Re: xmlmd:foo

"Simon St.Laurent" wrote:

> At 12:41 PM 5/30/00 -0400, Paul W. Abrahams wrote:
> >> I like it very much, but I'm concerned that URI-related issues identical to
> >> the ones we've been discussing here will crop up in xmlmd:foo work.
> >
> >I don't think so, because we agree at the beginning that the URI isn't just a
> >name; it locates something.
>
> That's where I start to get nervous that the URI community might not be so
> happy about this approach.  Claims like 'but URNs are dereferenceable'
> could cause some continuing controversy here.

Suppose we say that in `xmlmd:foo="bar"', bar has to satisfy the syntax (and
semantics) of an ExternalID (XML[75]).  Then the task of interpreting bar and
retrieving its contents is just like the task of retrieving an external entity,
and we're dealing with something we already know about.

> I'd be happy to see this happen, though, basically as you've outlined it
> here.  (It seems to be in sync with similar proposals made earlier.)

Any thoughts on the two alternatives: a family of `xmlxx:' attributes, one for
each metadata type, versus a single `xmlmd:' attribute with something at the
other end that's required to identify itself?

Paul Abrahams

Received on Tuesday, 30 May 2000 18:14:11 UTC