OWL "Sydney Syntax"

Dear All,

 

Firstly let me say how much I enjoyed OWL-ED 2006 and the opportunity to
meet and converse with many of you in person. Thanks Bijan, Ian, Peter,
Bernardo and crew.

 

Following on from Manchester's lead and the proposed "Manchester Syntax"
which was presented at OWL-ED 2006,

there was some discussion as to the possibility of an extension of this
approach, using whole English sentences with a controlled vocabulary

to comprise a syntax which is more intuitively natural for non-logicians,
AND has a precise DL semantics.

 

During the OWL-ED discussion, I mentioned a project called PENG (Processable
ENGlish) as a possible basis for this approach.  PENG is the brainchild of
Rolf Schwitter who is currently based at Macquarie University in Sydney
Australia.  The PENG website is at:

 

http://www.ics.mq.edu.au/~rolfs/peng/

 

Quoting from the website:

"PENG is a computer-processable controlled natural language designed for
writing unambiguous and precise specifications and use cases. PENG covers a
strict subset of standard English and is precisely defined by a controlled
grammar and a controlled lexicon. In contrast to other controlled languages,
the author does not need to know the grammatical restrictions of the
language explicitly. ECOLE, a look-ahead text editor, indicates the
restrictions while the specification is written. The controlled lexicon
consists of domain-specific content words that can be defined by the author
on the fly and predefined function words that build the structural backbone
of the language. Texts written in PENG can be deterministically parsed and
translated into discourse representations structures and also into
first-order predicate logic for theorem proving."

Rolf has previously indicated an interest in collaborating with myself and
my NICTA colleagues in Sydney, and Thomas Meyer (my co-supervisor) and I
have contacted Rolf to discuss this and hopefully scope out what I am
currently referring to as a "Sydney Syntax" for OWL.  I expect this work
would be a subset of PENG's current capabilities, and the hardest part would
just be to make a nice interface.  Which brings me to the next point...

 

I very recently came across Avi Bernstein's work on "GINO" which may be
relevant for  this angle.  This was a paper presented at ISWC06 entitle
"GINO - A Guided Input Natural Language Ontology Editor" by Bernstein &
Kaufman available at: http://iswc2006.semanticweb.org/items/paper_41.php

 

At this stage, I am looking for feedback from list members as to what they
think of this proposal, would they support it and would they have any
interest in participating?*   I appreciate that time and resources for
working on OWL are in high demand, but I am willing to make the effort
personally at least to scope this out and I suspect it should be a case of
minimal effort for a really good return in terms of making OWL much more
accessible and thereby increasing takeup.  I'd also like to specifically ask
Bijan, as our overall "ring master" to respond with his thoughts about how
to move forward with this.

 

I look forward to your responses.

 

Best Regards,

 

Anne Cregan

National ICT Centre of Excellence Australia 

Sydney, Australia

 

*Just like to point out that Sydney is now going into beautiful warm summer
weather, and our NICTA Sydney office is only a couple of kilometres from
beautiful Coogee beach - sun, sand, surf, speedos, etc.   If you are
interested in being a NICTA visiting researcher for any amount of time
(days, weeks, months, etc) just let me know...