From: Randall Randall (randall@randallsquared.com)
Date: Tue Oct 13 2009 - 13:58:49 MDT
On Tue, Oct 13, 2009 at 08:31:46PM +0100, Mu In Taiwan wrote:
> If you assume that 'in general' means "For all X...", why would JC use
> phrases such as 'given algorithms' and 'random programs'? These distinct
> phrases seem to imply he is referring to predictions about samples, rather
> than a universal claim.
I don't know. It seemed that there was a way to interpret the post
such that it was consistent. Perhaps I'm wrong.
> Finally, consistency between JC's posts may not be important anyway, given
> that they are simply factually inaccurate: JC claims: "what Turing proved is
> that in general the only way to know what a random program will do is to
> watch it for eternity and see". This is absolutely not what I see AT
I'm inclined to give Mr. Clark the benefit of the doubt where possible,
considering his reputation on, for example, the Extropians list of the
1990s. I do think it's still an open question whether a fixed hierarchy
of goals is workable for a GAI.
-- Randall
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:04 MDT