[
Lists Home |
Date Index |
Thread Index
]
John Cowan also did a rather good presentation on Relax NG at 2004.
His points on why RelaxNG is good were well made.
I asked in a meeting I chaired how many thought it would be a
good idea if Relax NG were more prominent (or something like
that). Not many hands went up. Not good.
1) We beat the standards message into a lot of heads in the
last decade. It worked.
2) Predictably, the W3C As Good Standards and ISO As Clueless
Standards messages converged with message 1. Toss in the Microsoft
Does W3C Schema as an externality with positive feedback
effects for XSD and....
Result: RelaxNG Bad. W3C Schema Good. Schematron: whatever
(we have business rules but ok).
It's a nonlinear network out there. Given the inability to
figure out the details, managers, buyers, etc., punt to the
greatest signal. They can't afford multiple ways to do the
same thing.
So now what? Seems obvious. The W3C should endorse RelaxNG
as a preferred alternative. That will be as good as it gets.
The lock is not on the schema type but on the tools.
len
The weird thing is, Betamax didn't just lose, it disappeared.
The interesting question is why? Video stores couldn't afford
two formats because they were in a low margin market. Given
a limited supply of movies, video stores were the externality that
pushed the Betamax all the way out of the market. See
"Betamax Revisited: A Contextual View of the Battle for the
Home VCR Market"
From: DuCharme, Bob (LNG-CHO) [mailto:bob.ducharme@lexisnexis.com]
> Likewise, just doing a text search on the "agenda at a glance", XML
>2002 seems to have had 3 presentations on RELAX NG, XML 2003 had 1, and
>XML 2004 had none.
My talk "Documents vs. Data, Schemas vs. Schemas"[1] didn't mention RELAX NG
in the title so that it would show up in the schedule summaries, but it came
to the conclusion that lots of people out there would be better off using
RELAX NG than XSD.
A "why RNG is good" presentation would end up preaching mostly to the
faithful, and I thought that a talk structured around more generalized
processing issues would be better for the conference than one that focused
on pushing a specific tool, especially if the tool is not as new as it was
in 2002.
|
