[MURG] Re: X-Prize

[Eric Zilli]

On Wed, 20 Oct 2004 23:50:29 +0800, Major <murg at audry2.com> wrote:
> 
> Sim Bamford <sim at plaything.co.uk> writes:
> 
> > 2) A panel of judges will get to know the individual prior to
> > scanning. They will then be able to cross examine the individual
> > and the simulation by video link. They will not be able to guess
> > which is the simulation.
> 
> Anyone can guess. A better wording might be "They should cross examine
> the individual and the simulation by video link and be asked to
> identify which is the simulation and which is the individual. Their
> answers should not be more accurate that chance."
> 
> Major
> 

Perhaps we should make the analogy to the Turing test explicit.
Also, I'm not sure the judges should communicate with the copy and the
original by video. The simulation would likely be easily visually
identified. I think delayed text communication would be best (delayed
so that the slow human responses and fast simulation responses [or the
other way around depending on how fast the computer is] seem to come
in the same amount of time).

-- 
Eric Zilli
Hasselmo Lab - Computational Neurophysiology
Center for Memory and Brain
Boston University
2 Cummington St.
Boston MA, 02215
digfarenough at gmail.com -- www.digfarenough.com