a sample game for the sake of furnishing the type of content that is expected of the portion of the readership interested enough to replay the game, then by all means, I say let a sample game remain. I have heard a wide range of arguments, such as providing a sample game is contrary to the Not A How To guidelines.
Claiming a sample game is a "How To Play Chess" example is about as meaningful as claiming you can learn how to become a painter by looking at the Mona Lisa.
I have also heard that including a sample game is "not encyclopediac", yet in my 1963 Encyclopedia Britannica, I see a sample game of Reshevsky listed. So, it must be "encyclopediac" if an analog source included it from long before there was "the net".
In my opinion, some Wikipedia editors take it personally when you confront them with such facts that refute their arguments, so they change their arguments, and persue other avenues to keep content from being furnished. I have seen this done numerous times by some of the aforementioned posters who removed the Rybka sample game.
To them I say: Why don't you add some content of value? I can take any homeless man wandering the street, give him 10 seconds of training with a mouse and keyboard, and he can remove content from Wikipedia also.
GothicChessInventor 04:57, 16 August 2007 (UTC)
I'm fairly sure that people who are looking forward to become painters do study Mona Lisa at some point. —ZeroOne (talk / @) 12:09, 16 August 2007 (UTC)

And, in typical fashion, ZeroOne implies that I have said something that I have not, with his own misconstrued reply. I never said anything about studying the Mona Lisa. I'm sure art students or those looking to become better painters study it also. That doesn't mean you can look at it and learn how to paint. You must practice painting no matter how much you stare at the best example of a great work of art. The same is true of sample games of chess (and Gothic Chess.) One sample game is not a violation of the "Not A How To" pillar of Wikipedia. You can't look at one sample game and know how to play chess, nor a chess variant. If that was the case, you could look at the Trice-Polgar game on the Gothic Chess page then give me a battle. I think it is safe to say you are not claiming to be able to hold your own against me in a game of Gothic Chess. And if this is the case, the argument you have put forth regarding "Not How To" falls apart.

GothicChessInventor 13:09, 21 August 2007 (UTC)

Sample games are an established precedent in many chess articles, especially chess biographies. Probably the best place to start a discussion about the relevance of sample games would be on Wikipedia:WikiProject_Chess. --Ryan Delaney talk 03:12, 17 September 2007 (UTC)
I don't agree that sample games are an established precedent. I am very concerned about them, and I think that at least one other active WP:CHESS contributor is as well. Concerns have been mentioned in passing at WT:CHESS, although no one has had the interest to start an in depth discussion of the issue. I do agree that it would be good to hash that out in a central place to get some guidance as to the consensus view about sample games. Also, as an unrelated aside: GothicChessInventer, would you PLEASE QUIT WITH THE DIV BOXES. You have been asked nicely many times, and we are tired of asking. Actually I'd be happy if everyone involved could try to confine the mud-wrestling with Ed Tice to the Gothic Chess and Capablanca Chess ghetto rather than letting it spill out elsewhere in Wikipedia. Quale 05:59, 17 September 2007 (UTC)
Whether or not there is a precedent for such things, no-one has yet answered my original question above, namely what the purpose of this section is in this instance. I don't see what useful information it imparts; it's just one game. I can possibly see the merit in some annotated examples and discussion of particular strategies the algorithm uses. But as it stands, I can't see how the section imparts anything useful, even to a seasoned chess expert. Oli Filth 09:27, 17 September 2007 (UTC)
As no-one has yet answered this question (which was the original justification for removal of the section), I have once again removed the sample game, as I don't see what purpose it serves in the article. Oli Filth(talk) 23:04, 25 September 2007 (UTC)

[edit] Claiming "strongest" violates WP:NPOV, WP:PEACOCK

The following line keeps being deleted and re-added:

Rybka is the world's strongest commercially available computer chess engine.

