Re: Computability theory question
- From: paetz.spam@xxxxxxx (Thomas Paetz)
- Date: Thu, 11 Oct 2007 18:27:07 +0000 (UTC)
Snis Pilbor <snispilbor@xxxxxxxxx> wrote:
Fix a Godel numbering of the computable functions, say phi_n is the
nth computable function.
Let superscripts denote iteration: phi_n^1(i)=phi_n(i), phi_n^j(i) =
phi_n(phi_n^{j-1}(i)).
Is it possible that there is some n, and some k>1, such that for every
i,
phi_n^k(i) = phi_n(i)? (In particular, phi_n is total)
If the answer is "no", then I can prove something nice.. =) Not
homework related.
Hmm, I'm afraid this is very possible: just let n be an index for the
identity function... Sorry...
--
Cheers, Tom
*** PGP Key available at <paetz.sdf-eu.org> ***
.
- Follow-Ups:
- Re: Computability theory question
- From: Snis Pilbor
- Re: Computability theory question
- References:
- Computability theory question
- From: Snis Pilbor
- Computability theory question
- Prev by Date: Re: Second order arithmetic
- Next by Date: Re: Second order arithmetic
- Previous by thread: Computability theory question
- Next by thread: Re: Computability theory question
- Index(es):