RE: Comments on history system descriptive note

On Fri, 2003-05-09 at 12:50, Tansley, Robert wrote:
> > > I think it is clear that *either* N3 *or* graphical representations 
> > > are more intuitive than XML. Wouldn't N3 be easiest in email, wiki, 
> > > etc. discussions, esp. where iteration on an example might be 
> > > required?
> > 
> > N3 is fine for email. But for all 'official'? SIMILE 
> > documents, I would strongly suggest using the XML serialization.
> 
> Would you mind explaining why?  As I've already mentioned, I find the XML serialisation almost impenetrable.  It is apparently not meant for human consumption (and IMHO needs a lot of work if that is an intent.)
> 
> Having said that, N3 doesn't look much easier.
> 
> A graph is a fundamentally non-linear thing so a human 'reading' a serialisation (or at least a visual thinker like me) will always have to perform some mental gymnastics to reassemble the serialisation into a graph.  Accordingly where RDF must be included in a document to demonstrate some point, I'd strongly suggest showing graphs where at all possible.

I may have scanned the email thread a bit too fast. I thought the
discussion was wrt RDF/N3 and RDF/XML. I would strongly suggest
graphical representations of the RDF model be included in the
documentation.  We've done both the Graphical and RDF/XML serializations
in the Primer and feedback from this experience has been extremely
promising.

-- 
eric miller                              http://www.w3.org/people/em/
semantic web activity lead               http://www.w3.org/2001/sw/
w3c world wide web consortium            http://www.w3.org/

Received on Friday, 9 May 2003 12:55:43 UTC