News for hackers. Stuff that matters.

A comment on Non-computability of solutions of certain equations on digital computers (2022). Parent comment

Re:Non-computability of solutions of certain equations on di... by ogogmad on Sunday August 04, 2024 @11:33AM (#41154181)
> A much simpler version is f'(x) = {-1 if x<0; +1 if x>0; Chaitin's constant if x=0}. f(x) = abs(x).

This is wrong. The function f you've defined is not continuously differentiable. The one in Myhill's paper is.

[ Parent | on HN ]
Re:Non-computability of solutions of certain equations on di... by a1369209993 on Wednesday August 07, 2024 @03:39PM (#41184610)
Evidently I should have been more explicit:

> > A much simpler[-and-more-nitpickable] version

Myhill uses a parlor trick to make his version continuously differentiable. It's a pretty good parlor trick, but it doesn't have much to do with computability.

[ Parent | on HN ]