Quote:
Originally Posted by YouKnowWho
I feel incredibly dumb, as I have absolutely no idea about what is going on in both the blog post and especially the PDF. I read the blog entry, I was able to comprehend words, but I don't understand the point behind it at all.
So, please answer a simple man's question - what is the point behind that blog post? I would be glad if you used simple words too.
The issue is essentially the question "What would somebody who claimed to play perfect chess have to do to convince you?" It isn't really possible to actually play him over and over again to convince yourself. You can be convinced that he will always beat you, always beat the top 5 players in the world working together etc... but you really don't know if some theoretical being exists that can beat him.
It turns out that there is a much easier way to figure out if he has solved chess than actually playing him. You can ask him a series of math questions (randomized based on his answers) that on the surface have nothing to do with chess that if he gets correct you can be 99.9% sure or higher in a few minutes that he is unbeatable. One the the important concepts is
zero knowledge proofs (The abstract example part of that is a nice intro to the concept imo)
This result is actually not specific to chess either. You can replace chess with a ton of other games or even pure math problems, like somebody claiming to have a proof of the Riemann Hypothesis. Even if we are all too dumb to understand it, or if the proof has too many symbols for a human to read in a lifetime, we can still be sure somebody actually has the proof if they can answer the questions we ask correctly. This is a hugely nontrivial result, just discovered in the last 20 years or so, so don't worry if it isn't perfectly clear, I don't fully get all the details of the proofs involved myself.
Last edited by Max Raker; 02-15-2011 at 12:20 AM.