W3C

TAG telcon

29 Jan 2009

Agenda

See also: IRC log

Attendees

Present
Larry Masinter, T V Raman, Jonathan Rees, Stuart Williams, Henry S. Thompson (in part), Dan Connolly, David Orchard, Noah Mendelsohn, Tim Berners-Lee, Norm Walsh (in part)
Regrets
Ashok Malhotra, Norm Walsh (in part), John Kemp
Chair
Stuart Williams
Scribes
Henry S. Thompson, Noah Mendelsohn

Contents


<raman> CACM Article: The Rest Of The Story -- See http://emacspeak.sourceforge.net/raman/publications/beyond-web20-cacm-2009/

<scribe> Agenda: http://lists.w3.org/Archives/Public/www-tag/2009Jan/att-0108/tag-weekly.html

Admin

SW: Agenda?

DC: Flyby of OAuth

SW: At the end, if poss.

NM: Requests came in to fix some broken links
... in the uriMediaType-9 finding: http://lists.w3.org/Archives/Public/www-tag/2009Jan/0118.html.

SW: Minutes from 22 Jan?

DO: Pending. . .

SW: Hold approval to next week
... This is my last call or meeting of any kind in the chair, NM will chair our meeting next week, 5 Feb, and going forward

DC: Regrets for 5 Feb

SW: JR to scribe

DO: Also last official meeting for DO and NW. . .

SW: Traditional allows outgoing TAG members as guests until the end of the first F2F

NM: I'm happy to go with that
... Anyone with a concern can say so now, or in private email

SW: No obligation to attend outside official terms, but informal overlap helps the transition

ISSUE-57 (HttpRedirections-57): The use of HTTP Redirection

<Stuart> http://www.w3.org/2001/tag/group/track/issues/57

<noah> http://lists.w3.org/Archives/Public/www-tag/2009Jan/0114.html

<DanC> ACTION-200 due next week

<trackbot> ACTION-200 Revise "Uniform Access to Metadata" (needs title change) to add XRD use case due date now next week

JR: ACTION-200, to add a use case, is nearly ready, but not done yet

SW: Topic for f2f?

JR: Yes, I think ISSUE-57 should be on the f2f agenda

<DanC> (darn; date of next ftf is not on http://www.w3.org/2001/tag/ )

SW: ACTION-184 appears to have generated some activity

JR: I've worked through the 303 story with Lisa D of IETF in a series of emails
... DBooth has pointed out the value-add of having a URI for the redirected-to URI as well

JR: I was accused of undermining httpRange-14, I'm in favor in general, but I thought pushing hard on IANA was going too far

LM: What's the issue?

SW: Entries in IANA registry for link relations

<DanC> e.g. http://www.iana.org/assignments/relation/describedby

<noah> FWIW: that URI provided by DanC is 404

SW: Given that they are intended to denote relations
... we would like 303s, per httpRange-14

<jar> Larry asks: What problem does the httpRange-14 rule solve?

SW: [glosses the httpRange-14 resolution]

HT: Is it the case that you explained the situation to Lisa, she understood, but you got no agreement to implement 303 redirects at these URIs

JR: yes, oversimplifying a bit

TBL: [example of a URI for a protein returning with 200 leading to confusion]

<Stuart> <protein> dc:creator <someDocumentAuthor>

LM: IETF/IESG have a complex relationship with IANA
... It's not always easy for IESG people to make things happen on that website

<timbl> In way, W3C would be more logical advisor to IANA's web site.

LM: [Some discomfort over the assumptions behind the httpRange-14 resolution, and with the resolution itself --- scribe failed to record in detail]

TBL: The issue resolution may not be perfect, but it has the advantage that we can stop spending huge amounts of person-time continuing to discuss it

<jar> sure... but I just did (with Mark & Lisa) - the resolution didn't help me avoid the talking-time...

<noah> I fear we are about to back into the whole httpRange business. If it's worth reopening, I think we should do it with great care, and perhaps after a few weeks' of sitting on the preliminary decision to do so. History suggests that alternative, equally imperfect, solutions will be difficult and time consuming to do better than we already have.

<jar> Is relation:describedby a relation or a document?

DC: If the argument didn't persuade, then maybe we should reopen the issue

<Stuart> Larry... your opinion is certainly noted... and it's quite in order for you to offer it.

DC: If the IANA website were abusing web security guidelines, we would push hard until we got a resolution

LM: I think it's at least worth getting a writeup of the outcome of this effort

<DanC> (Is the URI standard quite clear on " Why can't a single string identify a relation for some purposes and a document for others?")

