W3C

TAG Telcon

24 Mar 2011

Agenda

See also: IRC log

Attendees

Present
Peter Linss, Jonathan Rees, Ashok Malhotra, Noah Mendelsohn, Yves Lafon, Larry Masinter, Jeni Tennison (in part)
Regrets
Henry Thompson, Tim Berners-Lee
Chair
Noah Mendelsohn
Scribe
Peter Linss

Contents


<scribe> Scribe: Peter Linss

<scribe> ScribeNick: plinss

Date: 24 Mar 2011

late arrivals from JT, HST

WWW conference

NM: we usually cancel calls during conferences
... regrets from TBL, anyone else not available?
... regrets from LM
... will check and cancel later

YL: can scribe next week or week after

approval of last week's minutes

<noah> Minutes of the 17th http://www.w3.org/2001/tag/2011/03/17-minutes

NM: objections to approval of minutes from 17th?

<noah> RESOLUTION: Minutes of 17 March 2011 http://www.w3.org/2001/tag/2011/03/17-minutes are approved

Review of Ashok's draft #5

NM: additions to the agenda?

AM: comment on Dan's stuff, this works with client side storage. Probably worth speaking about when we talk about offline

Agenda item 4

NM: do we need to discuss preparation for IETF meeting

LM: I noticed there was a URN-BIS group, does anyone else think this is worth interest?

NM: will add as a another business item
... briefly consider action items relating to IETF meeting
... who best to report back form IETF meeting?

LM: we should work together

<noah> ACTION Larry and Henry to report on IETF Meeting Due 2011-04-07

<trackbot> Created ACTION-542 - And Henry to report on IETF Meeting Due 2011-04-07 [on Larry Masinter - due 2011-03-31].

<noah> ACTION-530?

<trackbot> ACTION-530 -- Henry S. Thompson to draft slides for IETF meeting, with help from Larry Due 2011-02-22 -- due 2011-03-15 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/530

NM: propose close action 530

<noah> close ACTION-530

<trackbot> ACTION-530 Draft slides for IETF meeting, with help from Larry Due 2011-02-22 closed

<noah> ACTION-519?

<trackbot> ACTION-519 -- Peter Linss to frame architectural opportunities relating to scalability of resource access Due: 2011-03-15 -- due 2011-03-15 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/519

NM: peter status of action 519

PL: been head down on CSS issues, haven't gotten to it

NM: keep this open

LM: don't think we need this for IETF

<noah> ACTION-519 Due 2011-04-05

<trackbot> ACTION-519 Frame architectural opportunities relating to scalability of resource access Due: 2011-03-15 due date now 2011-04-05

<noah> ACTION-519?

<trackbot> ACTION-519 -- Peter Linss to frame architectural opportunities relating to scalability of resource access -- due 2011-04-05 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/519

discussion of Ashok's draft

<noah> Ashok's draft: http://www.w3.org/2001/tag/2011/02/HashInURI-20110228.html

<noah> Concentrate on sections 4,5 & 6

NM: Ashok you suggested we concentrate on sections 4, 5 & 6
... show of hands who has read this

<noah> I have read the pertinent sections, not necessarily the older ones

NM: Yves and I have read, other's have not
... take comments form me, then Yves

YL: section 4 there is a point about using ? or #

<noah> YL: I think in section 4 we need to put more emphasis on distinction of identifying a document vs. an application

<noah> Reminder of my blog posting on this: http://blog.arcanedomain.com/2011/03/identifying-documents-in-web-applications/

AM: I had a question about what you were asking there, do you want more detail or were you asking for a fundamental difference in message

YL: I think it creates an architectural issue about follow your nose vs having to use javascript to understand the url

<Zakim> noah, you wanted to comment on new APIs

NM: I would like to see some more nuance in the analysis
... if you use ? you have to reload, so we have to use #
... that's true in old browsers, not in new browsers
... lets imagine a world in which apis let you take your choice
... as long as the domain doesn't change the page doesn't reload
... problems with # go away when you use ?

