Newsgroups: comp.lang.smalltalk,comp.lang.c++,comp.object,comp.client-server
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!news.kei.com!news.mathworks.com!tank.news.pipex.net!pipex!sunsite.doc.ic.ac.uk!susx.ac.uk!news.bton.ac.uk!Simon.Shurville
From: sjs16@bton.ac (Simon John Shurville)
Subject: Problem with mass-nouns in OOA/OOD/OOP
X-Nntp-Posting-Host: tomcat.itri.bton.ac.uk
Message-ID: <DJ27wK.6p3@bton.ac.uk>
X-Posted-From: InterNews 1.0@diamond.bton.ac.uk.
Lines: 19
Sender: -Not-Authenticated-[3206]
Organization: The Composer Project
References: <30AF370B.1AE@vmark.com> <jzqcvc@bmtech.demon.co.uk> 
 <30B246CB.1144@vmark.com> <48vfqc$l0f@jobes.sierra.net> 
 <RMARTIN.95Nov23085040@rcm.oma.com>  
 <30be02d2.9841218@netnews.voicenet.com>
Date: Mon, 4 Dec 1995 11:29:07 GMT
Xdisclaimer: No attempt was made to authenticate the sender's name.
Xref: glinda.oz.cs.cmu.edu comp.lang.smalltalk:31497 comp.lang.c++:163228 comp.object:41958 comp.client-server:14238

In article <30be02d2.9841218@netnews.voicenet.com>
jgerber@omni.voicenet.com (James Gerber) writes:

> Maybe people need a more robust system than Smalltalk, which still clings to
> the obviously false idea that the world can be modeled by single inheritance.

Can the (whole) world or the world of language be modelled in an OO way
at all? I attended a philosophy seminar recently where the speaker
argued that it was hard to represent mass-nouns, e.g., fire, water,
death, love, in logic or set theory. That argument set me to wondering
how one could model such mass-nouns in an OO way. I have yet to find an
answer. I realise that this might not cause application programmers to
miss much sleep. However, I think that it is an interesting problem in
knowledge representation at the pure AI end of the Smalltalk
spectrum. Does anyone have any ideas or pointers to some literature?

And please, please don't imagine that I am language-bashing or
paradigm-bashing. I am just perplexed and fascinated.


