W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2007

Re: owl xml serialization

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Tue, 10 Jul 2007 16:07:43 +0100
Message-Id: <6F13FB5F-6CED-4954-91F6-C8553298F5D3@cs.man.ac.uk>
Cc: public-owl-dev@w3.org
To: Matthew Pocock <matthew.pocock@ncl.ac.uk>

On 10 Jul 2007, at 15:45, Matthew Pocock wrote:

> Hi,
> I've reached the point where I'm writing the xml serializer/ 
> deserializer for
> my haskell owl 1.1 library. I've read through the syntax document  
> [1] and
> also through the xsd but these alone don't give enough info to be  
> able to
> work out exactly what goes where.

Not to mention that the XML syntax documents examples are  
horrifically unfriendly, what with qnames from hell scattered about.  
I find it much pleasanter to read/write with a default namespace for  
the syntax.

> In particular, I couldn't find examples in
> these docs about how to treat data values or annotations.
> I'm resorting to generating little bits of OWL 1.1 in protege4 and  
> then making
> my code round-trip to this. Not ideal, as I'll end up reproducing  
> any bugs in
> the OWL 1.1 library used by protege4.
> Is there a normative 'how to roundtrip the uml/functional syntax  
> with xml'
> document?

No, but there should be. I keep meaning to take up the XML syntax  
document which was quite neglected, but other things intervene.  
Presuming there will be a WG, I would imagine this is something they  
could take up. For now, if you are taking notes on what you're doing  
you could either file them as issues or you could set up a wiki page  
(or a side document) to keep track on:

I have a sneaky idea on how to specify this...

Received on Tuesday, 10 July 2007 15:06:38 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:58:15 UTC