1. I’m watching Flash Forward on ABC.
I read the novel too. It’s not Robert Sawyer’s best novel, but it’s okay. (I haven’t read all Sawyer’s books, so I can’t say which is the best exactly — but Flash Forward definitely isn’t.)
Here are a couple of interesting passages from Sawyer’s Flash Forward (pp. 166-7) that reference Tipler’s Omega Point:
…Tipler says there are 110,000 genes that make up a human being. That means that all possible permutations of those genes…amount to about ten to the tenth to the sixth people…[and] you could reproduce all possible humans that could ever exist, and all possible memories that they could ever have, in ten to the tenth to the twenty-third bits.
Ten to the tenth to the twenty third is a very, very large number. But it is, nevertheless, a finite number: a number with a limit.
And here are some more, not quite so large but equally interesting numbers, from an article published in 1963 by John Senders…
Three calculations of the number of different things stored in the world’s libraries yield estimates from 7.5 x 107 to 7.7 x 108. At 105 words per volume, five letters per word, and 12 bits per letter, the information capacity used for storage is between 4.6 x 1014 and 4.6 x 1015 bits, and is increasing at about 2 x 106 bits per second.
These are, again, very large numbers. Once upon a time, both sets of numbers might have been inconceivably large. Now, however, the last set of numbers — the number of bits in the world’s libraries — is not only conceivable, it is manageable.
2. The most recent round of Google Book Search Project settlement negotiations were completed on Friday, November 14, just before the midnight deadline set by the judge overseeing the settlement.
The idea behind the Google Book Project is to make all books in all libraries available to the public. As ideas go, it’s a very good one. Lawyers, guns, and money have cut that idea back a bit –although not yet erased it as a good idea or as an achievable goal.
I am now wondering, however, based on the numbers above, which will come first: The ability to search through the accumulated knowledge of all mankind with something like the Google Book Search Project OR the ability to generate all possible knowledge that could ever exist, from scratch.
3. The first possibility — cataloging all available human knowledge — would be, in chess, sort of like taking all chess games that have ever been played and creating a index of those chess games that can then be searched for good moves and bad moves. We more or less have something like that right now. It looks something like this: Current Index of All Chess Positions.
The second possibility — generating all possible human knowledge — would be, in chess, sort of like taking the rules of chess and then, based on those rules, generating all possible chess moves that could ever be played. We are now working on doing just that. It’s called a brute strength method of playing chess, and chess programs that use this brute strength approach, beginning with Deep Blue, include Rybka, the current world chess champion.
Rybka can refer to a large database of chess games and chess moves, as necessary. More critical to its ability to play chess, however, Rybka can generate lots and lots of chess moves that have never been played. And, at least potentially, chess programs like Rybka may be able generate all chess moves that ever will be played.
The rules for generating all possible human knowledge are, of course, likely to be a bit more complex than the rules for generating all possible chess moves.
But, if those rules are determinable and, most importantly, if those rules are finite, then there seems to be little doubt that those rules can and will be used to generate knowledge. And, given enough time, that generation of knowledge can and will become complete.
It’s going to take a while, of course, because there are a lot of bits involved. But then the Google Book Search Project is taking a while too, not only because of all the bits, but because of all the lawyers and the money.
So here’s a question to ponder: Which sorts of problems are likely to be solved sooner — the inconceivably large bit problems or the incredibly obtuse legal problems?
Or, an even better question: Which sorts of problems, when and if solved, make the other sorts of problems irrelevant?