[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Namespaces Revisited...

• From: Rick Jelliffe <ricko@a...>
• To: Tim Bray <tbray@t...>
• Date: Tue, 15 Sep 1998 13:48:33 +0800

Tim Bray ¼g¹D¡G

> At 11:13 AM 9/14/98 -0400, John Cowan wrote:
> >Unfortunately, we're stuck with it, until some ISO/IEC committee
> >can be persuaded to take up XML (there is already one trying
> >to create ISO HTML).

The ISO HTML is in fact already created. It was created in conjunctionwithHTML 4;
it is basically strict HTML 4 with additional position contraints so
that heading elements must be used to follow the ranking number
(H1, H2, H3, etc). From memory it has OBJECT, SCRIPT removed,
and no formating elements, and no frames.  It is suitable for very conservative
technical documentation.  The ISO people involved received full cooperation
from the W3C people, as far as I am aware.

Some organisations can only quote ISO or national standards as part of
their tendering requirements: ISO HTML was developed to allow such
organisations the benefit of HTML. Note that XML, even though it
is not an ISO standard, can be specified as part of tendering requirements:
ISO 8879 specifically has facilities to bring in subset specifications
like XML (i.e. the SEEALSO parameter, see ISO 8879 Annex L:
it is less than a full profile mechanism, which was felt to be overkill
in the light of SGML's existing "toolkit" customizability .)

On the issue of whether W3C is a standards body, I think W3C has been pretty
scrupulous to call their technologies "specifications" not "standards".
Every company who invents something that they think will be widespread
calls it a "standard", but it is useful to restrict the term to only those things

which have been through some broadbased, open procedure.  In practise,
some W3C efforts (e.g. XML) have been very broadbased and open; but
once there is widespread interest in a technology there needs to be some
vetting procedure to keep idiots out--a national standards body approach
like ISO uses is such a mechanism (and thereby falls open to the accusation
of becoming an "old boys clubs", which some say about IETF).

In the XML effort, I note that now national standards bodies can make
submissions to W3C concerning XML. There is a strong level of interaction
between ISO committees and W3C and industry consortia now (which you
can see from the recent CGM report  at www.w3.org/TR, so W3C
specifications are becoming more like standards.

> ISO/IEC WGs may be horribly slow and politics-ridden,
> >but they can't just *ignore* comments if they are properly submitted.
> >They *have* to process all of them.

But it is quite possible for a committee to stiffle comments, even ifthey are
good.  To call it "politics" is just to say that technologies
exist in the human world, where there are high stakes in the direction
of technology.

> I fully acknowledge that you disagree with the conclusion that the
> committee came to, but it is incorrect and damaging to allege that
> the process ignored the input.

I can certainly vouch for that. The namespaces discussion seemed
to take almost as long as the XML discussion did; along the way
many people changed their minds.

Rick Jelliffe


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• Follow-Ups:
  • Re: Namespaces Revisited...
    • From: John Cowan <cowan@l...>
  • W3C and the public (was Namespaces Revisited...)
    • From: "Simon St.Laurent" <simonstl@s...>

• References:
  • Re: Namespaces Revisited...
    • From: Tim Bray <tbray@t...>

• Prev by Date: Re: Proposition: "SGML is Gumming Up the Works"
• Next by Date: XSchema: Final Review (1-3, A, B, & D)
• Previous by thread: Re: Namespaces Revisited...
• Next by thread: W3C and the public (was Namespaces Revisited...)
• Index(es):
  • Date
  • Thread