At 06:49 PM 8/15/2001 +0100, you wrote:
>physical files for the RDF(S)/XML representations of the namespace schemas
>in the different languages - the alternative of adding them (cumulatively,
>as translations are made) to one schema file seems very unwieldy!? But then
>this means multiple RDF schema files associated with the same namespace.
>Maybe this is unavoidable?
I'm not sure I understand this... Namespaces are a syntactic simplification
for providing uniquely identified terms. Having instance data (Schemas,
examples, profiles, etc.) that "describe" different characteristics of
these identified terms allows for a very flexible and efficient means for
managing decentralized description.
This is a feature, no?
>At one of the breakout sessions, there was a strawman proposal on the table
>(from Tom B or Rachel?) for a "typology of schemas" which included
>"translated namespace schemas" as a third class of schema, alongside
>"namespace schemas" and "application profiles".
This is a useful idea for trying to help understand how DCMI might
partition / create various bits of instance data, however its slightly
confusing from the standpoint of implementation. From an implementation
standpoint its all simply more metadata. Now, this metadata may be using
different vocabularies (identified using Namespaces) in its respective
description, but its simple just more metadata. And if this metadata
happens to all follow standard conventions (e.g in the case of RDF), it can
be parsed, loaded into common databases and queried using the same tools,
independent of the actually "typology" (using the above term) of the
instance data.
--
eric miller http://www.w3.org/people/em/
semantic web activity lead mailto:[log in to unmask]
w3c world wide web consortium tel:1.614.763.1100
200 technology square, ne43-350 fax:1.208.330.5213
cambridge, ma 02139 usa
|