Re: what RDF is not (was Re: RDF/XML Syntax Specification (Revised) W3C Working Draft published)

From: Brian McBride <bwm@hplb.hpl.hp.com>
Subject: Re: what RDF is not (was Re: RDF/XML Syntax Specification  (Revised) W3C Working Draft published)
Date: Thu, 20 Dec 2001 20:46:21 +0000

> At 14:39 20/12/2001 -0500, Peter F. Patel-Schneider wrote:
> >From: Brian McBride <bwm@hplb.hpl.hp.com>
> >Subject: Re: RDF/XML Syntax Specification (Revised) W3C Working 
> >Draft     published
> >Date: Thu, 20 Dec 2001 18:30:29 +0000
> >
> >How is your solution any different from the solution to a related problem 
> >below?
> 
> I don't know.
> 
> 
> >As far RDF is concerned both solutions depend on magic.
> 
> That puzzles me, as I'm sure I can implement the solution I had in mind, so 
> it can't be that magical.

As far as RDF is concerned both your solutions might as well be magic.
Your second solution, of course, does *result* in the appropriate RDF, but
the mechanism it employs is completely outside of RDF.

> As far as I can figure, Mike is considering building an application using 
> RDF.  This is good news, and to the extent that I can, I'd like to 
> encourage and support him.

Yes, sure.  Just make sure that you explicitly label what you are doing as
not being part of RDF, particularly in discussions where the abilities of
RDF have been misrepresented.

> Mike has asked for guidance on a specific *practical* problem, and building 
> on suggestions from others, I've suggested two possible solutions, with 
> different characteristics.  Both will work, but one may suit his 
> circumstances better than the other, which is why I've called out the 
> differences.
>
> Peter, if you have a better solution to offer, I'd be happy to hear it.

Well, if someone really wants to have shared properties of resources, then
it would be better to move to some formalism that explicitly supports such
sharing.   If someone only wants a way to collect commonalities of
resources and have a single edit point, then it would be better to use a
file-include mechanism that does not look like RDF at all.  (I think that
even your second mechanism, without appropriate disclaimers, can mislead
users into thinking that RDF can do more than it really can.)

> Brian

peter

Received on Friday, 21 December 2001 06:58:54 UTC