RE: OWL "Sydney Syntax", structured english

Kendall, Dan, Kaarel and everyone,

 

I agree with Kendall:

 

> I'd prefer the WG standardize the concise,

> scribblable sufarce syntax than the quasi-natural language

> presentation syntax, if it couldn't do both. I think the quasi-

> natural thing is *much* harder and should be outside the scope of OWL

> 1.next.

 

This is what we would be aiming to do with a "Sydney Syntax" for OWL, and
what I intended this thread to be about when I started the discussion.  

 

In Sydney, we have now had meetings with Rolf Schwitter (creator of PENG)
and he is keen to contribute to our effort.  We are aiming to prepare
something in time for the next OWL-ED workshop.  It seems that there are
already several tools around that are up to the task, or would be, with a
little adaptation, and the hard question is to decide what the syntax should
look like.

 

Judging from the responses, there seem to be quite a few people who are
interested in pursuing the idea, or at least something in a similar space,
and several who have pointed to existing work which is potentially very
helpful.  

I will try to get around to thanking you all individually for your comments
and pointers, but let me thank you all now collectively: Thanks :-)

 

I think it's fair to say that what has come out of the dialogue is no
surprise: everyone has their own particular slant on requirements and
priorities for such a syntax.  Some choices, like the choice between
quasi-natural vs concise, controlled syntax, are obviously at tangents,
whilst other options will require some analysis to determine whether they
are ultimately compatible and if not, decisions will need to be made
regarding which path to pursue.

 

If the general consensus is that there should be a W3C-recommendation around
this (I personally support this approach, and have the feeling that most
people would be prepared to at least give it a go), then perhaps it is time
to ask the W3C to comment on formulating the requirements and co-ordinating
the activity. 

 

** That said, I would like to invite Dan Connolly from the W3C to comment on
what he would see as the next steps for moving towards a W3C-endorsed
standardization on some kind of controlled English syntax for OWL. **

 

I'm pretty new to W3C processes, but in my opinion, the first order of
business would be to determine what "OWLglish" or whatever we end up calling
it should look like, and how it should behave.  Sifting through the
postings, several design considerations/requirements have already come up.
I would suggest as a starting point, that we collect these into a list, and
post it somewhere on the W3C site (if Dan agrees) or else at some other
convenient location, to provide a basis for focused discussion, debate and
tasks.

 

 

Regards,

 

Anne Cregan

NICTA Sydney Australia

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

> -----Original Message-----

> From: Kendall Clark [mailto:kendall@monkeyfist.com]

> Sent: Wednesday, 29 November 2006 8:09 AM

> To: Dan Connolly

> Cc: Kaarel Kaljurand; Anne Cregan; public-owl-dev@w3.org

> Subject: Re: OWL "Sydney Syntax", structured english

> 

> 

> On Nov 28, 2006, at 2:47 PM, Dan Connolly wrote:

> 

> > I agree that rendering to a constrained dialect of English is an

> > interesting user-interface technique;

> 

> For my money, using quasi-natural language as UI is orthogonal to a

> more human-friendly syntax for writing OWL (and for some display

> contexts). What I'd like to see a new OWL WG standardize is something

> that plays a role analogous to the role N3 (well, Turtle, really)

> plays for RDF. Something like Manchester OWL Syntax is more like an

> "N3 for OWL" than one of these quasi-natural language things.

> 

> The market needs both, but I'd prefer the WG standardize the concise,

> scribblable sufarce syntax than the quasi-natural language

> presentation syntax, if it couldn't do both. I think the quasi-

> natural thing is *much* harder and should be outside the scope of OWL

> 1.next.

> 

> Cheers,

> Kendall