W3C home > Mailing lists > Public > public-xg-emotion@w3.org > January 2008

Re: OWL version

From: Bill Jarrold <jarrold@AI.SRI.COM>
Date: Wed, 30 Jan 2008 17:16:28 -0800
Message-Id: <DB144304-61CB-4E0A-AD81-787D29D1DA7B@AI.SRI.COM>
Cc: public-xg-emotion@w3.org
To: Ian Wilson <ian@neon.ai>


On Jan 30, 2008, at 6:15 AM, Ian Wilson wrote:

> Bill,
>
> At this point in time our investigations and examples are just to  
> show what a "typical" use case might look like (i.e. our  
> requirements use cases). The details are not too important (I don't  
> believe), we just need to see what an example would "look" like, to  
> get the flavor of the differences between an XML, RDF and OWL  
> representation. My own example just has dummy data in. So don't  
> worry at this point about the mechanics of any particular model.
>
> wrt your specific questions:
>
> a. Differing models - We are trying to accommodate as many models  
> as possible while being model agnostic. However there are 3 main  
> model types that are widely used and represented here, Catagory/ 
> Label models, Dimensional models (like my own) and Appraisal type  
> models. Ideally our representation would be flexible enough to  
> allow it to any variation of those types of model.

Okay.

Btw, It seems that under any of those three schemes we might want to  
start off representing them as occurants ("things that  happen")as  
opposed to continuants ("things that are").

>
> Our use cases were a way to find who the languages users would be  
> and what *their* needs might be so we can ensure we are building  
> something people may actually use and need.
>

Yes.  Ontologists like me can sometimes go overboard with the desire  
for generality and expressivity.

Btw, I feel unclear if there is an important distinction between  
markup versus ontology RE the goals of the EMOXG.

> b. "Do the other annotation options (e.g. XML or RDF) allow for  
> namespaces?" : Yes for RDF, it is a central part of the rdf idea.  
> XML also has name spaces (xmlns).

Ah, thanks.  So, if XML or RDF did not allow for namespaces then that  
might make our choice of markup language clearer.  So, if it turns  
out that we wanted to use namespaces to implement encapsulation of a  
given approach (e.g. Sherer vs Douglas-Cowie) we still have the  
freedom to chose between OWL, RDF, or XML.

>
> c."Can we import between xml files? Can we import between rdf  
> files?" : Yes and Yes. This will be an important part of our effort  
> as we wish to inter-operate with out languages (for example SMIL I  
> believe, see June 2007 posts for more details).

Okay thanks.

Bill

>
> Best,
>
> Ian
Received on Thursday, 31 January 2008 01:16:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 31 January 2008 01:16:49 GMT