RE: And the DTD says, "I'm NOT dead yet!!"

[Arjun Ray <aray@q2.net>]

On Mon, 8 Jan 2001, Simon St.Laurent wrote:

> At 10:23 AM 1/8/01 -0600, Bullard, Claude L (Len) wrote:

>> One can privately agree to disagree, but the public statements
>> make for political problems which will turn into technical
>> problems.

Only if we forget Harry Frankfurt's celebrated essay:)

> Which only gets more complex when the design issues docs at the
> W3C and early Notes get cited as axiomatic.
> 
> (http://www.w3.org/DesignIssues/Architecture.html)
> (http://www.w3.org/TR/NOTE-webarch-extlang)
> (etc.)

More than axiomatic, given the relevant druthers.  Scriptural.

> Fortunately, pointing this out on a regular basis does some to
> keep it down somewhat.

We should certainly hope so, no?  As Rick says:

| W3C technological specs are developed by W3C Working Groups in which
| W3C staff only have the same voting rights as other members.

In Summer 98, yet another draft of the XML-Names spec was finally
finding consensus in the XML WG, when somewhat peremptorily a brand
new proposal was thrust up for consideration, by someone who, we are
to believe, had the same "voting rights" as everyone else.  Here are
excerpts from The Forever Forbidden Archives about how "not really
more equal than others..." may work in practice.

Arjun


==[19]==

> The subject of the meeting is, of course, namespaces.  It is now
> generally agreed both within the WG and without it that the thing
> to do is finish the draft more or less as it stands for 1.0 and
> continue work on a defaulting/scoping mechanism for version 1.1 in
> a subcommittee.

It's agreed within the WG, but not so much without... I took
Jon's reports to the Hypertext and Metadata CGs and discussed
them with TimBL and some of the other staff:

In particular, while a few folks are minimially satisfied
by the status quo (18 May draft less src=) nobody's particluarly
happy: not this WG (who knows there are larger issues to
tacke), not RDF (who can live without defaulting and scoping,
but don't want to) not P3P (who can't live with PIs that
show up in extant HTML software, nor the requirement to
prefix all namespace'd names) not Micropayments (who can't
live with PIs) and not HTML (who can't live with PIs).

Jon offered the alternative that this WG work out the
larger issues for a 1.0 spec, and offered a lower bound
of 3 to 4 months for that work.

In the discussion with TimBL and other staff, we felt it was
unreasonable to just say "hurry up." But I believe the larger
issues can be solved in short order, and I have a concrete
alternative proposal that includes scoping, defaulting, and
meets the external requirements that I've seen (double-checking
in progress to get explicit say so from other WGs).

So... I give you:

========
Namespaces in XML
                   W3C Working Draft 08 Jul 1998
This Version:
     http://www.w3.org/XML/Group/1998/07/xml-names
     $Id: xml-names.html,v 1.13 1998/07/08 17:00:51 connolly Exp
     $
Latest Version:
     http://www.w3.org/TR/1998/WD-xml-names
Previous Version:
     http://www.w3.org/TR/1998/WD-xml-names-19980518
Editor:
     Dan Connolly <connolly@w3.org> W3C

Copyright   1998 W3C (MIT, INRIA, Keio), All Rights Reserved.
W3C liability, trademark, document use and software licensing
rules apply.

Status of This Document

This is a proposal to the W3C XML Working Group.

This document is CONFIDENTIAL to the W3C Membership.

This is part of the XML Activity.

Please send comments to the editor.

This document results from a discussion of the constraints and
requirements on the 18 May draft, and proposes a solution which
meets the external constraints (from HTML, P3P, and RDF).

We're working on an implementation. A larch specification is in
progress as well.

Abstract

This specification extends XML 1.0 to bind attribute and element
names in XML documents to global identifiers.
========


==[26]==

I have just received a reminder that, as AC rep, I am supposed to
vote on the XML Activity proposal and indicate whether representatives
from my company want to participate.

At this moment, I cannot understand why I would want to.

The major investment my company makes in W3C is the large amounts
of time good people spend trying to work towards W3C Recommendations.
[We from our company] have spent good chunks of the last year on
namespaces, as have many others on the XML WG.  If the W3C team can
veto WG work and replace it with their own, I question why my company
should invest more time in developing W3C recommendations.


==[27]==

> The major investment my company makes in W3C is the large amounts
> of time good people spend trying to work towards W3C
> Recommendations. [We] have spent good chunks of the last year on
> namespaces, as have many others on the XML WG.  If the W3C team
> can veto WG work and replace it with their own, I question why my
> company should invest more time in developing W3C recommendations.

Veto? I'm a member of the WG; as such, I can propose things, no?

Also, I'm the staff contact, which gives me the responsiblity
to liase with the other W3C working groups etc. via the director
and staff.


