[gurps] Failing autopilots and RVO

[David Scheidt]

On Wed, Mar 4, 2009 at 1:16 PM, Onno Meyer <Onno.Meyer at gmx.de> wrote:
> Travis replied to me:
>> I see no reason that the proof is only for machines, it also applies
>> to methods or procedures that involve one or more people.
>
> If those people follow an algorithm. Humans can decide not to
> do that. If computers deviate from their program, that must be
> programmed to start with.
>

If the human isn't following an algorithm, they're merely guessing.
Merely guessing isn't a proof, so they haven't solved the problem.

>> If you require a solution to the halting problem as a proof of
>> intelligence, then I can easily prove that intelligence cannot exist
>> by your definition.  I would consider that proof to be sufficient to
>> make your definition useless.
>
> I think the halting problem is just one example for a class of
> problems where computers (as we know them) fail and humans can
> succeed (or not).

It's a an example of a problem that can't be solved by computation,
whether that computation is mechanical or biological.  If you hold
that mentation isn't computation, then they might be solvable by that
extra-computational mentation.  But then you've got to demonstrate
that mentation isn't computation, or you're merely throwing up your
hands and saying "It's magick!".

-- 
David Scheidt
dmscheidt at gmail.com