From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!yale.edu!jvnc.net!darwin.sura.net!europa.asd.contel.com!uunet!mcsun!uknet!edcastle!aisb!jeff Tue Jan 28 12:17:22 EST 1992
Article 3118 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!yale.edu!jvnc.net!darwin.sura.net!europa.asd.contel.com!uunet!mcsun!uknet!edcastle!aisb!jeff
>From: jeff@aiai.ed.ac.uk (Jeff Dalton)
Newsgroups: comp.ai.philosophy
Subject: Re: Table-lookup Chinese speaker
Message-ID: <1992Jan24.151338.4883@aisb.ed.ac.uk>
Date: 24 Jan 92 15:13:38 GMT
References: <11828@optima.cs.arizona.edu>
Sender: news@aisb.ed.ac.uk (Network News Administrator)
Organization: AIAI, University of Edinburgh, Scotland
Lines: 40

In article <11828@optima.cs.arizona.edu> gudeman@cs.arizona.edu (David Gudeman) writes:
>In article  <1992Jan23.220442.24200@aisb.ed.ac.uk> Jeff Dalton writes:
>]In article <1992Jan22.205804.39265@spss.com> markrose@spss.com (Mark Rosenfelder) writes:
>]>tester says, but merely a few particularly good ones.  (It can't store
>]>just one response, or we would notice it making exactly the same responses
>]>in repeated runs, which be a failure of the Turing test.)
>]
>]I think that's a good point.  If we're allowed repeated tests,
>]and they're suspiciously the same, then we'd be, well, suspicious.
>
>No it isn't a good point.  "Repeated runs" are not part of the
>experiement.  If you want to experiment today and again tomorrow, then
>you have to come back tomorrow and start at the point where the
>conversation left off.  Even if that point is "good bye", it will
>still have a history of what has happened before.

I agree completely that the Turing Test doesn't involve repeated
runs.  However, I don't think the TT is a very good test.  For me
it's a "good point" against the TT to say that repeated runs might
detect something that the TT doesn't.

I also agree completely that the system could have a history of
what came before.

However, _if_ repeated runs _were_ suspiciously the same (perhaps
because the system didn't remember, had been rebooted, or whatever),
then we might well be suspicious.  Or suppose all the machines from
the same factory made the same response.

That sort of thing may not be part of the Turing Test, but it 
would certainly be evidence we ought to consider if we were 
trying to determine if machines were conscious (or whatever).

When I made such remarks in my earlier message, I was thinking
about machines in general, not specifically the lookup machine.
Perhaps I should have made that more explicit.  Moreover, I
don't think the lookup machine will necessarily do worse than
other computers in those respoects.

-- jd


