trw wrote on 06/21/09 at 21:32:39:
CraigEvans wrote on 06/18/09 at 21:15:03:
You learn 20 moves from a GM's book, you go into the game and after that 20 moves, you are on your own.
You learn 20 moves using a computer, you go into the game and after that 20 moves, you are on your own.
I'd actually like to seriously discuss LSA's point because it is a good one.
I agree with it and don't think I can refute it but for the purpose of fun... i'll play devil's advocate.
I would point out the main difference for me between a GM book and a computer line for the 20 moves is as follows:
1) The gm's moves usually have a sense of a concrete plan... you can see the pawn structure, space, piece play, endgame even taking shape. To boot, on top of the moves there is usually explanations and sidelines that help prove the point of endorsed plan. And in the end... the evaluation is as much your own opinion of the resulting position as it would be the GM's notated "=" or "+=" etc. What i'm trying to say is I don't believe it is possible to simple read a book by a GM and have only retained rote memorization. Either you will be able to understand the opening and the ideas behind it... or frankly you won't have the skill to memorize.
2) I see more rote memorization from databases/computers which leads to the next point. It is very difficult to follow computer analysis with understanding as often times the computers will endorse dumb plans that only make sense if you can see 15 moves ahead in all lines. Sometimes the computers themselves can't see this due to the horizon. I would further point out that your own creativity is stifled by computers as it is next to impossible to ignore = "+=" when it comes with a number in the very line you're analyzing. To this effect, once you start analyzing a position with a computer it is hard to stop!
As per the argument itself. Obviously no form of preparation for your game can be considered either cowardly or cheating only effective and non effective.
An interesting attempt at fighting LDZ's battle for him, Wizard, and one that made me stop and think for a little while. Here are my comments:
1) I am one of a number of players I know who have never ever sat down and read a chess book properly. By that, I have almost never sat down and played through the lines on a boar,d I've never taken it away and looked in databases for typical positions, I've never really even read over much of the text other than the moves themselves. A lot of my old dragon theory (some of which I still weirdly remember) I gleaned from a particularly long and boring train journey when I was 18, accompanied only by The Complete Dragon and my MP3 player. Did I have a real understanding of what I was doing in the opening... perhaps, perhaps not, that would have to have come from my understanding of chess in general. Did I know 20+ moves of theory in some lines? You bet.
2) Hence I actually think that (similar to Smyslov's comments) rote learning is far more likely with books. As he says, the computer is used by any good player as a "partner", and moreover, a partner who you need to guide. As you rightly say, the computer's evaluations only make sense to himself and are based on concrete variations under specific pre-defined parameters, which are often misled by the horizon effect, material considerations, simple positional factors such as passed pawns etc. In other words, any good player worth his salt can completely look past =+'s given by the computer - if you play some lines of the French or Caro-Kann or Alekhine through a computer it will often give white a nonsensically large advantage because it understands nothing. A human who prepares his openings based solely on this number will actually be disadvantaging themselves compared to just trying to work things out OTB.
Either way, I suppose the main point is that even if the GM has an idea of a concrete plan as well as the moves - if someone else is giving you that sort of help to learn moves then under LDZ's definition, that would be cheating. In fact, one could argue that books and magazines are even more a "cheat" in this sense, exactly because of the additional guidance, tips, plans and comments they give. The computer takes whatever moves you give it, and outputs a number at the end - it is up to the human to interpret this however.
An interesting case in point (very brief digression) is in the very detailed analysis done by Markovich, LG, Kam (and myself in places) of some Alekhine lines. Sometimes even as far as 25-30 moves eep the computer is giving white a huge "+/-" and then suddenly realises his mistake. A computer alone could never find some of the lines which we have found on those threads which actually make black look quite reasonable in, for example, the Voronezh variation. Realistically, a human alone would probably also struggle. But to say that using a computer to help you prepare for games is cheating... well, I think we're all agreed it's a pretty weak position to take. Computers, books, magazines, GM seconds/thirds/fourths (if you're Kasparov et al) - it's all the same. It's absolutely no different to revising for an exam, getting a tutor to help you in areas you are struggling with - it is help for the pre-test preparation. As long as you don't have the answer book then it isn't cheating. Since you're on your own as soon as you're inside the tournament hall, it isn't cheating. It's just replacing paper with bytes, and GMs with a calculator.
As for the freedom of speech issue... well, we've been over this a thousand times. Think Markovich, Smyslov et al have covered that one nicely.
I think you'll have confused some people by calling me LSA, however, Wiz. That alone is justification enough to do it...