xml-dev - Re: W3C, responsibility (Re: Why the Infoset?)

[Michael Champion <Mike.Champion@softwareag-usa.com>]

----- Original Message -----
From: "Winchel 'Todd' Vincent, III" <winchel@mindspring.com>
To: "Simon St.Laurent" <simonstl@simonstl.com>; "John Cowan"
<jcowan@reutershealth.com>; <xml-dev@xml.org>
Sent: Thursday, August 03, 2000 11:01 PM
Subject: Re: W3C, responsibility (Re: Why the Infoset?)


>
> I am not saying that the W3C is misbehaving like Microsoft or Malboro, et
> al.  I do, however, tend to agree that some of the W3C's behavior is
> questionable, at best, self-serving, at worst.  Further, whether or not
"it
> is written anywhere", it is in the W3C's best interest to be many things
to
> many people even if many of those people are not its paying members.

Questionable, yes, but I'm not sure what special interests of the W3C
members are being served in the examples noted.  The W3C is not some
monolithic organization doing things for obscure and sinister purposes, it's
a bunch of people trying to produce Recommendations that reconcile an
*immense* network of constraints -- including the prior promises of the W3C
management, the stated objectives of a specific activity, the interests of
the participants' companies, individuals' personal sense of
aesthetics/heuristics/interests, purely pragmatic considerations of timing
and resources, etc.  There is NEVER a solution that meets all the
constraints, and it is a laborious process to find a solution or sequence of
solutions -- that is "good enough".

Big companies get flamed for the gap between their promises and performance
because a single organization could in principle avoid making misleading
predictions (by keeping quiet), and clearly benefits from locking in
customers to a vision of the future that is never delivered.  An
organization such as the W3C *needs* to spin its vision, because that is all
it has to define itself, and doesn't benefit from locking in "customers" to
that vision with vapor.  I can see how any company benefits from promising
that its vision will emerge seamlessly from future incarnations of its
products, so "just keep buying those updates, folks!."   I don't see how the
W3C benefits from over-hyping a promised spec that proves underwhelmingly
capable or overwelmingly complex in practice.

In short, I see the "questionable" behavior that was noted (some of it which
I have no trouble with, such as the deprecation of SGML ... and other that I
deplore, such as the growing complexity of the XML-related specs) as the
result of ordinary mortals trying to do so much in so little time, not
self-serving calculation by the W3C or its members.  They are not "big lies"
to be repeated until skepticism is overwhelmed, they are "overly optimistic
forward-looking statements" that should have been taken with considerable
skepticism in the first place.  Something like XML Schema is so complex
because it is the product of 35 (or whatever) people with at least that many
separate agendas.  There's no simplicity because there's no single mind that
can grasp the totality of the requirements and technology, conceive of a
solution, and convince the world to adopt it.  W3C specs fully demonstrate
the truth of the adage "a camel is a horse invented by a committee".  If you
want something swift, graceful and beautiful, don't require it to go without
food or water for a week while surviving extremes of heat and cold.

I fully agree that the W3C has to seriously improve its performance to
maintain its credibility, but an effective prescription generally depends on
an accurate diagnosis.  I believe the problems noted are rooted in the size
of the requirements lists and working groups and not the motivation of the
W3C management or membership.  Ruthless elimination of complicating
constraints will be much more effective than attempts to eliminate
self-serving behavior, even if such a thing were possible.