FYP

Fine, but you'd be surprised how many people make that error because of misunderstanding, not laziness.

What if humans evolve into computers while at the same time computers evolve into humans?

Computers will start programming humans.

i think comps will get dumber in the next 20 years due to overuse and stamina problems.

it's a vicious cycle, regressive evolution

In before Devo.

You are overestimating the impact of a temporary reovertaking by humans. I think you would agree that people would simply assume that there isn't the same urgency to improve chess computers as there once was. And you would also agree that if hypothetically that urgency returned, human players would have no chance a few years hence. It doesn't have to actually happen to put a damper on the game.

^^^ Makes sense

and holy **** Sklansky posted in a chess thread.

Haha, funny enough, I bought NLHE TAP today and it looks awesome. Some of it seems to confirm some ideas that I've come up with on my own and some of it is new and interesting.

David,

As for a temporary re-overtaking only serving to spur the programmers to "lock down" the game once and for all, I'm sure you're right. I still want to see it happen in my lifetime, though! I feel that it will somehow vindicate my love of the game. Too many people think that chess is like checkers, which is blasphemy.

Of course, you might well say that I shouldn't need validation for my love of the game. And I don't. I will always play chess and enjoy it. But it would be nice to have.

Edit: wow, when I first posted this I spelled bought "bot" in the first paragraph. o_O I must be going soft in the head.

Why not just come up with a simple wrinkle that will thwart computers but not chess talents. Nothing random or anything like that. Something simple like allowing a rook to move like a knight but only once during the game should separate the men from the boys (and the programmers).

so my grandmother was on to something

That's actually clever. I've thought about this problem before and here are some of the things that I came up with:

- Before the game, the players first place their pieces on the back rank however they choose, alternating. Obviously the advantage here is to drop second, so White has to place first. This, I think, would almost be enough to give humans the edge over computers. Programmers might think that they could just copy White's set-up, but this leak would be obvious and good players would choose a set-up that has several undefended pawns, giving White an easy attack.

- Give each player an extra queen that they may, on any move, drop on an empty square instead of moving one of their pieces. Humans would adapt by not leaving holes around their king until after the enemy queen had dropped, and this would add so many extra permutations that the engines would be back to seeing maybe 5 full moves ahead in all variations.

I like your idea, though. However, I would give the queen one knight-move; or, to make things even more interesting, give every piece the right to move like a knight once during the game (except to capture the enemy king, of course), perhaps limiting the number of non-knight knight moves to a small number. Or maybe giving every piece one knight's move to use. The pieces would then be categorized as having used their knight's move or not, and the burden of keeping track of this would likely be too much for the engines.

I think you drastically overestimate how much people care that computers are better than humans in chess. Given the major advances in computers in the last decade I'm surprised it's as close as it is right now.

Also the idea of adding an extra rule or two should have very little effect and have the unfortunate affect of bastardizing the game of chess for a cause that almost no one cares about.

You don't think that David's and my proposed changes would give humans a significant edge over computers? You underestimate how hard it would be for computers to keep track of these one-time knight moves and their implications.

I think that it could be done in an elegant way that would improve the game rather than "bastardizing" it.

I also think it's a bit silly to talk about "bastardizing" a game that has gone through myriad rule changes throughout its history. The classic games evolve: that's why they're classics. A few hundred years ago castling wasn't the same as it is now. And a couple thousand years ago the game bore little resemblance to the chess that we know and love today.

Both Capablanca and Fischer were worried about a "draw death" and advocated rule changes. Many players express distress at the opening preparation that's become necessary to truly excel at the game. To be honest, I think that a rule change is inevitable, if not in the next few years then certainly within the next century. Either that, or one of the many variants played online will overtake "classical chess" in popularity and become, essentially, the "new chess".

I already think that bughouse is a better game, strictly from a gameplay perspective. However, I am new to the game and may have to re-evaluate after becoming more experienced. It's possible that bughouse doesn't offer the same depth as classical chess, but it's both fun and fascinating.

In the next century maybe, but a fischer-random type of change seems much more natural to me. Although I don't think it'll happen....chess is very hard, and there is still nothing too close to a draw death.

But the opening preparation already necessary is very annoying; I don't think many serious players contest this point.

I'll let you decide which two pieces should be chosen. But there is no doubt in my mind that given the goals you want to achieve my way, or an equally simple variation is the the way to go. To get the public interested you want the rule change to be instantly explainable and obviously intriguing.

And even though I barely have a right to talk chess strategy, I would think that giving the queen the one time option creates the drawback that a good classic player who is weak in Sklansky Chess will quickly try to get you to swap queens.

I think plenty do.

A minority. Gluttons for punishment who enjoy poring over pages upon pages of Najdorf theory.

Well, maybe players with good memories who don't mind boring study like the necessity of opening preparation because it gives them an edge over their opponents.

Unless I'm completely missing something, all of these ideas would obviously make it harder for humans to beat computers.

Yeah this is probably true but only because it would take some time for humans to get used to the new movements. In theory it'll make the computers slightly slower, but considering that they will have no problems instantly adjusting to how a new piece moves, they will have the edge at first.

No, no, no. The rule changes would give humans a significant edge. They would not make computers "slightly slower", they would hamstring the engines. Look at how weak computers are in bughouse...

Most players drastically overestimate how much opening theory is needed to be extremely competitive.

I presume you mean that people should adopt a simple (yet reputable) system against each opening they're likely to encounter and just get good at playing the resultant positions? e.g., you could go with Bb5+ against the Sicilian, Exchange Variation against the French, Scotch against e5, Panov-Botvinnik against the Kahn, Austrian Attack against the Pirc. Wouldn't require too much theory and you wouldn't be playing crappy openings.