This is not yet another "whether chess can be solved..." thread. Rather this is hypothetical novel ridden scenario. I'd originally posted this in Rybka forum. But I'd like to post it here as well since there are people who are mainly CCC members, and don't visit Rybka forum often.
Hypothetical Situation - The Year is 2015. And you're Vas.
So, Can you DARE take up this challenge...?
Code: Select all
Let's say you're a 50 elo away from 'solving' chess. Beyond which there's an oblivion. And nothing could be improved further on no matter how hard you tried.. even years after.
'solving' is ambiguous word. But let's set up the limit. Let's stipulate that 4000 on a comp chess rating scale is the maximum. So in distant future you find yourself in a situation where your engine couldn't be improved further on... and you've tried real hard slogging days and nights working on it, and still you find your engine stuck at 3950 even after 5 years of trying. You implement different concepts, speed up or faster hardware, genuine algorithm or something, but it's still stuck at 3950. You find that the nearest competitor is at 3400 and he had already long since given up any attempts to imrpove his program. That gives you the sense of freedom. You'd no longer be pulled in to elo-hunting rat-race ever. You'd no longer feel pressured.
Even with higher advancement in hardware department, you find that your engine is independent of the hardware speed or programming techinques. It still plays the same with faster hardware or slightly less faster hardware. Hardware issue is deemed irrelevant. It is of no help. And people continue reporting some obscure bug that program doesn't understand certain pattern in positional chess. They post several different positions where that bug in the program is evidenced. And you find that this is the *only* thing to do to finish solving chess beyond which there would be no conceptual mistakes encountered. Bug in positional pattern... that even humans could easily avoid making that positional mistake...well, strongest GM, for that matter. But your program couldn't not, no matter how much you've tried to overcome this defect.
After years of trying, you become frustrated and you decided to make an announcement that it is hard to defeat the bug, but you plan to do it anyway... people, mainly chess enthusiasts who play active chess all over the world, upon hearing your statement, stage internet-strike. They make campaigns while at the same time persuade you to kindly give up improving the program further, for the sake of them and the thousand year old (chess) board game being alive and played over and over in the future. They don't want you to destroy chess. They don't want you to 'solve' it and throw it the checkers-way. They want you to give up - and leave the game- They want you to exit - The Dark Knight- style. They would be really happy if you did that.
In this situation, What would you then?
- (A) Release the source code to the public so it could be improved upon further by group of new programmers - like the way it is done with Fruit derivatives (Toga etc).
(B) Stop programming, with the engine still stuck at 3950, with 50 elo away from solving chess. And make thousands of chess enthusiasts all over the world happy.
(C) Hire a competitive programmer to work as a team, to find the way to solving the mystery behind the missing 50 elo and overcome the strange positional bug.
(D) Keep trying despite knowing that it is hopeless. Hoping beyond hope that you'd find the secret mystery surrounding the missing 50 elo, and later if, provided that ever happens, take credit once you found it.
(E) Declare the game unsolvable and Close the project and move on to some other software field, thereby saving your time spent on chess.
(F) Other - Specify what.
On a sidenote, I've gotten quite a few messages from friends asking me to return back to forum and start being active here or organize test tournaments. etc If anybody is ever going to wonder about me, I'm taking a break from computer chess but I can't make that promise. I could occassionally make posts when something catches my attention and prompts me a reply. Anyway, so much stuff happening in real life, I hardly find time to hang around here. In Spare time lately, I have been watching movies more often rather than test engines or read message boards. Ofcourse, I will be there at CCT's operating the engine Bright, also I could work on test suite alongside Dann Corbitt, in the background when I find time.
So see ya later, people!
Regards,
Swami
