I wasn't suggesting that for this TC. I should have made it more
clear that the paper I'm writing has to do with the HPCDML
subcommittee and some other work, and that what I am looking at is
how to do THAT work, and it was in that context that I concluding
that namespacing with a verbose naming convention was what would be
arrived at in the end. In our meeting this week I volunteered to pass
along some references for naming conventions, so what I intended for
this TC was only those listed at the end of that excerpt.
However, you do bring up the essential question we need to answer if
we decide to formalize a vocabulary ourselves.
Ciao,
Rex
>At 07:24 AM 4/18/2003 -0700, you wrote:
>>[...] I have found that the combination of namespacing for
>>standards which have produced an xml schema, using the namespace
>>prefix, and a verbose naming convention within the standards
>>themselves, such as the one I suspect we will produce, is the most
>>workable if not the only workable method to use to conform to what
>>standards have emerged and are emerging.
>
>Rex
>
>Seems to me this is equivalent to declaring failure at the outset!
>Names qualified with lots of namespaces do accomplish uniqueness,
>but only by obscuring any overlap in meaning that may in fact exist.
>
>We should all agree that it is impossible to standardize on XML
>tags, just as it has proven impossible to create all-inclusive
>database schemas for complex systems. But, it is quite feasible
>to agree on common concepts (e.g., person, organization, address).
>Then, one simply requires system interface developers to register
>those data elements that are used at the interface, noting which
>common concepts underlie those data elements. This is the essence
>of ISO 11179.
>
>Eliot
--
Rex Brooks
GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth
W3Address: http://www.starbourne.com
Email: rexb@starbourne.com
Tel: 510-849-2309
Fax: By Request