Lists Home |
Date Index |
Micah Dubinko wrote:
> Since there are multiple versions of some vocabularies that can't be
> distinguished by the namespace, only by the DOCTYPE, something like this
> would help processors figure out what to do.
If this is an accurate statement of the problem, then I think that there
are solutions out there. There are three separate issues:
* multiple "related" namespaces
* multiple versions of a namespace
* multiple schemas for a single logical namespace
If you model your versions as different namespaces altogether, then you
have no problem, just use different namespace declarations.
If you want to have a concept of multiple versions, then you could add a
"version" attribute to your namespace and figure out what version you're
dealing with through that.
If you want a single namespace and multiple schemas (whether for
versioning or otherwise), you can use the XML Schema SchemaLocation
attribute...or equivalent for other schema languages that have such a
The current concensus view is that document types/vocabularies are
distinct from schemas. The DOCTYPE declaration conflated them in a way
that was harmful.