Re: 2001-09-07#5 Literals (use cases/test inputs, please?)

Brian McBride wrote:
> 
> Dan Connolly wrote:
> 
> [...]
> 
> >>Doesn't this just represent a deficiency in cwm?
> >>
> >
> > Perhaps, or perhaps it represents a deficiency in
> > the design of literals. Why is it we can query
> > on any other property of anything using RDF properties,
> > but we can't query the language of a label using
> > RDF properties?
> 
> Yes and there are a few more like that, aren't there:
> 
>    o Why can't I query the namespace of property using properties?

Actually, we do have a property for that
use case: rdfs:isDefinedBy.
It has to be explicitly stated, but that's the only
thing that makes sense; it doesn't make sense to speak
of *the* namespace of a property, in part because
it doesn't make sense to speak of *the* name of
a property: a property can have lots of names.

>    o Why can't I query the uri of resource using properties?

Er... in cwm, I can. it's called log:uri.
(and it gives *a* URI, not *the* URI of a resource, provided
it's been told one).

As TimBL explained (Fri, 19 Oct 2001 14:00:37 -0400),
I could enhance cwm to destructure
literals similarly, but the result is extremely distasteful.


> > Regardless, my request stands:
> >
> > | please give an
> > | example (or several examples) of how RDF users use xml:lang,
> > | and what software they use to process it according to
> > | their expecatations?
> 
> Why are you bringing this up again?  We have been round this already.
> My recollection is that we examined this issue, we looked at use cases, we found
> some where the M&S spec of how to handle xml:lang was useful and we found no
> major problems with it.

Then please point me at the relevant test inputs
and/or the software that handles these test inputs
to the satisfaction of the users.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/