W3C home > Mailing lists > Public > public-owl-wg@w3.org > April 2008

RE: Proposal to close ISSUE-82: UML diagrams

From: Boris Motik <boris.motik@comlab.ox.ac.uk>
Date: Wed, 16 Apr 2008 20:25:29 +0100
To: <conrad.bock@nist.gov>, "'Alan Ruttenberg'" <alanruttenberg@gmail.com>, "'Elisa F. Kendall'" <ekendall@sandsoft.com>
Cc: "'Peter F. Patel-Schneider'" <pfps@research.bell-labs.com>, <public-owl-wg@w3.org>
Message-ID: <002d01c89ff7$a1cc2bf0$d012a8c0@wolf>

Hello,

I agree with Conrad here. I have had many quite positive comments from users and, in particular, from developers of OWL tools about
the style in which the OWL 2 specification has been presented. The object model in the specification can be almost directly cast
into an OWL 2 API; furthermore, aspects of the specification such as various consistency criteria (i.e., the general restrictions on
axioms) can be defined in a much easier and more precise way in terms of an object model. Therefore, I really believe that we should
keep the diagrams.

I also agree that the diagrams themselves should be written in the correct UML. The present errors were unintended; however, I
haven't fixed them yet because I wanted to get a chance to talk to UML experts. In particular, I was hoping that either Elisa or
Conrad would be present at the last F2F meeting, so that we could have a brief chat. I have a few technical questions about UML that
I would like to ask before I go and modify the diagrams.

Regards,

	Boris

> -----Original Message-----
> From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org] On Behalf Of Conrad Bock
> Sent: 16 April 2008 20:01
> To: 'Alan Ruttenberg'; 'Elisa F. Kendall'
> Cc: 'Peter F. Patel-Schneider'; public-owl-wg@w3.org
> Subject: RE: Proposal to close ISSUE-82: UML diagrams
> 
> 
> Alan,
> 
>  >  Speaking as someone who is not UML savvy, I can say that they also
>  >  create confusion if you don't use UML. Frankly, unless there is
>  >  compelling evidence that these are useful to segment of our users, I
>  >  would consider dropping them altogether.
> 
> That would be a separate issue.  Since UML is so widely understood in
> the software community, and appears even in the ISO Common Logic spec,
> I'd say the compelling evidence should be needed for dropping them, not
> for keeping them.
> 
> Conrad
> 
Received on Wednesday, 16 April 2008 19:27:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 16 April 2008 19:27:49 GMT