W3C

Technical Architecture Group Teleconference

29 Nov 2012

Agenda

See also: IRC log

Attendees

Present
Masinter, Noah_Mendelsohn, ht, Ashok_Malhotra, plinss, Yves, TimBL
Regrets
jeni
Chair
noah
Scribe
Larry Masinter

Contents


<scribe> scribenick: masinter

<scribe> Scribe: Larry Masinter

date: 29 November 2012

Administrative items

NM: do we want to have a Jan F2F? and what do we need to do?

<Noah> LM: Would like to discuss whether Polyglot spec is normative

ht: there's a "rag" going on ... (ht breaking up)

ashok: is this about the schema.org thing?

<ht> microdata going to CR

<ht> A certain amount of upset on public-html about this

<ht> [quite a lot, actually]

<Noah> Shall we discuss later?

<ht> We should discuss, but perhaps not w/o Jeni

LM: I think the topic for the tag is "what does it mean for a spec to be 'normative'"

and CR exit criteria

<Noah> http://www.w3.org/2001/tag/2012/11/08-minutes

RESOLUTION: minutes of 8 nov 2012 approved

<Noah> ACTION-743?

<trackbot> ACTION-743 -- Noah Mendelsohn to schedule F2F discussion of issues report to Jeff Jaffe, see ACTION-563 -- due 2012-10-04 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/743

<Noah> close ACTION-743

<trackbot> ACTION-743 Schedule F2F discussion of issues report to Jeff Jaffe, see ACTION-563 closed

Will there be a Jan F2F?

LM: I'm prepared to host, even if I don't get reelected

(discussion about swapping West Coast vs. Cambridge for Jan & March meetings)

<Noah> ACTION: Noah to investigate hosting Cambridge meeting Jan 15-17 [recorded in http://www.w3.org/2012/11/29-tagmem-irc]

<trackbot> Created ACTION-768 - Investigate hosting Cambridge meeting Jan 15-17 [on Noah Mendelsohn - due 2012-12-06].

(discussion about Sunday, Monday Tuesday or monday tuesday wednesday for cambridge meeting)

LM: any feedback on our FPWD drafts? can we move them forward?
... solicit explicit review from targeted individuals?
... i'm just surprised we've gotten _no_ feedbaack

agenda review: jaffe list, polyglot, microdata

polyglot to rec? the term "normative"

<Noah> LM: There was formal objection to moving the polyglot doc to REC, with the objection that it shouldn't be normative

<Noah> LM: Suggestion was Note rather than REC. I think the WG has the scope to decide that if they can come to consensus.

<Noah> LM: That said, the nature of the arguments raised some broader issues the TAG might clarify:

<Noah> LM: E.g. whether doc has had sufficient review? What if docs might conflict with other docs? What it means for a document to be normative and make normative reqts?

<Noah> LM: Might SHOULD/MUST/MAY apply outside scope of document.

<Noah> LM: Could send my thoughts to TAG list. I'm suggesting we focus not primarily on the decision directly, but on clarifying the criteria, in part to help inform the Director should the issue come to him.

<Noah> LM: Reading would be HTML WG Bug, formal objection, subsequent discussion.

NM: the process document uses 'normative' all over the place, but without definition
... reading would also include existing process & recs

ashok: is there any time constraint on this? do they want an answer by 1 jan ?

lm: i imagine yes

<Noah> http://www.w3.org/2005/10/Process-20051014/process.html#rec-modify

<Noah> "The following sections discuss the management of errors and the process for making normative changes to a Recommendation."

<Noah> A correction is first "proposed" by the Working Group. A correction becomes normative -- of equal status as the text in the published Recommendation -- through one of the processes described below.

lm: i'll take an action to come up with a summary of topics in the next couple of days, if people want to discuss this next week

(Tim joins)

moving Jan TAG meeting

<Noah> RESOLUTION: The TAG will meet in Cambridge, MA 14-16 (Mon-Wed) January 2013.

<ht> March is still blocked for me

<ht> Let me check earliest date I could make

RESOLUTION: The TAG will tenatively move its March 2013 meeting to Adobe in San Jose

<Noah> Rumor is 19-21 March.

<ht> Right, I can't make it until 9 April

The term Normative

<Noah> TBL: What are the concerns?

<ht> +1 to talking about this after prep. material from Larry

<Noah> LM: What does it mean for a document or reference to be normative.

<Noah> LM: ...missed some other details..

<Noah> LM: There are details I'd like thijs

<Noah> TBL: To me, it's a relationship between specs. If you conform to a spec, then you also must conform to what it references normatively

<Noah> LM: One case that concerns me is overlap. What are the rules for normative specs overlapping.

<Noah> LM: Recommendations are from W3C, that happen to be prepared by WGs.

<Noah> LM: ...scribe is missing some points about community review...

<Noah> LM: Questions like that.

<Noah> TBL: That's a longer term thing, vs. dealing with this particular one.

<Noah> LM: I think the WG has scope to take next steps on this one. If a formal objection were to be lodged, we (and you as director) should (have good basis for responding)

<Zakim> Noah, you wanted to say we have a precedent on overlap

<ht> Meant as an endorsement

nm: there's a precedent with the authoring spec

lm: we have the precedent, we just don't have the policy

<Noah> NM: Right, agreement was reached that both HTML5 and Authoring spec are normative RECs, though overlapping. Disagreement is evidence of bug. I think that until bug is resolved, the rules say HTML5 wins.

issue list "for Jeff"

<Noah> F2F Discussion: http://www.w3.org/2001/tag/2012/10/08-minutes#item04

<Noah> ACTION-753?

