From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!haven.umd.edu!darwin.sura.net!convex!mips.mitek.com!spssig.spss.com!markrose Thu Feb 20 15:22:18 EST 1992
Article 3884 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!haven.umd.edu!darwin.sura.net!convex!mips.mitek.com!spssig.spss.com!markrose
>From: markrose@spss.com (Mark Rosenfelder)
Newsgroups: comp.ai.philosophy
Subject: Re: Look-up tables
Message-ID: <1992Feb19.234103.63728@spss.com>
Date: 19 Feb 92 23:41:03 GMT
References: <299C3CC6.15601@orion.oac.uci.edu> <1992Feb17.175002.38249@spss.com> <1992Feb19.200727.26155@neptune.inf.ethz.ch>
Organization: SPSS Inc.
Lines: 11
Nntp-Posting-Host: spssrs7.spss.com

In article <1992Feb19.200727.26155@neptune.inf.ethz.ch> santas@inf.ethz.ch (Philip Santas) writes:
>Note that your set of conversations includes silly ones too 
>ie. conversations that cannot pass the Turing test.

Of course it does; but these need not be entered into the table.

This was discussed ad nauseam about a month ago.

Suffice it to say that we enter conversations into the table such that the
machine's statements are all reasonable; we put no constraints on the
tester's statements.