<Stuart> Larry... FYI some of the working consensus beyond httpRange-14 is detailed in http://www.w3.org/TR/cooluris/

DC: The crucial point is "Why can't a single string identify a relation for some purposes and a document for others?"

LM: I think it's worth distinguishing between identify and denote---when I say "I'm parked out back" it's not me, it's my car, and that's not a problem

JR: True, but not I think exactly relevant
... Having summarized the standard arguments, I went to the RFCs
... I think RFC 2616 says you can't return a 200 for a URI which identifies a relation
... I also tentatively canvassed suggesting a 404 response for non-information resources. There was, not surprisingly, pushback.

<DanC> -1 404

JR: But Mark Nottingham basically said that 2616 wasn't meant to be read that way
... So that line didn't fly either
... Lisa didn't think going directly to IANA would help
... Note that the registry hasn't been published yet, which is why the URIs aren't there yet

TBL: They could use a hash?

JR: No, because they want to use relative URIs

<timbl> "If the relation-type is a relative URI, its base URI MUST be considered to be 'http://www.iana.org/assignments/relation/'" (in http://tools.ietf.org/html/draft-nottingham-http-link-header-03)

NM: As JR said, the points we're hearing are the ones we discussed at length in the httpRange-14 discussion
... Uniformity is a real value, and "using the same string..." compromises that
... Either we should re-open this, and prepare to spend a substantial amount of time on it
... or we should just accept that our advice will never be completely adopted

TBL: Given that IANA haven't spent a lot of time using URIs like this in ways that raise the problems, we could spend a lot of time trying to educate them, and then we would have to do that for many other people
... It's not that different from the move from plain text to HTML
... We could just wait
... Or we could ask to have the registry run at www.w3.org instead, and then we can do the redirect

TBL: There is real growth in systems, particularly in the Linked Data area, which depend on the 303 convention, and I would not like to make trouble for them

JR: I believe putting a w3.org URI in an RFC is not allowed

<noah> As chair for next week, I would like to come out of discussions like this knowing whether we expect to schedule further discussion next week, and if so with what goals?

SW: So, JR, is ACTION-184 done?

JR: Yes.

<DanC> (yes, ACTION-184 is done to my satisfaction)

TBL: We could send them a HOW-TO for Apache servers. . .

JR: No-one objected on the grounds of difficulty

SW: Should the "move registry to w3.org" be put on the W3C-IETF Liaison call agenda

DC: Too slow

<Stuart> close action-184

<trackbot> ACTION-184 contact Lisa D of IESG, cc www-tag, to explain about 303, with cool URIs and webarch as references. closed

<DanC> ACTION: Jonathan to raise moving the registry to w3.org with Mark Nottingham [recorded in http://www.w3.org/2001/tag/2009/01/29-minutes.html#action01]

<trackbot> Created ACTION-217 - Raise moving the registry to w3.org with Mark Nottingham [on Jonathan Rees - due 2009-02-05].

<timbl> Jar, is your discussion with IANA folks in email or unrecorded?

<jar> Tim, I didn't talk to IANA. The conversation is in private email, with Mark N (Yahoo!) and Lisa D (IETF).

SW: We've had a reminder from the POWDER WG that they are nearing the end of their Last Call period, about to request CR: http://www.w3.org/mid/497DD071.2070707@philarcher.org
... Anyone interested in reviewing

<Zakim> DanC, you wanted to note http://www.w3.org/2007/powder/Group/powder-dr/20090120-diff.html#semlink

<DanC> "http://www.iana.org/assignments/relation/describedby"

DC: In section 1.4.1 of their document, the very URI we were just discussing appears

<DanC> "http://www.iana.org/assignments/relation/describedby" in powder...

DC: Are we happy that they think it's a relation and IANA are serving it as a document?

<timbl> No!

<DanC> (wierd... which is the document in last call? I'm confused...)

<jar> I think it's http://www.w3.org/TR/2008/WD-powder-dr-20081114/

<timbl> http://www.w3.org/TR/powder-dr/

<jar> well, i thought so, because that's the 'latest version' link

<DanC> "This is the Second Last Call draft"

<DanC> no iana link in http://www.w3.org/TR/2008/WD-powder-dr-20081114/#semlink

http://www.w3.org/2007/powder/Group/powder-dr/20081205.html

<DanC> what's wdrs:describedby ?

<DanC> ah... http://www.w3.org/2007/05/powder-s#describedby

<HST> I think http://www.w3.org/2007/powder/Group/powder-dr/20081205.html#atom is later

<timbl> Bug: You click on "latest version " and you get an earlier version from http://www.w3.org/2007/powder/Group/powder-dr/20081205.html

<timbl> http://www.w3.org/TR/2008/WD-powder-dr-20081114/

<DanC> http://www.w3.org/TR/powder-dr/#semlink http://www.w3.org/TR/2008/WD-powder-dr-20081114/#semlink

DC: I propose to focus on the published Last Call draft, dated 2008-11-14
... describedBy is central to their design, right?