<trackbot> ACTION-753 -- Larry Masinter to do first draft of technical issues list for Jeff -- due 2012-10-22 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/753

<Noah> Note from Larry: http://lists.w3.org/Archives/Public/www-tag/2012Nov/0043.html

<Noah> NM: TAG has standing request from Jeff to highlight technical/arch issues that should concern him

<Noah> NM: Larry has drafted an initial list for discussion

<Noah> LM: Five issues:

<Noah> LM 1) We did some work relating to Publishing and Linking that may be more general. Quoting Larry's note, which he is reading:

<Noah> "The TAG has been discussing areas where governance - the desire of legal, regulatory, administrative, or contractual relationships to regulate communication - affects web architecture. While W3C has ongoing efforts in privacy, security, accessibility and internationalization are oriented toward insuring that standards specified by W3C specifications can accommodate at least some of the

<Noah> governance requirements, there is a need to do more. For example, the "publishing and linking" FPWD from the TAG addresses some of the issues of linking vs. copying vs. embedding that have been at the center of some controversies with regard to the application of governance around copyright, censorship of unwanted material, identity, logging, and many other issues. While the TAG discussed a

<Noah> more general framework (http://www.w3.org/2001/tag/doc/governanceFramework.html) the larger topics seem outside the scope of what the TAG can take on alone."

ashok: at TPAC, we had lunch with the AB, and this question came up, and Steve Zilles (Either speaking for himself or the AB), thought it would be a good idea if we went further in this direction.
... the question was: should we be doing more work on web/governance was?

<Noah> Noah: What is it we need to alert Jeff about?

nm: this may be good work for the TAG, but what do we need to alert jeff about?

LM: the one i have this time for issue 1 is censorship

<Noah> LM 2. Security is an arms race, and the bad guys are winning. (and we aren't doing security reviews early enough or thoroughly enough before RECs are baked)

<ht> HST regrets for next week (6 December)

(HT drops)

<Yves> larry, examples?

<Noah> LM: Suggest hiring more people, giving editors and chairs security training

<Noah> YL: Do you have a specific spec in mind other than HTML5

<Noah> LM: Well, HTML5

<Noah> YL: There is cooredination on CORS, for example

<Noah> LM: Uh, OK.

<Noah> LM: Yves, how do you see the overall situation?

<Noah> YL: I think there is coordination for some clearly identified points like cross-site in CORS, but we may be missing others.

LM: The problem is not "lack of coordination", it is "insufficient resources"
... there isn't sufficient review in IETF, either

<Noah> NM: http://www.w3.org/Security/

<Yves> http://www.w3.org/Security/Activity.html

<Noah> NM: Don't you think the right way to look at this is that the Security Activity "owns" this, with whatever help from the TAG

<Noah> LM: There are resource issues

<Noah> Web Application Security Working Group

<Noah> Web Cryptography Working Group

this is more a matter of publishing documents which open security exploits

<Noah> AM: Do these cover the ground?

<Noah> LM: The issue is, it's easy to publish a spec that opens unanticipated security problems. Working groups chartered to produce "new locks for new doors", while not checking that we're unlocking the old doors.

http://lists.w3.org/Archives/Public/www-tag/2012Nov/0043.html

<Noah> NM: Are we telling Jeff "There's a problem that specifications may be have or be causing security problems not covered by existing WGs, and that are not being suitably considered before the recs come out"

we recommend more security training? more security staff?

before WDs are implemented and deployed

<Noah> NM: Are there other TAG members who think we should put it on the list for Jeff?

<Noah> Silence.

<Noah> LM: 3. Boundary between "Web" and "Internet Applications" evolving

<Noah> LM: It's blurring. Web capabilities are expanding with ???

<Noah> "As web capabilities expand, the boundary between "Web" and "not web" continues to blur, and the overlap between W3C and IETF also becomes more problematic. The definition of what the "Web" is compared to other internet applications requires a more principled discussion and coordination. This applies in general to a wide variety of areas (WebRTC RTCWeb) but specifically for two technologies

<Noah> fundamental to the web but work in the IETF without significant W3C involvement:"

<Noah> NM: Reads items 4-9 and invites expressions of interest from TAG members

<Noah> AM: Interested in Web of things

timbl: RFID things

<Noah> NM: What's the message to Jeff?

timbl: The URL spec is quite specific, the rest are high level

<Noah> TBL: [says a bit about each one]

yves: what is new since the last report we made to jeff?

<Noah> http://lists.w3.org/Archives/Public/www-tag/2011Aug/0030.html

<Yves> http://lists.w3.org/Archives/Public/www-tag/2012Feb/0049.html

LM: I want to close my action item
... I would like someone else to draft another list :)

<Noah> ACTION-753?

<trackbot> ACTION-753 -- Larry Masinter to do first draft of technical issues list for Jeff -- due 2012-10-22 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/753

LM: I think Jeff's heard about most of these
... we shouldn't forward a list we're not willing to defend

peter: Jeff doesn't want a formal list twice a year
... It isn't what Jeff asked for, what he wants is a timely heads up, when things come up

<timbl> bye

<Noah> ACTION: Noah to informally inform Jeff that we did not at this time identify urgent technical matters for his consideration [recorded in http://www.w3.org/2012/11/29-tagmem-irc]

<trackbot> Created ACTION-769 - Informally inform Jeff that we did not at this time identify urgent technical matters for his consideration [on Noah Mendelsohn - due 2012-12-06].

Summary of Action Items

[NEW] ACTION: Noah to informally inform Jeff that we did not at this time identify urgent technical matters for his consideration [recorded in http://www.w3.org/2012/11/29-tagmem-irc]
[NEW] ACTION: Noah to investigate hosting Cambridge meeting Jan 15-17 [recorded in http://www.w3.org/2012/11/29-tagmem-irc]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2012-12-12 19:11:34 $