W3C home > Mailing lists > Public > www-tag@w3.org > March 2002

Re: [namespaceDocument-8] 14 Theses, take 2

From: Norman Walsh <Norman.Walsh@Sun.COM>
Date: Mon, 04 Mar 2002 14:05:47 +0000
To: "Tim Berners-Lee" <timbl@w3.org>
Cc: "Jonathan Borden" <jonathan@openhealth.org>, "Simon St.Laurent" <simonstl@simonstl.com>, "TAG" <www-tag@w3.org>
Message-ID: <878z98pgw4.fsf@nwalsh.com>
/ "Tim Berners-Lee" <timbl@w3.org> was heard to say:
| 1.  We cannot now define the languages which will be available later to say
| things about namespaces.

Yep.

| 2. Human readable information is good.  Humans can sue it

Sue it? Use it? Use it, probably.

| 3. Machine-readable information is good.  It allows automatic processing.

Yep.

| 4.  Where there is machine-readable information, it must be
| machine-accessible from the namespace URI, either directly or indirectly,
| but without human intervention.

I think this argues strongly for indirection. Document types for which
there is only a single kind of useful information are going to be rare
(I think Tim Bray called them uninteresting edge cases). So even if I
publish something today for which I only have a single kind of schema,
one can imagine that I'll eventually have other sorts of information.

| 5. Where all the information available can be expressed in one (not too
| long) document then an indirection for the sake of it is an engineering
| mistake.  So clients should be prepared to accept information directly or
| indirectly, ideally.

Architecturally, I expect the benefits of requiring (or strongly
encouraging) indirection even for the initial single resource outweigh
the cost. Especially when you consider that resource retrieval only
comes into play for those namespace names that are *ever* retrieved.
(This plays into the use case where a namespace name is known by an
application and the application never, ever retrieves anything, it
just *knows* about it.)

| 6. Where content negotiation is used, it should only be used to negotiate
| between documents which really are equivalent - they basically say the same
| thing in a different language.  For example, it would not be appropriate to
| give and RDF schema and XML schema for a namespace because they really
| contain different information, and a machine or human would be fooled into
| thinking it knew the import of a document, when really it had been given
| something different.

So what counts as equivalent? If the RDF Schema and the XML Schema are
different, are the XML Schema and the DTD also different? What about
the RELAX NG Grammar and the Schematron schema?

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM   | Men are more like the times they live in than
XML Standards Engineer | they are like their fathers.--Ali Ibn-abi-talib
XML Technology Center  | 
Sun Microsystems, Inc. | 
Received on Monday, 4 March 2002 10:05:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:05 GMT