> From [log in to unmask] Tue Aug 13 14:14 MET 2002
> MIME-Version: 1.0
> Date: Tue, 13 Aug 2002 08:11:26 -0400
> From: "Wagner,Harry" <[log in to unmask]>
> Subject: Re: Registry Update
> To: [log in to unmask]
>
> Hi Andy,
>
> > Please, please, please... can we remove the incorrect RDF
> > schemas that the
> > namespace URIs resolve to. Incorrect data is *worse* than no
> > data - and
> > much more confusing.
> >
> > It is totally crazy to have released
> >
> > http://dublincore.org/usage/terms/dc/current-elements/
> >
> > but to have left an incorrect schema at
> >
> > http://purl.org/dc/terms/
> >
> > A simple 'rm' should do the job - 2 minutes at most!
>
> I am in favor of this also - and I would call it more of a 30 second job
> than a 2 minute job. The problem is the fallout this would cause.
> Applications reference those links and would have problems if we remove
> them. It is equally as difficult to change them. I know this from
> experience.
Why we had all this discussion on RDF Schema on this group and elsewhere, if
no action results from it?
It's wrong to say, that classification and note are official DC refinements.
This has virtually nothing to do with RDF!
>
> The reliance on schemas (even though no one is validating!)
RDF is validated by Parsers!
RDFS is a for vocabulary declarations - are you claiming one could
do any kind of reliable metadata application without term declarations??
> is one of the
> many things that irk me about rdf. IMO, we could have (should have?)
> produced the Registry using XML in less time and with less $$$, and all been
> better off.
How you will do XML validation without an XML Schema or DTD?
I can't believe it!
rs
That's my 2 cents worth...
>
> Regards,
> Harry
>
>
|