A claim like this violates WP:NPOV and WP:PEACOCK. The best you can do is present objective evidence (which is already done: the list of titles gained and rating lists topped). --IanOsgood 16:57, 26 September 2007 (UTC)

I disagree. It does not fall under WP:NPOV or WP:PEACOCK because it is both important (see WP:PEACOCK#Do not hide the important facts) and quantifiable; in particular, it is a perfectly valid claim when backed with references, and is preferable to circumlocution (has won this or that tournament, tops this or that list). I'm even considering reinserting it, but I'm not sure at the moment; the current intro is reasonably clear. GregorB (talk) 14:06, 27 May 2008 (UTC)
I agree. Leaving out this kind of peacock phrase does not hide an important fact, because "Rybka is the best" is not an observable fact. Rybka's achievements in the rating lists and matches against other engines should speak for themselves; making this kind of assertion does not help us, unless we are selling something. The following passage from the Wikipedia:Neutral point of view policy is illuminating:
Assert facts, including facts about opinions—but do not assert the opinions themselves. By "fact" we mean "a piece of information about which there is no serious dispute." For example, that a survey produced a certain published result would be a fact. That there is a planet called Mars is a fact. That Plato was a philosopher is a fact. No one seriously disputes any of these things. So we can feel free to assert as many of them as we can.
By value or opinion, on the other hand, we mean "a matter which is subject to dispute." There are many propositions that very clearly express values or opinions. That stealing is wrong is a value or opinion. That The Beatles were the greatest band in history is an opinion. That the United States is the only country in the world that has used a nuclear weapon for military purposes is a fact. That the United States was right or wrong to drop the atomic bomb over Hiroshima and Nagasaki is a value or opinion. However, there are bound to be borderline cases where it is not clear if a particular dispute should be taken seriously and included.
When we discuss an opinion, we attribute the opinion to someone and discuss the fact that they have this opinion. For example, rather than asserting, "The Beatles were the greatest band," we can say: "Most people from Liverpool believe that the Beatles were the greatest band," which can be supported by references to a particular survey; or "The Beatles had many songs that made the UK Singles Chart," which is also verifiable as fact. In the first instance we assert a personal opinion; in the second and third instances we assert the fact that an opinion exists, by attributing it to reliable sources.
Simply put, we should stick to the facts. That Rybka is the strongest engine is not a fact; it's an opinion that depends on many other assumptions (What makes an engine the "strongest"? Rating performance, or match performance? Under that metric, Zappa would be "stronger") that are up to the reader. --Ryan Delaney talk 22:17, 8 June 2008 (UTC)
"Rybka is the strongest chess engine" is as appropriate as "the Pacific Ocean is Earth's largest ocean". Relative chess playing strength can be (and has been) measured in a reliable manner, I wouldn't say that it's a matter of opinion. GregorB (talk) 23:10, 8 June 2008 (UTC)
Here, this is not the case. We are not talking about reasonable dispute, but what is in principle a possible subject of dispute. What is reliable is that Rybka leads all the major rating lists. But for the sake of argument, suppose I wanted to say that Zappa is stronger, since Zappa defeated Rybka in a match, and that's how we determine strength among humans. The ensuing hairyness is what the neutrality policy rescues us from; when we only talk about results, we can't get ourselves into trouble. From that perspective, I think we can all agree that Rybka's results speak for themselves. --Ryan Delaney talk 23:56, 8 June 2008 (UTC)
There's one thing here, though: saying that Rybka is the strongest based on chess engine rating lists implies that chess playing strength can be reliably measured that way. I believe it is true, but that's precisely the problem: it is neither obvious nor widely accepted. So yes: in a way this would only shift POV one notch further (and add an OR problem, perhaps).
On a side note: intro to Haile Gebrselassie says he is "widely considered as one of the greatest distance runners in history". (Not my line, but I provided five references; I could have added five more without a problem.) This could be construed not only as WP:POV and WP:PEACOCK, but also WP:WEASEL ("widely considered"). However, in this particular case, I believe his intro would be worse without it. GregorB (talk) 08:04, 9 June 2008 (UTC)
You might be right in the case of Haile Gebrselassie. What I think is different about that statement is that it conveys public perception which is a fact that can be reported -- if we had enough reliable sources. If you can find reliable, third-party sources that report on a public perception that Rybka is the strongest chess engine, then we would have something to go on. --Ryan Delaney talk 16:01, 9 June 2008 (UTC)
Correct. But apparently there aren't any, so I'd have no trouble leaving it as it is. GregorB (talk) 16:27, 9 June 2008 (UTC)