Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!haven.umd.edu!news.umbc.edu!eff!news.duke.edu!news.mathworks.com!newsfeed.internetmci.com!vixen.cso.uiuc.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!nagle
From: nagle@netcom.com (John Nagle)
Subject: Re: [??] Re: CYC performance
Message-ID: <nagleDKDDu7.LqD@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <4bqqva$212@tribune.usask.ca> <4c1c6s$ni6@villa.fc.net>
Date: Fri, 29 Dec 1995 22:44:31 GMT
Lines: 12
Sender: nagle@netcom11.netcom.com

"William D. Gooch" <goochb@rwi.com> writes:
>The real question, IMO, is "does CYC do *anything* useful
>reliably?"  I ask this in all seriousness, as I have been 
>out of touch with the CYC work for a while.  This is not
>a troll - I'm not in a position to comment meaningfully
>on the current status of CYC in any case.  

      Prof. Vaughn Pratt at Stanford wrote up an evaluation of Cyc
last year, after doing some tests.  It's on the Web; I don't have the
URL handy but you can probably find it with Lycos.

					John Nagle
