[
Lists Home |
Date Index |
Thread Index
]
On Mon, 1 Nov 2004 22:14:51 -0000, Michael Kay
<michael.h.kay@ntlworld.com> wrote:
> The big software vendors all committed themselves heavily to XML Schema back
> in 2000/2001 and have huge investments in it. The more you spend, the less
> your inclination to change direction. Also, they see the big spenders in the
> user community equally committed (or locked-in, if you prefer).
I dunno .. look at the reports from the Microsoft-centric XML
Developers Conference I linked to in the Partyin' thread. There seems
to be a lot of disaffection with XSD, at least in its current form.
Anyway, support and maintenance costs can dwarf initial investments,
and the pain that users feel may be starting to cost big vendors
money.
As best I can read the tea leaves, what is most important for the big
vendors is standardization in the de-facto sense: many customers
demand NOT to be locked in, and want assurances that what they are
building can be ported or made to interoperate with other platforms,
tools, etc. without undue pain. The other major need is a certain
amount of predictability -- big vendors have big customers, and they
don't turn on a dime, or throw out technology investments just because
they aren't fashionable this year. Advocates for the
out-of-the-mainstream specs do have to work wirth these hard facts:
build the tools to make them de-facto reality on the ground, figure
out how to import/export XSD or otherwise interoperate with the
"legacy" XSD products to the greatest extent possible, and somehow lay
out a logical roadmap from where real XML users are today to where
they could be with RNG/Schematron being more pervasive.
Big vendors and their customers aren't going to adopt Schematron or
RNG just because they are official international standards, but
neither are they going to scorn them because they don't come from W3C.
My guess is that they will adopt them when (if?) it becomes obvious
that a significant chunk of people are getting their work done
better/faster/cheaper with them than with XSD.
If these technologies are really as good as most people on this list
seem to think, it may not take much to get the ball rolling.
|
