True or false? Chess will never be solved! why?

It is?  I'd love to see your proof.

Why are we wasting our time trying to answer a question none of us can answer instead of trying to work together to find out how to start solving the problem?

Those of us who appreciate the fact that the problem can't be solved won't be very productive contributors to that discussion.

(Well, no-one will, really, but we realize our time is better spent.)

It is?  I'd love to see your proof.

I have the proof.  It won't fit in your universe, though, so you'll have to lump it ;).

i love speculating !!

So would I. Is it in a book somewhere? 

Just a passing reference to it -- the margin is not large enough to hold the actual proof.

A book or a "Shannon number" taken on faith is enough proof for holding a stick-in-the-mud position about requiring absolute proof, apparently :).  Ironic. 

You have never been to Mongolia.  Can I prove to you that it exists to your satisfaction?   No.

Splendid! 

The routes to Mongolia are already well mapped. Nothing equivalent exists for that opening.

Isn't it funny when you're just sitting calmly and all of a sudden men of straw start marching before you?

So the same essential joke I just made is suddenly splendid coming from another quarter...one that you agree with ;).  Interesting.  I am picking up all kinds of cool nuances of human nature in this thread...carry on.

The straw men started on page 82, I guess.

The whole analogy is silly.  Mongolia is not a mathematical proof.  I've seen enough evidence can take it's existence as a near certainty without having to see a proof.  I've seen no evidence of a solution to chess existing, nor even of a feasible strategy for obtaining one.

The analogy works as intended.  I am merely pointing out that it is impossible to prove anything to someone that refuses to accept it, since the universe only "exists" via each person's perceptions inside their own brain.

No more, no less.

Nope.

The burden of proof is on the one who, well, claims to have a proof.

In that case, you claim that you will have within say 100 years a proof of the evaluation of each position - hence you have to prove it.

The Mongolia analogy is pretty relevant actually. Indeed, we have no absolute proof Mongolia exist (Mongolia being defined by "something above sea level including points of coordinates blah blah on the Earth"). Even those who have went there, because all the information they had comes from their eyes that are fallible detectors. No absolute proof.

But who needs one ? Everyone can guess that Mongolia exist, and the probability is well enough for most applications.

Here, we want to know (not to guess) the evaluation of the positions. 99.9999% is not enough to have a satisfactory mathematical answer.

We all know Mongolia exists because of the talilaserbillichess expansion pack, of course.

In that case, you claim that you will have within say 100 years a proof of the evaluation of each position - hence you have to prove it.

Not quite.  I said that it is was very possible that "best play" will be achieved and that holding the position that it is impossible is not without flaws ;).  I don't think I have argued any absolutes at all (and I usually don't).  All I did was poke holes in the opposing absolute position. 

If you're playing white, then it *might* be possible. If you work backwards from a position of checkmate, then you *might* be able to "solve" chess by forcing different variations that lead to the same position on the board.

The anti-Mongolia example : http://en.wikipedia.org/wiki/Miller%E2%80%93Rabin_primality_test

Testing if a number is a prime can be done with admitting a small probability of mistake. The running time increases when you decrease the authorized probability of failure.This algorithm has a running time of the order (k.log(n)^3) to obtain a certitude of 4^-k that a number n is prime.

Saying k = 500 (ie you accept a probability of failure of 2^-1000 which is around 10^-300, which looks fairly safe), you have an algorithm of the order log(n)^3, when the shortest deterministic algorithm known is aroung log(n)^6. Yet somehow mathematicians consider that the deterministic algorithm is worth more.

Thanks, a nice example.  According the Mr. Grobe, though, you can't prove an integer is prime unless you try to divide it by every integer less than it.  You can't even make a rule saying "even numbers are never prime, skip those", because that would not be enough proof, because somehow it might magically come to pass that there's an even prime number and that the laws of mathematics don't apply at some astonomically high number (kind of like relativity?), so you cannot take shortcuts.  You must apply brute force calculation to every number without using square roots to eliminate possibilities, etc.  Anything less than that leaves room for doubt.