[
Lists Home |
Date Index |
Thread Index
]
2/7/2002 12:15:29 PM, "Bullard, Claude L (Len)" <clbullar@ingr.com> wrote:
>The telling part is the "commitment to an aggressive schedule".
>
>My guess: they are terrified of the fragmentation over XML Schemas, the
>slow pace of the W3C at this time, and the centrality of the
>TAG.
I suspect that's true. The W3C was favored by the movers and shakers as a
place to hammer out modus vivendi in the days when HTML and DOM
interoperability was the critical issue. For various reasons, the W3C is
no longer a useful forum for "lets just make this WORK" discussions. As
the debates over W3C XML Schema, the Semantic Web, and XQuery here (and
the SOAP/REST discussion on xml-dist-app) have made clear, the W3C is
focused these days on trying to Do the Right Thing and build a
theoretically grounded foundation for the future more than a pragmatic
accomodation for the present.
>
>Meanwhile, a self-sustaining market emerges and the
>lock-in solidifies based around a handful of workable
>standards. It's called "colonization". It worked for
>the W3C and has worked ever since. Game as played.
>
>Ever wonder why the old guys used ISO? Nation/state
>customers. Slow but dammed hard to derail.
OK, WSIO is about the web services hype purveyors needing to get some
interoperable reality on the ground NOW rather than wait for the ISO or
W3C to define the Right Thing. That's dangerous in that the potential is
there for the biggest players to ram quasi-proprietary specifcations down
our throats and call them standards, and we'll find ourselves locked into
them for some time to come. But the alternative isn't everyone sitting on
their hands for 3-5 years while the Right Thing is defined, it's having
the biggest players shove REALLY proprietary, heavy-duty nasty lock-in
technologies down their customers throats. These things won't
interoperate, so we (except for the consultants) would be screwed even
worse than we will be under the WSIO regime.
Sorry, but I do think that the essay-that-shall-not-be-named (because its
title muddies the waters) at http://www.ai.mit.edu/docs/articles/good-
news/subsection3.2.1.html is quite instructive in this regard. Let's call
it "Good is Better" (following voltaire's famous dictum "the best is the
enemy of the good"). The Best would be a theoretically grounded, well-
designed, elegantly formulated, gracefully written specification for how
web services can work. The Good is more or less what SOAP/WSDL/UDDI is
today with a lot of bashing to find the subsets/refinements that actually
work together. The problem is that hype waits for no man; expectations
have been raised, and if the industry doesn't work together to produce a
"good is better" solution, they will work separately to produce
interoperability hell that makes kludgy compromises look like paradise.
Don't like W3C HTML? Think about a world where numerous HTML 2
derivatives and HTML 3-esqe variants exist, with Blackbird being installed
on every copy of Windows. Best is Best, but Good is a *lot* better than
that scenario.
BTW, this is not an endorsement of WSIO, just the notion that if the W3C
isn't going to be the place to hammer out short term tactical
arrangements, SOMTHING has to take its place ... as Simon mentions in his
weblog, there's a fishy smell of patents in the air ...let's hope that
WSIO's hidden agenda isn't the big patent holders taking their ball to a
new field now that the W3C has said "no" to their preferred patent policy.
|
