Microsoft's responce to XML.com article

[Tim Bray]

At 09:45 AM 1/17/00 -0500, Ann Navarro wrote:
>And the "resources" Tim was (most likely)  referring to are W3C paid
>resources -- staff members, engineers, etc.  Those aren't free. They could
>raise dues for the vast majority of memberships (the affiliates), but then
>we'd just hear more exclusionary ranting, so....

Not only - I'm talking about reps from paying members; it's well-known
that even if someone is doing a titanic amount of work editing a spec,
when they have an in-house product-release crunch, well, they're just not
going to be doing any editing for a while.  That is unless they're a 
"standards drone" a well-known species of engineer who isn't trusted
to do any real work but is dispatched to standards committees, found
in some but not all large technology vendors.

I'm not convinced that opening up the W3C process would really make that
much difference in terms of having serious quantities of work cycles
available from qualified, level-headed people.  

But on the other hand openness is a virtue in and of itself, much like
honesty and charity, and I don't think you really ought to have to give
reasons why processes that affect the public interest should be 
open.  They just should.

In fact, the shoe should be on the other foot; if you want to run some
process behind closed doors, you need to have some good, persuasive arguments
as to why this is a good thing.

So, let me offer one such argument, but one that applies to a very limited
application domain.  WHEN you have a consensus-building process that
involves competitors that are energetically trying to wipe each other off
the map, AND one of the weapons in this struggle-to-the-death is dueling 
press releases, AND you've got a trade press that just loves to run stories
of the form "company XYZ {winning|losing} battle over FooML" AND you're 
trying to build consensus around some technology that is expected to have 
real commercial import:

THEN, the effect of having a non-public working group mailing list is
[1] more of the discussion takes place on the list, and less in 
    private back channels
[2] fewer possible emerging consenses [pl of consensus?] are crapped on
    from a great height by some combination of management and marketing

SO, is this effect worth the cost of not being open?

Sometimes.  It depends.  Maybe.   -Tim

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ or CD-ROM/ISBN 981-02-3594-1
Please note: New list subscriptions now closed in preparation for transfer to OASIS.