W3C XML Protocol Working Group teleconference, 13 March 2002

1. Roll call

Present 28/23 Excused Regrets Absent

2. Review of Agenda and AOB

3. Approval of March 6 telcon minutes

Approved.

4. Review of Action Items, two things worth noting in minutes:

[i] RDF folks are waiting to see the Data Model.
Yves pointed them to our editors' copy (which contains Data Model) on Monday.
PaulC: how dangerous for us is waiting for somebody to resolve this?

Noah: We should clarify our position on how RDF should proceed - use SOAP
Encoding or their XML serialization?
Discussion.
We (DavidF) will suggest that they can use their current (or any) XML
serialization.
[ii] No responses to XMLP-Comments resolution posts. That's good, if there
is any response, we need to know about it.

5. Status reports:

-- Primer
Nilo: I included comments from various people. There was satisfaction with
my taking care of ednotes, and especially an RDF serialization example
that was added. Generally - it's on track. Primer depends on the spec
which is changing but I'm following the changes closely so I should only
need a few days to catch up after the spec is frozen.

-- Spec editors
HFN: we're incorporating the resolutions, going through comments etc.
DavidF: I'll send my comments to you
HFN: public editors' copy (XML) is being updated on the fly, HTML version
is updated every day or so.

-- TBTF
Holding a meeting later this week, so nothing to report yet.

-- ETF
HFN: we don't seem to have any issues to meet over. So there is nothing to
report.

-- Conformance
Oisin: We have 143 assertions, we haven't got a test for every assertion
  yet. If anybody has comments, they're welcome. Some problems with linking
  to the editors' drafts.
DavidF: can we now parallelize the work?
Oisin: The document will be split in parts (each assertion, probably) in a
  few days and those can be worked on separately.
DavidF: we may want to set up a Conformance Task Force (scribe's name).
Paul: MS has comments yet to be submitted (delayed for various reasons).
  Also implementors might submit their already existing tests.
Oisin: The current expectation is that we must cover all assertions with
  tests (at least one test covering every assertion, one test may cover a
  number of assertions).
DavidF: are there volunteers for helping the Conformance document editors?
  Mark Jones volunteered, John Ibbotson will contact IBM folks who might be
  able to help.

-- Usage Scenarios
JohnI is going through Dietmar's and Jacek's comments on Usage Scenarios,
this and other work will be incorporated in the next version of the
document.

-- Requirements
Bob Lojek: just started with my work, have some little question. Will take
off-line with DavidF.

-- Email Binding
Highland: JJM reasked a question about why not SMTP binding. Also Mark
Baker talked about the abstractness of the binding. There is an action for
Mark and Jacek to provide an appendix, after we see
that we'll revisit this question.

6. Issue gleaned from xmlp-comments and xml-dist-app, all have proposed

   resolutions/next-actions.

From [3]:
-- General, (a) - we accept Gudge's proposal.

-- Part 1, (a) - Gudge to check that Noah's response to the commentator
was satisfactory, and to report back or post to xmlp-comments accordingly.

-- Two editorial comments, to be dealt with by editors

From [4]:
-- Part 1, (a) - Henrik proposes that we don't disallow the attributes
everywhere, we'll say instead that they have the specified meaning only in
the given places. Some violent agreement followed.
Noah: our schema has lax validation of content which may have the wrong
semantics. We might want to have "skip" validation or both versions of our
schema.
Resolution: Noah will post his comment and we'll contact the commentator
that we accepted what we violently agreed on. (Action on HFN)

-- Part 1, (b) - this is an issue on the MIME type draft. We might be
unclear on what is the issue. HFN will contact the commentator (MarkB).

-- Part 1, (c) - both comments seem to be true. There may be some
understanding that the default encoding, if not indicated explicitly, is
the SOAP Encoding. Our opinion is that this understanding will gradually
diminish. We might want to explain this a bit more in the spec. Gudge took
an action to generate the appropriate text and to respond to the
commentator.

-- Part 1, (d) - MarcH (the commentator) thinks this is no longer
applicable, he will write the appropriate text to xmlp-comments.

-- Part 1, (e) - HFN: this is handled by our resolution to issue #137 from
last week. The WG agrees.

-- Part 2, (a) - Gudge's proposed resolution accepted. He'll write the
responses.

-- Part 2, (b) - Gudge: we're actually going the other way from what was
requested, but the commentator is quite happy with this. We will postpone
until the section 2 rewrite is available.

From [5]:

-- Part 1, (a) - Henrik proposes that the encodingStyle value need not
apply to the element it's on, just to its children.
Noah counterproposed that encodingStyle, used on Body, for example, really
means that the Body is encoded using the given encoding.
Gudge proposed that we disallow encodingStyle on Body (and the scribe
presumes on Header and Envelope, too).
Resolution: we open a new issue.

-- Part 2, (a) - dependent on the ongoing rewrite of Encoding and Data
Model sections. Postpone discussion until the rewrite is available.

-- Part 2, (b) - Gudge: this is partly affected by the ongoing rewrite,
too. But we might want to weaken the naming requirement on the RPC result
element.
Ray: this topic is related to issue #180.
Gudge will send in a new issue text to the issue list maintainer.

Meeting closed.