Oh, god, the book. There will be a book.

Show Post

Okay, I think I see your point. But I would argue that it's impossible to get a representative sample of a Turing complete language, since by definition they can do everything that's computable and, via the Church-Turing thesis, can simulate anything that's physically possible.

Show Post

Yes, I think it's true that there are no non-cyclic BF programs of length <= 11. But we can just extend the experiment and get to them. For Rule 110, I modified the machinery, so no matter how long we run the exhaustive program search, we will stay limited to FSA programs.

Show Post

I mean that the machinery is universal, independent from the program length. If we take the program length into account, any limit on it makes any programming language automatically non-universal (because there are infinitely many computable functions).

Show Post

who becomes *proficient* in Brainfuck? you play with it for one afternoon, say "well that was amusing but mostly annoying," and then never think about it again

Show Post