Re: We need a META schema registry!

At 01:18 AM 3/30/97 -0800, Andrew Daviel wrote:
>What I meant was, a single page which is definitive for all possible 
>values of META attribute [across the entire Web] is unworkable.

No disagreement on this point!

>  A page which lists URLs of
>metadata specifications is more-or-less what I had in mind.

While this would be a great resource to have (and I applaud your
efforts at, we
cannot afford to build a mechanism that _requires_ such a registry.

The point of Dan Connolly's examples was that the Web itself can
serve as a global -- and distributed -- registry, with the descriptions
of any particular element available but not "centralized" (in the
usual sense).  There is a natural tension between those who would
prefer to limit the number of different metadata elements in use
(hoping, probably, that a centralized list of element descriptions
would encourage maximal re-use) and those who would allow the Web
to grow an essentially unrestricted (and unregulated) metadata
dictionary.  I will claim there is nothing fundamental about the
technology that forces a sole registration authority.

>  What I'd like to see is some
>kind of middle ground between the hopelessly vague and the academically
>precise, that could be generated using an authoring tool by "Joe Sixpack"

Yes; allowing every user to generate useable metadata is the only way
that we'll get enough metadata coverage on the Web.  "Joe Sixpack"'s
authoring tool should also allow him and the other members of his team
to define their own (perhaps task-specific) metadata elements and use
these new elements with ease equal to the "common" elements.  Others
outside of Joe and Josephine's work group need not recognize these
new elements but should equally be able to use them if the work group
chose to publish its dictionary.

>META attributes are going to be dominated by whatever the authoring and
>conversion tools generate. If the authoring tool designers think there's
>no standard, we get a hodge-podge of different tags.

We can't wait for there to be one single standard. I agree with you;
we would probably be waiting forever.  But we can create a mechanism
that allows standard_s_ to be shared and propagated easily and let the
(multiple, and perhaps non-overlapping) user communities pick which
one(s) best suit their needs.  And to add their own refinements to
standards that come close but aren't quite enough.  All this while
preserving interoperability at all but the semantic level.
(Semantics being the thing that will be debated ad infinitum :)


Received on Monday, 31 March 1997 10:17:19 UTC