Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.sprintlink.net!howland.reston.ans.net!ix.netcom.com!netcom.com!whitten
From: whitten@netcom.com (David Whitten)
Subject: Re: NetCYC?
Message-ID: <whittenDA8DBL.5KA@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
References: <3rnie0$nlh@news.cs.tu-berlin.de> <3rpd0o$gb1@vixen.cso.uiuc.edu>
Date: Thu, 15 Jun 1995 20:09:20 GMT
Lines: 25
Sender: whitten@netcom20.netcom.com

Dan Putnam (dputnam@firefly.prairienet.org) wrote:
: Maximilian Spring (sepp@cs.tu-berlin.de) wrote:

: On the other hand, I also see problems in taking an open approach
: to knowledge-entering.  First of all, the same people who write
: viruses would delight in seeing how confused they could make the
: knowledge base.  The people who spam the net with commercial messages
: would figure out some way to inculcate the knowledge base with
: a false sense of the importance of whatever garbage they are trying
: to purvey.  I also worry about the "fringe" elements on the net,
: whose ideas are of questionable quality.  And how would the knowledge
: base engine be able to sift through the tens of thousands of
: mutually contradictory religious and political doctrines that
: would no doubt be the FIRST areas of input?

As I understand the CYC project, it is able to handle mutually contradictory
facts, it is just necessary to have a consistent context (a set of facts and
rules).  Guha did some work on taking facts that are known to be true
in one context and referring to them in a different context.

There is also support for models in CYC, which are explicitly intended to
store information that there may be a differing opinion or perspective 
available to interpret the data.

David (whitten@netcom.com) (214) 437-5255
