- From: Ian B. Jacobs <ij@w3.org>
- Date: Wed, 30 Jan 2002 18:51:26 -0500
- To: www-tag@w3.org
TAG teleconference
28 Jan 2002
Present: Tim Berners-Lee (TBL, Chair), Tim Bray (TB), Dan Connolly
(DC), Paul Cotton (PC), Roy Fielding (RF), David Orchard (DO), Norm
Walsh (NW), Stuart Williams (SW), Ian Jacobs (IJ)
Absent: Chris Lilley
Previous meeting 21 Jan 2002
http://lists.w3.org/Archives/Public/www-tag/2002Jan/0193
Next meeting: 4 Feb 2002
See also IRC log:
http://www.w3.org/2002/01/21-tag-irc
A summary of open action items may be found at the end of this
message.
---------------------
Agenda:
1. Setting expectations about TAG responsiveness
2. Lightweight process for publishing TAG findings
3. nsMediaType-3: Relationship between media types and
namespaces, character encodings
4. Web Architecture evangelism
5. uncefactLiaison-5: Request to liaison with UN/CEFACT ebTWG
Architecture Group
6. xformsReview-4: Request to review XForms in Last Call
---------------------
------------------------------------------------
1. Setting expectations about TAG responsiveness
------------------------------------------------
The TAG discussed the (significant) amount of discussion
currently going on on www-tag. While the TAG encourages this
discussion, it is also necessary to set expectations about TAG
responsiveness. In general, the TAG only expects to be able to
address 2-3 issues at its weekly teleconference. The TAG commits
to the following:
* The TAG will respond to each request brought to it by
a W3C Working Group.
* The TAG should respond to each request brought to it by
an external organization.
The TAG does not guarantee that all other requests brought to it
will be answered, nor that every TAG participant will read every
message on the list. For more information about how to raise
issues in a manner likely to get the TAG's attention, please
refer to:
http://www.w3.org/2001/tag/#tag-attn
Some of the tips suggested were:
* PC: I'd like requests for review to include links to relevant
discussion in a WG's archive.
* TB: I'll read things on the list that have an assigned
issue number.
RF asked: Should every WG and IETF be repeating all of their
arguments on this list?
DC: I am hoping that no one person can cause too much damage by
sending too much mail. I'm hoping that a community will form
around the TAG, that people will do their homework, that people
will point to existing discussions, etc.
------------------------------------------------
2. Lightweight process for publishing TAG findings
------------------------------------------------
The TAG touched on (but did not pursue in depth) a proposal from
Tim Bray [1] for a lightweight proposal for publishing "findings"
in a timely manner. This discussion is related to a discussion of
how the TAG expects to create an outline view of Web
Architecture, and then fill in that outline using past and future
experience.
[1] http://lists.w3.org/Archives/Public/www-tag/2002Jan/0083
------------------------------------------------------------------
3. nsMediaType-3: Relationship between media types and namespaces
------------------------------------------------------------------
The TAG started to discuss Tim Bray's email about principles and
corner cases of media types and namespace processing [2].
DC: The first two questions from the XML Protocol Working Group
were easy, the last one (nsMediaType-3) was hard.
TB: The issue to consider is when a protocol header says one
thing about a resource and the resource says something else.
The TAG reviewed an SVG diagram by Tim Berners-Lee and Martin
Dürst describing how to determine the character encoding of
content:
http://www.w3.org/2001/04/roadmap/xml-charset.svg
Note: If your browser doesn't yet render SVG images, please refer
to SVG implementation information [3].
TBL: This is the flow diagram that includes the bits from MIME
and from XML.
See also Appendix F. of XML 1.0 (Second edition): "Autodetection
of Character Encodings (Non-Normative)" [4].
DC: First choice in this flow chart is "check if there's a
charset". That presumes that charset means the same for all
mimetypes. Is that the case? As far as I know, parameter names
are local to each MIME type.
Action DC: Verify that parameter names are local to each MIME
type.
TB: After you've looked at the first 4 bytes, you pretty well
know the encoding.
TBL: The EBCDIC stuff is not standard.
DC: I can make up a new encoding right now and write an xml
document in it.
TB: TBL, the algorithm for determining character encodings
represented by the SVG diagram is correct. There's a real-world
difficulty, however: your average XML processor is more apt to be
able to tell the encoding by looking at the content than the
server is from looking at the local file name.
TBL: What follows from flow chart: the sender shouldn't specify a
charset unless the sender is absolutely sure.
TB: I agree.
/* On proxies and transcoding */
TBL: One of the ideas about putting the encoding at the top was
that a proxy could do it.
RF: I don't think an HTTP proxy can modify the content type.
TBL: Can it translate the character encoding?
RF: If you know what the encoding is, by declaring it up front,
this allows you to process the media type in that encoding alone.
The basic problem is that media type scanners that are working at
the level of the message can't afford to know the 5 bazillion
rules of 'charset' analysis. If there is a 'charset' defined for
that particular media type, it's supposed to use it. But aside
from text/*, there is no requirement that a media type have a
'charset' parameter.
TBL: Thanks, that clarifies that there's no requirement that a
third party be able to transcode (e.g., taking JIS and converting
to UTF-8). This could break digital signatures, etc.
TB: I think we've almost disposed of this issue:
a) Do we want to make a definitive statement on namespaces and
media types? I don't want to.
b) Should we say something about the case where the two
disagree?
TB: I think our position probably should be "There is an error
when the MIME type and the character encoding don't agree. The
MIME type is supposed to be right."
