3 ms·
Paul Graham is running out of ideas these days. (Partly joking). This goes on to my same discovery: Life itself is a constant process of incremental trial and
by csomar 9y ago
Paul Graham is running out of ideas these days. (Partly joking).
This goes on to my same discovery: Life itself is a constant process of incremental trial and error builds-up. Over a few billions years, you get something out of it (human being playing snapchat)
If you keep counting from 0 to an infinitely large number, at some numbers you'll "discover": Windows 7, Windows 7 Korean Language, Microsoft Office, OSX, and every other game that human designed and developed. Plus a bunch of other software from alternate realities that compile on our machines. Cool.
The process the human are doing is not much different (albeit one can argue that it is much more efficient than stupidly counting) "Everything" is already there. The new Audi A4? It was already "possible".
If there is one thing that will change the future, it is our discovery of the link between "abstract" and "real" concepts. They share some important points that I'd not be surprised that "real" and "abstract" are the same thing/continuum. Like space and time.
- edanm 9y agoThat's an interesting thought. Just want to point out that complexity theory from Computer Science fits nicely into this, because it asks, not what is "possible" (e.g. finding all programs by counting all the integers), but what is possible within limited resources, like e.g. time. A nice addition to the thought that, theoretically, even if you can find the Windows 10 source code by counting all integers, you can't actually do it within the lifetime of the universe without a smarter algorithm.
- csomar 9y agoMore like infinite resources. But there is a lifetime for the universe?
- hammock 9y agoBefore you lay one thousand blocks of marble. Inside every one is a world famous sculpture. Eventually you will find it...
- andrewflnr 9y ago> keep counting You can always find the successor of a given integer, but there's not really an analogous process for human discovery, where it's easy to go in circles. The process really is that different. Are there theoretical situations where random searches are guaranteed to be complete on infinite spaces given "infinite time" (probability of finding solution approaches 1)?