<DanC> the full URI is http://www.w3.org/2007/05/powder-s#describedby

<timbl> xmlns:wdrs="http://www.w3.org/2007/05/powder-s#"

<timbl> collapsedescribed by

<timbl> Type expandloadedProperty

<timbl> Comment An RDF property to link to a POWDER document. Provided for use in RDFa, ATOM etc.

<timbl> IsDefinedBy expandfetchsemlink

<timbl> Label described by

<timbl> Range opt off expandloadedPOWDER document

<timbl> --------

<DanC> (wierd... I can't find an HTML spec citations)

"using the link element to relate an XHTML document"

<HST> http://www.w3.org/2007/powder/Group/powder-dr/20081114.html#assoc-markup

<timbl> http://www.w3.org/2007/05/powder-s#describedby works

<Stuart> Full HTML source is at http://www.w3.org/TR/powder-dr/example_4_2.html and we ought to be able to follow noses to specs :-)

[The above reflects a period when the discusion was fragmented as different participants chased pointers and looked at examples -- scribe did not catch very much of the discussion, sorry]

<DanC> I propose we say: at a glance, we can see some struggles around HTML spec, but that's understandable; otherwise, noting we didn't do a thorough review, we don't find any architectural issues

DC: Straw poll on the above suggestion

LM: Only just looking at this for the first time, it's interesting, I have a lot of questions

<timbl> I would prefer from taste and UI "described by" to be called "description document"

TVR: They should get their story right wrt what part of (X)HTML they are depending on

<DanC> (I'm already up to 2 saying "let's study this more"; doubt my proposal is going to fly)

<jar> I assume they've come to peace with the wdrs:describedby vs. iana.org.../describedby issue - aliases are not so nice. but MNot's thing isn't an RFC yet, so they can't use it, as their pub date precedes his.

TVR: They need to be very explicit about the (X)HTML connections, which they haven't done
... it's a bit of a mess as of now

<timbl> http://www.w3.org/TR/powder-grouping/

<timbl> The set of documents

SW: I need a volunteer to coordinate, or this gets left to individuals

SW: OK, hearing none, I will tell Phil Archer that he may or may not hear from individuals, but no official TAG input will be coming

<DanC> +1 "several members are studying and may have comments" as stuart said

ISSUE-51 (selfDescribingWeb-51): (short) well known formats and URI based extensibility

NM: No progress, waiting on last week's minutes

<dorchard> irc is at http://www.w3.org/2009/01/22-tagmem-irc

<DanC> ACTION-216 due next week

<trackbot> ACTION-216 Publish SDW finding, with 4 changes as noted in minutes of 22 Jan 2009 tag telcon due 29 jan 2009 due date now next week

NM: I will publish internally to tag@w3.org to enable last-minute review

DC: Critical path to publish was HST, NM and NW

NM: NW dependency was for the diagram

TBL: I am happy with what NW did to the diagram

DC: Critical path is down to NM and HST

ISSUE-41 (XMLVersioning-41): (short) What are good practices for designing extensible XMLlanguages and for handling versioning?

SW: DO, where are we?

DO: I believe I am going to do some final cleanup, and then publish it as a white paper over my name, which did not command TAG consensus

SW: We agreed that we need to carefully minute what we decided at the f2f, as the record isn't perfectly clear

NM: No-consensus, but a TAG document, editor DO, or personal document, author DO, with substantial impact from TAG discussion

TVR: I prefer the latter

SW: That's what I thought we decided

<noah> Proposed resolution: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by Dave Orchard (not the TAG) as a note, acknowledging history of TAG work and input.

<DanC> works for me

<DanC> yes, W3C Note

<noah> Proposed resolution: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by Dave Orchard (not the TAG) as a W3C Note, acknowledging history of TAG work and input.

<noah> NM: I thought it would be a note from Dave as an individual?

HST: I don't think individual W3C Notes are possible

DC: Correct

<jar> I don't think WG notes require group consensus on content; just consensus on desirability of publication

<jar> this is from memory.

<noah> Proposed resolution: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by Dave Orchard (not the TAG) as a W3C Note, acknowledging history of TAG work and input, and making clear lack of TAG consensus on the contents.

<DanC> indeed, strike "not the TAG"

<noah> Proposed resolution: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by David Orchard as author as a W3C Working Group Note, acknowledging history of TAG work and input, and making clear lack of TAG consensus on the contents.

