[Top] [All Lists]

Re: [ontolog-forum] Musen's presentation & relations with "good" softwa

To: "[ontolog-forum] " <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Mark Musen <musen@xxxxxxxxxxxxxxxx>
Date: Sat, 11 Dec 2004 12:03:26 -0800
Message-id: <B8024300-4BAF-11D9-8AEC-0003930B01C0@xxxxxxxxxxxxxxxx>
On Dec 10, 2004, at 10:13 AM, Nicolas F Rouquette wrote:
> My hunch is that the similarity between "sound" software engineering 
> principles
> and "formal ontology" development is much stronger than one might 
> acknowledge.
>    (01)

I agree with the points that you are raising.  I personally like to 
think of ontologies as software components, but this view clearly is 
not universal.  Much of the "language wars" of the past few years 
concerning the "appropriate" level of expressiveness for ontology 
languages points to the tension in the community.  As I mentioned in my 
talk, the folks in the trenches tend to be reaching for impoverished 
knowledge representation systems that meet their immediate needs (e.g., 
DAG-Edit) whereas there are folks from the logic community who would 
like to promote undecidable representations in the name of maximum 
expressiveness.  I've heard certain logicians claim that the complexity 
of the representation system doesn't matter, since ontologies are meant 
for computers, not people.    (02)

But just as software engineers do not choose to program Turing 
machines, ontology builders are seeking representation systems and 
tools that have maximum cognitive tractability for them.  That's why 
DAG-Edit has become so popular among biologists, even though the 
resulting ontologies are rather crippled.    (03)

The Apelon tool that NCI currently uses for ontology development does 
not enforce scoping or conventions of software-engineering of 
object-oriented systems.  But I agree strongly for the need for that 
kind of development support, and we increasingly are building those 
kinds of capabilities into our Protege and PROMPT systems.    (04)

Although the OMG crowd does not use the "O" word, I think that as their 
model-driven architecture approach gathers strength, and as people 
realize that are reasons for modeling application areas beyond the 
restrictions of UML, we will see even greater overlap between 
traditional software-engineering and the development of ontologies as 
our community thinks of them.    (05)

Thanks again for asking me to talk on Thursday,
     Mark    (06)

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (07)

<Prev in Thread] Current Thread [Next in Thread>