W3C home > Mailing lists > Public > public-p3p-spec@w3.org > April 2004

MINUTES: 7 April P3P spec call

From: Lorrie Cranor <lorrie@cs.cmu.edu>
Date: Wed, 7 Apr 2004 15:49:05 -0400
Message-Id: <A0A064C2-88CC-11D8-9BE1-000A95DA3F5A@cs.cmu.edu>
To: public-p3p-spec <public-p3p-spec@w3.org>

Brooks Dobbs
Jeff Edelen
Giles Hogben
Lorrie Cranor
Jack Humphrey
Rigo Wenning
Dave Stampley

Lorrie asked about objections to points a-d.

a. http://lists.w3.org/Archives/Public/public-p3p-spec/2004Apr/0001.html

b. http://lists.w3.org/Archives/Public/public-p3p-spec/2004Apr/0002.html

c. http://lists.w3.org/Archives/Public/public-p3p-spec/2004Apr/0003.html

d. http://lists.w3.org/Archives/Public/public-p3p-spec/2004Apr/0004.html

There was consensus to adopt a-d. Giles asked about
primary puroposes. Lorrie said that this was a separate issue.

2. primary purpose specification
[Dave please send revised draft]

There was no new Draft.

3. P3P Generic attribute for XML applications

Rigo explained that the problem was how to use P3P in new protocols
and services. Liberty Alliance and Web Services were not the only
applications. Liberty Alliance has not done much but Web Services has
recently gotten very interested. Joseph and Patrick's draft had a way
to link P3P policies to WSDL via HTTP requests and URIs, but doesn't
work offline or without HTTP. Also to describe a complicated WSDL
interface requires writing a complicated P3P policy. But there are
simple things going on at each part of the interface, so we would
really like to address one P3P policy for a part of a WSDL
interface. This would also solve the XForms problem of having a P3P
policy per field. So... Rigo, Patrick, Steve Pemberton, Hugo, and
Phillipe had several calls and worked on drafting a P3P attribute
proposal which is in the first public P3P 1.1 working draft. Then
Massimo raised the issue that the P3P attribute doesn't have a defined
semantic meeting and it can have a different meaning in different
contexts. Lorrie raised similar concerns. Rigo had extensive
discussions with Massimo about this. Rigo thinks the concern has to do
a lot with chunks of XML being sent over the wire (as in CC/PP), but
this is not what he has in mind. WSDL and XForms are interface
description languages. The P3P policy would apply to what happens when
you use an interface. So then there was the suggestion to limit the
generic attribute to interfaces and not allow it to be used when XML
is sent over the wire. Massimo says it is a hint to XML schema
developers but not a MUST. So the question is, do we want just a hint
or do we want to try to do something stronger, at least for WSDL and
XForms. We could leave it to those working groups to import P3P into
their own specs.

Giles asked why HTML is not an interface. Rigo explains that WSDL and
XForms are interface descriptions because they are not the instance
data - that is your name, address, etc. that gets encoded in XML. HTML
is the instance data. Giles argues that HTML is actually similar. Rigo
says the P3P attribute makes sense for some elements and not others,
even in HTML.

We need to define interface descriptons. This is tricky.

ACTION: Rigo will see if another group has already done this.

ACTION: Lorrie and Rigo will work together to draft a proposal for
what to put in P3P 1.1. The working group will still need to discuss
whether this should be a MUST, MAY, or SHOULD.

Giles says there is a difference between saying "if you give me data I
will do x y z" and "I'm giving you my data and I want you to do x y
z." P3P does the former.

4. Who should be listed as authors of p3p 1.1?

5. Dataschema in XML Schema
Giles proposed to get it done for week of 23 April.

Action: Put out a new working draft with the things ready so far.

Action: Giles forward email to Massimo and Martin Presler-Marshall.

6. Publication of next working draft
Besides from the above, the only other open item I see is the schema
stuff that Giles is working on. Am I missing anything else?

ACTION Rigo: Next Draft for 16 April

Next working Draft by May 7 with the remaining issues.

7. Authors
Who should be listed as authors of p3p 1.1?

Number of people that have contributed sections.
Dave Giles Matthias Brooks Jack

Everybody should be in contributors.

8. Next call

14 April 2004

Scribe: Rigo
Received on Wednesday, 7 April 2004 15:49:51 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:02:18 UTC