W3C home > Mailing lists > Public > www-webont-wg@w3.org > March 2003

Re: DL Syntax proposal B.1 and B.2

From: Ian Horrocks <horrocks@cs.man.ac.uk>
Date: Fri, 14 Mar 2003 12:08:22 +0000
Message-ID: <15985.50742.451427.507871@merlin.horrocks.net>
To: Sean Bechhofer <seanb@cs.man.ac.uk>
Cc: www-webont-wg@w3.org

On March 14, Sean Bechhofer writes:
> 
> 
> I'd like to understand a little more about the ramifications of B.1 and
> B.2 as detailed in [1] and discussed briefly in the telecon last night.
> 
> As I understand it the proposal basically says that I can "reuse" bnodes
> that correspond to concept descriptions if those descriptions are the
> same. I assume here we're talking about them being *exactly* the same,
> i.e. if I have:
> 
>  intersectionOf( A B )
> 
> and
> 
>  intersectionOf( B A )
> 
> then I wouldn't be able to do this anyway.
> 
> So what does it mean to me in implementation terms?
> 
> If I'm producing triples from some internal data structure corresponding
> to abstract syntax constructions, then I can save myself a small amount of
> work by only doing the triple generation once for each description and
> then possibly reusing those elsewhere.
> 
> Conversely, when I'm trying to read a bunch of triples and determine the
> OWL structure, B.1 and B.2 reduce the amount of work that I have to do so
> see whether I've got valid OWL-DL, as I don't need to check that it's tree
> like and that description bnodes aren't reused.
> 
> Is this a fair comment and/or am I missing some important subtleties?

I believe that this is a fair characterisation of B.1 and B.2. 

From an implementation point of view, I don't think that there should
be any problem - when generating triples you would have the choice of
reusing or not, and when parsing triples there is no reason why you
would even notice the reuse (in fact, as you point out, it would be
extra work to check that no such reuse had occurred).

My concern was with respect to semantic layering considerations, and
the fact that Peter mentioned further changes to the proofs. If,
however, Peter is confident that this isn't a significant problem,
then I would be happy to support B.1 and B.2.

Ian




> 
>    Sean
> 
> [1] http://lists.w3.org/Archives/Public/www-webont-wg/2003Mar/0066.html
> 
> -- 
> Sean Bechhofer
> seanb@cs.man.ac.uk
> http://www.cs.man.ac.uk/~seanb
> 
Received on Friday, 14 March 2003 06:07:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:58 GMT