AM: the problem with that is that this state is not quite ready yet
... when the state changes, then we can do things differently

NM: for 6m people who downloaded FireFox 4, they got it
... there is the legacy problem
... but we are the architecture committee, not documenting current state
... given that this is the direction, can we compare pros and cons with both approaches
... with ? you can send to server, with # you cannot
... I think the TAG responsibility is to look to the future and tell that story

<Zakim> jar_, you wanted to warn that fragid FYN is a tar pit, do you really want to take it on?

AM: i can start thinking about that

JAR: I want to warn about 3rd bullet

<Yves> to respond to NM with ? you reload the page (or with a new URI you switch to a new URI), with # in many cases, you trigger retrieval of side data

<noah> JAR is talking about "A related fragment id meaning arises when one considers content-negotiation"

JAR: not sure thats' relevant
... we've been talking about that wrt 3023

<noah> JAR references RFC 3023: http://www.ietf.org/rfc/rfc3023.txt

JAR: there's the notion that frag id refers to an element
... my advice is to figure out how to avoid the subtext

AM: i could take out that bullet

<noah> Noah thinks this is yet another reason to avoid fragids...? does the right thing, more or less by definition, I think.

JAR: you refer to webarch or just remove it

NM: there are two ways we can read your comment
... go with encourage and #, but don't discuss conneg
... or use ? and conneg work as always
... we find places were architecture of frad is is fragile, conneg, ability send to server

JAR: there has to be a way to fudge this or you'll rat hole
... you want to be able to write this so RDF doesn't fail

<jar_> ... but you don't want to draw attention to RDF either...

<noah> My ultimate problem is that # is, strangely, architected pretty much to address into a representation, which is an odd thing to do into a URI. We just keep hitting ways in which that causes weakness. So, admitting that I'm belaboring this, ? seems stronger to me.

AM: you're pointing out a question we aren't going to be able to deal with in this document

YL: you pointed out when we change url in bar it triggers a reload, so we use #

NM: first class id for the we is URI without #

AM: you and i are looking at the world differently

<Zakim> noah, you wanted to respond to Ashok

AM; we aren't in a world of documents, but apps, state is document like

NM: we agree there
... what if we had ajax from day 1

<jar_> NM was saying the javascript # ids each feel like they're identifying a document...

NM: thing kid of look like documents, but they're just app states, so you can't email them around and such
... if the things you're navigating look like documents, lets make sure we can link to them
... its a huge loss if we can't do that
... we can't do this with tweet now with hackery with #!

AM: unless there are other comments, i'd like to speak about one of the other things form your blog
... because google apps uses ?, you can use it without javascript
... I did an experiment, I got a map, captured the link, turned script off and pasted the link
... what i got back was significantly different
... it works sometimes, but what you get without javascript is sometimes significantly less functional

NM: webarch tells a story about that
... take a URI and the server gets to decide what to do with it
... the server owns the resource
... they chose how to represent it
... we can warn people that will tend to be true

AM: that's worth speaking about in # vs ?

AM; you may get less function

NM: if we used #, you'd get nothing, because the # would not be sent to the server
... with ? theres a chance the representation you get will be useful
... tell that whole comparison
... you have enough input to think about redrafting?

AM: theres a fair amount to think about and type in

<JeniT> Noah, it's fine... hope to be there soon

<noah> ACTION-533?

<trackbot> ACTION-533 -- Noah Mendelsohn to schedule TAG discussion of #! (check with Yves) [self-assigne] -- due 2011-04-05 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/533

going through action items and revising dates

<noah> close ACTION-533

<trackbot> ACTION-533 Schedule TAG discussion of #! (check with Yves) [self-assigne] closed

<noah> ACTION-481?

<trackbot> ACTION-481 -- Ashok Malhotra to update client-side state document with help from Raman -- due 2011-03-01 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/481

<noah> ACTION-481 Due: 2011-04-12

<noah> ACTION-481?

<trackbot> ACTION-481 -- Ashok Malhotra to update client-side state document with help from Raman -- due 2011-03-01 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/481

<noah> ACTION-481 Due 2011-04-12

<trackbot> ACTION-481 Update client-side state document with help from Raman due date now 2011-04-12

<noah> ACTION-508?

<trackbot> ACTION-508 -- Larry Masinter to draft proposed bug report regarding interpretation of fragid in HTML-based AJAX apps -- due 2011-02-22 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/508

<noah> ACTION-538?

