From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!news.funet.fi!sunic!ugle.unit.no!nuug!ifi.uio.no!sics.se!sics.se!torkel Thu Feb 20 15:20:11 EST 1992
Article 3670 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!news.funet.fi!sunic!ugle.unit.no!nuug!ifi.uio.no!sics.se!sics.se!torkel
>From: torkel@sics.se (Torkel Franzen)
Newsgroups: comp.ai.philosophy
Subject: Re: Real-life Turing test
Message-ID: <1992Feb12.105015.21892@sics.se>
Date: 12 Feb 92 10:50:15 GMT
References: <1992Feb10.235221.56220@spss.com> <1992Feb11.102903.7771@sics.se>
	<1992Feb11.195902.150987@spss.com>
Sender: news@sics.se
Organization: Swedish Institute of Computer Science, Kista
Lines: 18
In-Reply-To: markrose@spss.com's message of 11 Feb 92 19:59:02 GMT

In article <1992Feb11.195902.150987@spss.com> markrose@spss.com 
(Mark Rosenfelder) writes:

   >I tend to agree; but I think this shows that the Turing Test is 
   >too ill-specified.

  As has been pointed out by Stanley Friesen, this wasn't really a
Turing test (a fact that I missed). In the conversation quoted, the
topic was apparently restricted to "whimsical conversation", which may
well be regarded as a synonym for "semi-random drivel". So my previous
comment should have been that the test isn't a test unless any
ordinary question is allowed, and the person who performs the test
knows enough about computers not to be impressed by an output of wordy
"whimsicalities". Still, it's true that the specification is
imprecise, but this also applies to all ordinary tests for
understanding in human beings. I think it will be clear enough when a
system appears that passes something that might "reasonably" be called
a Turing test.


