endgame tablebases

Why doesn't chess.com have the Nalimov tablebases?

Well, it would be a nice addition to the site even if it is lready available online, and it is completely legal and fair 

Not according to chess.com. See site-playing-rules-a-guide-for-the-perplexed.

I wonder why it isn't legal. In correspondence it is.

It seems to me, if opening databases are allowed, then the endgame tablebases should be as well. Or else neither should be allowed.

Wow - I can't believe the "fairness" of using tablebases in Chess.com games is actually being debated. Obviously, if tablebase assistance is allowed, anything goes (that includes having Deep Fritz 12 play every move in all your games), and we might as well do away with banning members for cheating. Better yet, do away with the whole site; we can easily play against our own computers without it, and there are more formal ways to conduct engine-vs.-engine matches ;)

...except that there are no "winning", "losing", or "drawing" moves in the opening database, and even OTB it's extremely rare that a game is truly won or lost during the opening phase. Furthermore, choosing which opening moves to play (even when using a database) is mostly a matter of personal taste, and one may deviate from known theory very early on if they wish. Using tablebases to ensure victory/equality in difficult endgames (i.e. Q+P vs. Q+P, Q vs. R+2P, R+N vs. 2N, Q+N vs. 2R, etc.), on the other hand, directly affects the outcomes of many games (especially between Class players).

Honestly, though, I don't think that either resource should be allowed... so I guess I'm agreeing with you to a certain extent.

When you say that tablebases are "fair" (see highlighted text), do you mean that people should be allowed to use them during online games (literally plugging in the current game position and then playing the best move that Nalimov gives)? If so, then RainbowRising is absolutely right - because if consulting Nalimov in a given endgame position is fair, consulting Deep Rybka 4 in the same position is fairer (at least the latter is [occasionally] fallible). And if you can use computers when there are just five or six pieces on the board, then you might as well be allowed to use them at any given point in the game. If chess were completely solved 10 years from now (Yes, I'm fully aware that won't happen.), would games be adjudicated (see post #7) before the first move?

Perhaps you totally disagree with the above logic. Well, then answer this: if a 1600-rated player reaches the position below in an online game, does he deserve to win if his 1300-rated opponent makes no further mistakes?

Of course, endgame tablebases are an important online resource that everyone ought to use - but only between games, not during them. And this can't be compared with including tablebase variations in books, as there are no rating points to be gained from it, and the authors generally remember to give Nalimov credit for such lines...

More interesting would be to arrange an official MMAC1  tournament, where each individual player decides which type(s) of artificial intelligence will dictate to him/her what moves to play, and the phase(s) of the game during which his/her selected resource(s) will be utilized. Then at the end of the tournament, the 1st, 2nd, and 3rd-place winners are required to explain the exact method(s) in which they cheated.   

1. MMAC - Mixed Martial Arts of Cheating   ;)                            

__________________________________________________________________________________

Btw, it would be pretty amusing if a mod. locked this thread, or simply deleted it altogether, imo. The OP's question might be the most ludicrous I've ever seen - having tablebases on this site would be silly given their easy accessability in so many other parts of the web.

Everyone is missing the major point here: people who use these things are lame non-chessplayers. They'd rather watch a computer play brilliantly than try to win a game on their own.

cobra, I don't know why you are equating the resource being perfect/fallible to it being unfair/fair. Fair refers to a level playing field. If all players have access to the exact same resource, the use of it in games is would be fair. Since the endgame tablebases are a standard resource (especially if chess.com hosted their own), the rules could allow their use while maintaining fairness.

Of course, just because something is fair doesn't mean it should be allowed.

To be precise, my suggestion about allowing endgame tablebases pertained to turn-based chess only (seeing that opening databases are already permitted there). I would be strongly opposed to any form of assistance being allowed in Live Chess.

I am not sure that winning/losing games in the opening is as rare as you suggest. In blitz it's actually pretty common, and even in turn-based I have had games that were won or lost in the opening. (If it's more rare, that is due to database use.) Even if you just obtain an edge out of the opening, it can go a long way, in my experience. On the other hand, one might argue that "even" endgames with 7 pieces (tablabases would hardly matter when one side is up a pile of material) don't occur that frequently.

I think it would be a real shame if this were allowed, as in practice you might as well just adjudicate any position that reaches seven remaining peices.

I'd like to be able to continue practicing my endgame conversions, and would have to go somewhere else to do it if this change were made.

chess.com routinely provides resources that are available elsewhere. There were already correspondence and live chess sites before chess.com came around. Databases of games and opening explorers were around before chess.com released theirs. I don't think it's part of chess.com's business plan to send their users to a different website to get relevant material.

Something being against the rules in gameplay doesn't seem to be a good reason to not have it on the site at all. For example, chess.com provides postgame computer analysis. However, I don't think (though I haven't seen anything recently) that this comptuer analysis includes endgame tablebases.

In terms of making it available without changing the rules regarding its use, I wonder if there aren't copyright/availability issues preventing it.

This seems a bit uncharacteristically rigid. I don't like the use of databases in correspondence, but I accept that these are the rules. Of course, I can play Live Chess where it is not permitted, as could you.

 I was half joking when I said engines were more fair than tablebases because of relative fallibility; you're right, "perfection" and "unfairness" obviously can't be equated. My point was that using a tablebase to find moves is no different than using a [perfect] engine for the same purpose.

But how is it unfair to allow tablebases (which would be available to everyone, maintaining a level playing field) in online games, you ask? Because it favors players with weaker endgame skills (just like engine use favors weaker tactical players). For instance, there are some players who can force a win with Q+P vs. Q, and some who can't: when these players reach such a position, the first group should score a win (assuming it's winnable, and they don't mess up), while the second group should split the point. A rule that allows both groups of players to win isn't technically "fair".

Hey, maybe they should change the distances for Olympic races: make the 100m dash just 5m.  After all, it's fair (since it applies to everyone) and, more importantly, I'd like my chances against Usain Bolt in a 5-meter race  ; )

Sure, but that's not at all my preferred time control.

Additionally, opening databases and endgame tablebases are fundamentally different in nature.  One can quite easily use an opening database poorly, and end up in trouble against a well-booked opponent who doesn't use one at all.  This is not the case with tablebases.  In fact, as I said, if endgame tablebases were allowed there's really little to no reason to play out any endgames that reach seven remaining pieces and I think that would be a shame. 

Yes, well, we already have opening databases that favor players with weaker openings knowledge, so it seems fair to level the field.

cobra, by the same argument, opening databases are unfair. They level the playing field between players who know and don't know their openings.

I almost wonder if it doesn't come down, at least in part, to enforceability?

The use of opening databases simply can't be detected.  The use of an endgame tablebase can.  As a result, effectively prohibiting the former would be impossible, as opposed the latter, which should be relatively straightforward.

Could it be that philosophically the preference might be to allow no outside help, but that practicality dictates opening databases must be permitted to maximize fairness?