<trackbot> ACTION-538 -- Noah Mendelsohn to schedule discussion of Ashok's post F2F client state draft for 24 March [self-assigned] -- due 2011-03-22 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/538

<noah> close ACTION-538

<trackbot> ACTION-538 Schedule discussion of Ashok's post F2F client state draft for 24 March [self-assigned] closed

NM: we should defer offline apps until Dan is available

urn-bis

<noah> JAR: I think Larry wanted to know if we had input on that.

JAR: I wrote to larry about the sha1 urn namespace
... we have been talking about registries, and the tension between urn and http

LM: I just saw it on their agenda

JAR: been trying to think about something coherent to say about http and namespaces

<jar_> issue-50, jar + ht

LM: i'd appreciate if someone could look more in to background and report on it

JAR: i'll look at it

Conformance terminology for HTML5

NM: particularly in the case where they refer to applicable specifications

<noah> My concern became: http://www.w3.org/html/wg/tracker/issues/140

NM: they have a proposed resolution

<noah> Proposed resolution of the issue is now available from HTML WG chairs: http://lists.w3.org/Archives/Public/public-html/2011Mar/0574.html

NM: if you same something is a conformant html5 document, a UA should be able to conform with only the html5 spec

overdue pending action items

<noah> http://www.w3.org/2001/tag/group/track/actions/overdue?sort=owner

<noah> ACTION-475?

<trackbot> ACTION-475 -- Ashok Malhotra to write finding on client-side storage, DanA to review -- due 2011-03-21 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/475

AM: lets get client side state stuff finished first

NM: will you have this for the f2f?

AM: yes

<noah> ACTION-475 Due 2011-05-24

<trackbot> ACTION-475 Write finding on client-side storage, DanA to review due date now 2011-05-24

<noah> ACTION-523?

<trackbot> ACTION-523 -- Ashok Malhotra to (with help from Noah) build good product page for client storage finding, identifying top questions to be answered on client side storage -- due 2011-03-01 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/523

<noah> ACTION-523 Due 2011-04-05

<trackbot> ACTION-523 (with help from Noah) build good product page for client storage finding, identifying top questions to be answered on client side storage due date now 2011-04-05

<noah> ACTION-515?

<trackbot> ACTION-515 -- Larry Masinter to (as trackbot proxy for John) who will publish http://www.w3.org/2001/tag/2011/02/security-web.html, slightly cleaned up, with help from Noah and Larry -- due 2011-03-07 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/515

<noah> ACTION-515 Due 2011-04-12

<trackbot> ACTION-515 (as trackbot proxy for John) who will publish http://www.w3.org/2001/tag/2011/02/security-web.html, slightly cleaned up, with help from Noah and Larry due date now 2011-04-12

<noah> LM: Note websec group meeting in Prague next week, might make some progress there. 3:30 local time 30 March 2011

<jar_> IETF websec https://datatracker.ietf.org/wg/websec/

<noah> ACTION-508?

<trackbot> ACTION-508 -- Larry Masinter to draft proposed bug report regarding interpretation of fragid in HTML-based AJAX apps -- due 2011-02-22 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/508

<Larry> http://tools.ietf.org/wg/websec/agenda

<noah> NM: Bug report against media type registration?

<noah> ACTION-508?

<trackbot> ACTION-508 -- Larry Masinter to draft proposed bug report against HTML5 media type registration regarding interpretation of fragid in HTML-based AJAX apps -- due 2011-02-22 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/508

<noah> ACTION-481?

<trackbot> ACTION-481 -- Ashok Malhotra to update client-side state document with help from Raman -- due 2011-04-12 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/481

<noah> ACTION-508 Due 2011-04-19

<trackbot> ACTION-508 Draft proposed bug report against HTML5 media type registration regarding interpretation of fragid in HTML-based AJAX apps due date now 2011-04-19

NM: is there anything relating to client side state you want to re-open?
... summary of previous discussion

JT: I don't have anything to say about that

NM: can you verify that HTML5 pushstate allows to to change uri without reloading?

JT: yes

AM: yes

JT: both the address bar and the history can be manipulated

AM: there was a question that tim asked, are there any restrictions to that?

JT: yes
... you can't change domain or scheme without reload

Review of Pending Action Items

<noah> ACTION-421?

