On Tue, 2001-10-16 at 11:23, Rachel Heery wrote:
>
> What about 'plain XML' with no RDF embedded? My concern with RDDL in the
> DCMI context is that resolution is in effect to multiple expressions of
> the 'schema' using different schema languages. Each expression will carry
> different emphasis, each will allow for different aspects to be expressed.
> I think DCMI needs a canonical schema from which others flow, and that is
> the schema to which the namespace resolves. So I would go for
> resolution to RDFS.
Yes,
But a first among equals. One isn't better, they mark up the same
concepts. I think that that you provide a pointer to the set and maybe
pick a default but it is up to the comsumer as to what their app
requires (or pick as default, what you do in the privacy of your own
machine).
>
> Which of course still offes possibility for someone to provide a source
> for multiple expressions of DCMI terms in different schema languages. I
> think this would be useful. However such schemas would not go thro the
> same 'approval process'. RDDL seems to me to be a 'registry' or catalogue
> of schemas in effect?
>
> Rachel
Will RDFS include links to the XHTML (HTML) prose or do we do that by
word of mouth (e.g., elements 1.1 spec lives at http://dc.org/blah and
some of it's markup is at http://purl.org/dc/blah or
http://abc.org/foo1.1).
Isn't nice to be able to describe the relationships of the how these
interact in a cononical location? Relationships are just as important
and hard to markup (hence the bloody state of RDDL).
Cheers Tod...
|