<dorchard> Jonathan, I had hoped that the TAG would publish as a NOTE. I'm still very disappointed in this result.

<DanC> aye

<jar> sorry? isn't that what we just said?

<noah> Can notes have authors, or just editors, per the process?

<jar> dorchard, I don't understand your disappointment

SW: I want to be sure this is what we decided in Dec., so discouraging input from those not there.

Proposed resolution: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by as a W3C Working Group Note, acknowledging history of TAG work and input, and making clear lack of TAG consensus on the contents.

SW: Anyone opposed? Any abstentions?

Silence.

RESOLUTION: Clarifying the resolution reached at the 9-11 Dec. 2008 F2F, the versioning finding will be published by as a W3C Working Group Note, acknowledging history of TAG work and input, and making clear lack of TAG consensus on the contents.

Issue-20 Error handling

SW: There's an action to Henry http://www.w3.org/2001/tag/group/track/actions/199
... There was some back and forth on the list about tag soup and error handling. Henry had action to follow up. Larry made a useful posting. Should we schedule further discussion?

<ht> HST will contribute to the resurrected XML / Errors / Postel's Law thread by the time his action is due, 30/1/09

[HST leaves the call]

LM: I'd like some chance to prepare for discussion.

<DanC> +1 2 weeks

<scribe> ACTION: Noah to schedule discussion of ISSUE-20 for 12 Feb 2009 telcon [recorded in http://www.w3.org/2001/tag/2009/01/29-minutes.html#action02]

<trackbot> Created ACTION-218 - Schedule discussion of ISSUE-20 for 12 Feb 2009 telcon [on Noah Mendelsohn - due 2009-02-05].

ISSUE-58 Scalability of URI access

SW: Let's skip since Norm isn't here.
... Norm, welcome.

[NW joins the call]

NW: Re ACTION-163, I am working with Ted, and will continue to do so after my tenure ends. Target 19 Feb 2009.

<masinter> missed last 10 minutes of IRC

I've updated date on action 163

Issue-1 w3cMediaType

<Stuart> I'm updating the status to pending review; I suppose we should announce the decision(s) that we have made and solicit feedback, esp from Mark Baker and the (heirs of?) the XMLP WG.

DC: Someone made a joke about this pending for almost a decade. We made a few pertinent decisions, especially to approve the finding.
... Tried to figure out status in July 2006, didn't write anything down.

SW: What decision did we make?

DC: To publish the finding.

SW: Does the finding answer the question?
... So proposal is to address the xmlp group now?

LM: Is this an architectural issue or process?

DC: We accepted as architectural. We goofed.

LM: There are both technical and process issues.

<DanC> Internet Media Type registration, consistency of use

<DanC> TAG Finding 30 April 2004

The finding is at: http://www.w3.org/2001/tag/2004/0430-mime

<DanC> mark baker's original question http://lists.w3.org/Archives/Public/www-tag/2002Jan/0063

We recommend that section 7.1 of [RFC3023] be amended to something like the following:

The use of the charset parameter, when the charset is reliably known and agrees with the encoding declaration, is RECOMMENDED, since this information can be used by non-XML processors to determine authoritatively the charset of the XML MIME entity.

LM: W3C is change controller of 3023, so you have authority to do that.

<Norm> I think we need to find way to get 3023 finished

NM: Write token for 3023bis is somewhere between Chris and Henry, right?

SW: So, we can't close this right now.

<masinter> thanks, all, need to drop off phone

Thank yous to outgoing members

<DanC> +1 thanks Dave, Norm, Stuart

<ht> HST would like to be recorded as joining in the motion of thanks to SW which he expects will be forthcoming

<Norm> Thanks to you all!

TBL: Thank you to Dave, Norm and Stuart for your wonderful service. Working with you has been a great pleasure.

<Norm> Au revoir.

NM: As incoming chair, I have growing insight for just how much great work you've done for us Stuart, thank you!

<jar> Thanks Stuart - it's been a pleasure to have you preside

<timbl> Never mind .. my machine has learned them

<jar> looking forward to it.

Summary of Action Items

[NEW] ACTION: Jonathan to raise moving the registry to w3.org with Mark Nottingham [recorded in http://www.w3.org/2001/tag/2009/01/29-minutes.html#action01]
[NEW] ACTION: Noah to schedule discussion of ISSUE-20 for 12 Feb 2009 telcon [recorded in http://www.w3.org/2001/tag/2009/01/29-minutes.html#action02]
Minutes formatted by David Booth's scribe.perl version 1.134 (CVS log)
$Date: 2009/02/05 19:41:35 $