==[28]==

>> The major investment my company makes in W3C is the large amounts
>> of time good people spend trying to work towards W3C
>> Recommendations. [We] have spent good chunks of the last year on
>> namespaces, as have many others on the XML WG.  If the W3C team can
>> veto WG work and replace it with their own, I question why my
>> company should invest more time in developing W3C recommendations.
>
>Veto? I'm a member of the WG; as such, I can propose things, no?
>
>Also, I'm the staff contact, which gives me the responsiblity
>to liase with the other W3C working groups etc. via the
>director and staff.
>

Perhaps [he] was referring to the earlier mail from TB-L which stated
what the XML WG will be working on in the future and it doesn't matter
whether the WG agrees with it or not.

That certainly seems to [our company] as if "the W3C team can veto WG
work and replace it with their own." We absolutely agree with [him].


==[33]==

| Veto? I'm a member of the WG; as such, I can propose things, no?

So the proposal that you sent out today is just a proposal -- right?

| Also, I'm the staff contact, which gives me the responsiblity
| to liase with the other W3C working groups etc. via the
| director and staff.

Monday morning we discussed this in the Hypertext CG and at more
length in the Metadata CG.  There was general agreement at that time
with the finding of the WG that it was better to move something
essentially equivalent to the current working draft to PR for 1.0 and
then have a subcommittee put a defaulting and scoping strategy in
place as soon as possible than to hold up the 1.0 PR until we could
reach agreement on a defaulting and scoping strategy.  This being the
case, and in the absence of any new data that would change the
substance of the discussions that we've already had about this in both
the WG and both CGs, it would seem that our disposition of your
proposal at this late date will either be to accept it by acclamation
as obviously the right answer or to refer it to committee and go ahead
with the part that has (after a year of work) gained consensus.  Are
there other alternatives that I'm missing here?


==[34]==

> | Veto? I'm a member of the WG; as such, I can propose things, no?
>
> So the proposal that you sent out today is just a proposal -- right?

Sure.

> | Also, I'm the staff contact, which gives me the responsiblity
> | to liase with the other W3C working groups etc. via the
> | director and staff.
>
> Monday morning we discussed this in the Hypertext CG and at more
> length in the Metadata CG.  There was general agreement at that
> time with the finding of the WG that it was better to move
> something essentially equivalent to the current working draft to
> PR for 1.0 and then have a subcommittee put a defaulting and
> scoping strategy in place as soon as possible than to hold up the
> 1.0 PR until we could reach agreement on a defaulting and scoping
> strategy.

There was agreement that this was the XML WG's plan, yes.
But that doesn't mean all the other WG's like the plan,
especially the groups that weren't on the call
(P3P, HTML-to-be, ...).

In particular, I passed on to the director the WG's
intention to ratify the namespace spec pretty
much as is (18 May less src=) and asked if he was
likely to promote it to PR. The answer I got was no.

So we talked about what he was looking for, and I wrote
it up as a proposal.

>  This being the
> case, and in the absence of any new data that would change the
> substance of the discussions that we've already had about this in
> both the WG and both CGs, it would seem that our disposition of
> your proposal at this late date will either be to accept it by
> acclamation as obviously the right answer or to refer it to
> committee and go ahead with the part that has (after a year of
> work) gained consensus.  Are there other alternatives that I'm
> missing here?

Not really.


==[37]==

|> Monday morning we discussed this in the Hypertext CG and at more
|> length in the Metadata CG.  There was general agreement at that
|> time with the finding of the WG that it was better to move
|> something essentially equivalent to the current working draft to
|> PR for 1.0 and then have a subcommittee put a defaulting and
|> scoping strategy in place as soon as possible than to hold up the
|> 1.0 PR until we could reach agreement on a defaulting and scoping
|> strategy.
|
| There was agreement that this was the XML WG's plan, yes.
| But that doesn't mean all the other WG's like the plan,
| especially the groups that weren't on the call
| (P3P, HTML-to-be, ...).

I'm sorry, but the agreement was rather stronger than that.  I
specifically put to the Metadata CG the question of whether they
would rather go now with the part on which we've reached consensus
and then add defaulting/scoping as soon after that as possible, or
wait on a PR for 1.0 until we had defaulting/scoping in place.
There was considered and clear agreement that the former course of
action was preferable to the latter, just as there had been earlier
when the same question was thought through in the WG.

| In particular, I passed on to the director the WG's
| intention to ratify the namespace spec pretty
| much as is (18 May less src=) and asked if he was
| likely to promote it to PR. The answer I got was no.

Well, that's useful to know.

| So we talked about what he was looking for, and I wrote
| it up as a proposal.

I'm having a lot of trouble distinguishing this from simply being
told how to design the spec.  Could you explain the difference?