<trackbot> ACTION-421 -- Henry S. Thompson to frame the discussion of EXI deployment at a future meeting -- due 2011-01-21 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/421

<noah> ACTION-509?

<trackbot> ACTION-509 -- Jonathan Rees to communicate with RDFa WG regarding documenting the fragid / media type issue -- due 2011-03-07 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/509

JAR: I sent email to www-tag
... in RDFa they're using frag id's with media type xhtml or html in a way that's not sacntioned by media type registrations

<jar_> http://lists.w3.org/Archives/Public/www-tag/2011Mar/0097.html

JAR: summarizes email

<Larry> maybe this is also my action item bug report

<noah> ACTION-508?

<trackbot> ACTION-508 -- Larry Masinter to draft proposed bug report against HTML5 media type registration regarding interpretation of fragid in HTML-based AJAX apps -- due 2011-04-19 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/508

<noah> LM: Related?

LM: we need to reconcile what RDFa is doing with frag ids

JAR: more serious issue is 3023 frag id semantics

<Larry> note that there are discussions at IETF application area about MIME and registrations etc.

NM: you said that 3023 and 2054 have not yet caught up

JAR: I haven't heard anyone object to idea of updating media spec to be consistent with RDFa
... 3020 says xpointer semantics applies, frag id have to be treated as errors if not defined with id=
... same thing we went through with RDFa+xml
... this makes the problem with rdf+xml infect all of xml
... not high on working groups priorities
... don't see it as part of their charter

NM: they're defining a URI space

JAR: the frag is practice they're promoting is not explicit, just used in examples

NM: the examples contravene important specs that no one is changing
... these are not good practice, why are they in the document?

JAR: not sure what to do about it

<Zakim> JeniT, you wanted to talk about cross-format fragment identifiers for RDF

JT: using uris with # is just very common across rdf
... when the uri is requested, you may get a different format back , but you want to keep the uri relevant

<noah> I'm starting to wonder whether we need to just change the claim that # is resolved relative to media type. That's a change that would nail a lot of things, including conneg, no?

JT: its a nasty thing about frag ids when you have multiple formats representing the same uri
... its not about a particular representation, its about identifying an abstract construct

JAR: I agree, that's what webarch says

NM: do we need to tell 3986 to change?

<Yves> noah, but it will raise lots of conflicts, if # was suddenly no longer bound to media type

NM: semantic web has deployed ignoring it, we need another story

<Yves> including needing a # registry

<Larry> there have been calls for URI-scheme relative fragment identifiers

<Larry> i don't understand how to make # *not* depend on media type

JAR: there are two different fixes, make 3023 more like webarch

LM: first step is to write down what the problem is
... my suggestion is look at some way of adding this issue, can we get a coherent story and examples
... i have an action to revise document

<noah> ACTION-472?

<trackbot> ACTION-472 -- Larry Masinter to update the mime-draft based on comments & review -- due 2011-04-09 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/472

LM: asks if someone else can take action

NM: summarizes action 472

JT: i'm willing to take a look

<noah> ACTION Jeni To propose addition to MIME/Web draft to discuss sem-web use of fragids not grounded in media type Due: 2011-04-05

<trackbot> Created ACTION-543 - Propose addition to MIME/Web draft to discuss sem-web use of fragids not grounded in media type Due: 2011-04-05 [on Jeni Tennison - due 2011-03-31].

<noah> ACTION-543 Due 2011-04-05

<trackbot> ACTION-543 Propose addition to MIME/Web draft to discuss sem-web use of fragids not grounded in media type Due: 2011-04-05 due date now 2011-04-05

<noah> ACTION-529?

<trackbot> ACTION-529 -- Noah Mendelsohn to schedule telcon discussion of a potential TAG product relating to offline applications and packaged Web -- due 2011-02-17 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/529

<noah> ACTION-534?

<trackbot> ACTION-534 -- Jonathan Rees to create issue page relating to Harry Halpin's concerns about 200/303 responses -- due 2011-03-03 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/534

<noah> close ACTION-534

<trackbot> ACTION-534 Create issue page relating to Harry Halpin's concerns about 200/303 responses closed

NM: jeni are you avaialable next week

JT: yes

NM: we will have a call next week

meeting adjourned

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2011/03/31 20:56:50 $