[2] http://lists.w3.org/Archives/Public/www-tag/2002Jan/0177
[3] http://www.w3.org/Graphics/SVG/SVG-Implementations.htm8
[4] http://www.w3.org/TR/REC-xml#sec-guessing
/* On the proper interpretation of text/plain etc. */
DC: If content is served with media type 'text/plain' and the
content happens to contain angle brackets with "TITLE", the
client should not interpret the content as HTML. I want users to
see the plain text. Today, if Internet Explorer sees <TITLE> in
the first 200 characters, it treats the content like HTML, even
if labeled 'text/plain'.
TB: Even if you label the content as svg or xml, IE will render
it as HTML. Back in the days when everything was HTML, it might
have been a forgivable sin for the client to correct something
from the server. Now with XML I think we should take a harder
stance. Also, this opens some glaring security holes.
TBL: I agree.
RF: A recent security patch to MSIE removed that feature.
PC: With MS hat loosely on, I'll track this down.
DC: Cf MS SDK 19 Dec 1997 version, which documents the
200 byte behavior.
------------------------------
4. Web Architecture evangelism
------------------------------
DC: When we decide that we're serious about what a specification
says, despite what popular software does, how can we get
Webmasterish folks to pay attention to us? Can we exploit the QA
Activity of W3C somehow?
IJ: Refer to "Common UA problems" [5]; we have talked about
having a "Common Server Problems" Note as well.
TB: W3C has a track record of not engaging much in evangelism.
DC: That has changed. The Membership endorsed both the QA
Activity and the TAG.
TBL: We do evangelize. We don't chase up miscreants. The WAI and
QA Activities are close to making lists of offending software.
PC: Can we point to Amaya behavior?
TB: That's not an effective way to get the word out.
TBL: We've done evangelism for SVG by demonstrating it in
Amaya. That's effective at the AC level.
PC: The XML Schema home page has links to a test suite
with 10,000 tests in it. That will help people do schemas
the right way.
DC: We need to be similarly aggressive for this.
TBL: Having a test area is a good idea.
DC: We could probably fit this test case in the HTTP/1.1
infrastructure (http://jigsaw.w3.org/HTTP/).
PC: Doing more basic Web QA would be useful.
See the paper at the QA Workshop on testing of HTTP.
http://www.w3.org/2001/01/qa-ws/pp/alex-rousskov-measfact
Action TB: Write up summary of findings about issues
w3cMediaType-1, customMediaType-2, and nsMediaType-3.
Done:
http://www.w3.org/2001/tag/2002/0129-mime
[5] http://www.w3.org/TR/cuap
-------------------------------------------------------------
5. uncefactLiaison-5:
Request to liaison with UN/CEFACT ebTWG Architecture Group
-------------------------------------------------------------
Invitation from Duane Nickull to participate in UN/CEFACT ebTWG
Architecture Group:
http://lists.w3.org/Archives/Public/www-tag/2002Jan/0021
PC: I suggest sending this request to Chris Ferris, chair of the
new Web Services Architecture group.
DC: Seconded.
Resolved: TAG declines this invitation to establish a liaison.
Action PC: Draft a response to Duane Nickull on www-tag with
recommendation to contact Web Services Architecture Working
Group.
-------------------------------------------------------------
6. xformsReview-4:
Request to review XForms in Last Call
-------------------------------------------------------------
Raised by Art Barstow:
http://lists.w3.org/Archives/Public/www-tag/2002Jan/0172
Resolved: The TAG declines this invitation to review the
document, but welcomes input from the XForms WG on specific
issues they want addressed.
Action SW: Respond to Art Barstow explaining our request
that the XForms WG make a more specific request.
Done:
http://lists.w3.org/Archives/Public/www-tag/2002Jan/0209
The TAG also discussed reviewing other documents, which
have now been added to a "Homework" section of the TAG
home page:
http://www.w3.org/2001/tag/#homework
=============================
Summary of action items
=============================
Open:
DC: Verify that parameter names are local to each MIME type.
Assigned: 28 Jan 2002.
PC: Draft a response to Duane Nickull on www-tag with
recommendation to contact Web Services Architecture Working
Group.
Assigned: 28 Jan 2002.
TBL: Find out what kind of editing access to the Web site will
be available to TAG participants.
Status: TBL Reports that CVS should be available. TBL
thinks that people should get collaborator
accounts at W3C.
Assigned: 7 Jan 2002.
Closed:
SW: Respond to Art Barstow explaining our request
that the XForms WG make a more specific request.
Done: http://lists.w3.org/Archives/Public/www-tag/2002Jan/0209
TB: Write up summary of findings about issues
w3cMediaType-1, customMediaType-2, and nsMediaType-3.
Done: http://www.w3.org/2001/tag/2002/0129-mime
DO: Take a first stab at writing a policy to summarize
resolution of issue w3cMediaType-1.
Assigned: 14 Jan 2002.
Subsumed 29 Jan 2002 by TB's action item to report findings
for issues w3cMediaType-1, customMediaType-2, and nsMediaType-3.
PC/IJ: Summarize input on www-tag (including technical
comments, liaison request). An initial categorization
of input may be found in the IRC log of the 7 Jan 2002
meeting.
Assigned: 7 Jan 2002.
--
Ian Jacobs (ij@w3.org) http://www.w3.org/People/Jacobs
Tel: +1 718 260-9447
Received on Wednesday, 30 January 2002 18:54:30 UTC