W3C home > Mailing lists > Public > semantic-web@w3.org > May 2007

Re: [Semantic_Web] Ontology Vs Semantic Networks

From: Martin Hepp <mhepp@computer.org>
Date: Wed, 30 May 2007 10:34:48 +0200
Message-ID: <465D3728.2000908@computer.org>
To: Dan Brickley <danbri@danbri.org>, Alan Ruttenberg <alanruttenberg@gmail.com>, Danny Ayers <danny.ayers@gmail.com>, semantic_web@googlegroups.com, SW-forum <semantic-web@w3.org>

 >> how about:  How have the terms "ontology" and "semantic network" been
 >> used historically?

In this respect, I would like to stress that Information Systems as a 
discipline and Computer Science have been using “ontologies” in slightly 
incompatible ways for more than a decade now (unfortunately, this has 
remained largely unspotted so far). This causes IS ontology papers to be 
frequently rejected by Semantic Web communities and vice versa.

IMHO, for most Information Systems researchers, the constituting 
property of ontologies has been the analysis of and consensus on the 
ontological nature of conceptual elements in a domain of discourse, in 
whichever way this consensus was represented, whereas in Computer 
Science, a large share of researchers regards as an ontology only such 
specifications of conceptualizations that excludes unwanted 
interpretations (i.e. invalidate unintended world models) my means of logic.

A well established example for the different notion of the term 
“ontology” in Information Systems research is the REA ontology project 
(http://www.msu.edu/user/mccarth4/rea-ontology/). Since the 1980s, this 
has been used in Accounting Information Systems research as a conceptual 
framework for various purposes. The exclusion of unintended models by 
means of an axiomatic specification and in general a specification of 
that model to be used for computation has only recently gained 
attention, in particular by the ongoing PhD work by Frederik Gailly 

It is a curio that two very related communities (often submitting to the 
same journals) have been using partly incompatible notions of the term 
ontology for so long, and have not even spotted that they are talking of 
something different (there are edited books around that include papers 
from both communities, which, side by side, speak of completely 
different things when they use the word "ontology").

Just my lengthy 2 cents ;-)


Dan Brickley wrote:
> Alan Ruttenberg wrote:
>> I think, in the direction Danny has started, that you would be better 
>> off considering the question differently than as posed.
>> Instead of:
>>>>> How can we distinguish between ontologies and semantic networks, and
>>>>> in what respects are they similar.
>> how about:  How have the terms "ontology" and "semantic network" been 
>> used historically?
> This is a very healthy reformulation of the question. If more people 
> thought this way, ... there'd be fewer wars and everyone would 
> understand each other 100x better. Well, I exagerrate, but only a bit :)
> Dan
Received on Wednesday, 30 May 2007 08:35:13 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 08:45:00 UTC