W3C

Technical Architecture Group Teleconference

04 Feb 2010

See also: IRC log

Attendees

Present
Dan Connolly, Jonathan Rees, Daniel Appelquist, John Kemp, Henry S Thompson, T.V. Raman, Ashok Malhotra, Larry Masinter, Tim Berners-Lee
Regrets
Noah Mendelsohn
Chair
Dan Connolly
Scribe
John Kemp

Contents


<trackbot> Date: 04 February 2010

<DanC> scribe: johnk_

Convene

DC: Can you scribe next week, Henry?

HT: OK

<DanC> minutes ok? http://www.w3.org/2001/tag/2010/01/28-minutes

<DanC> "DKA: As a member of that WG, not sure I can concur "

I can't understand DKA very much at all...

<DanC> PROPOSED: to approve http://www.w3.org/2001/tag/2010/01/28-minutes ammended to note that "DKA: As a member of that WG, not sure I can concur" should read "DKA: As a former member..."

<DanC> DKA, is it enough to note the correction in today's minutes?

<DKA> fine

<DanC> PROPOSED: to approve http://www.w3.org/2001/tag/2010/01/28-minutes ammended to note that "DKA: As a member of that WG, not sure I can concur" should read "DKA: As a former member..." and to note that the security stuff is unclear

JK: I found the 'security' section unclear

<masinter> if there are errors in the minutes, send the errors to me and i'll update

<DKA> It should be "when I sat in on the first working group meeting as an observer"

JK: Happy to just note that and move on

<DanC> ACTION: DanC to take approval of minutes 28 Jan offline [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action01]

<trackbot> Created ACTION-385 - Take approval of minutes 28 Jan offline [on Dan Connolly - due 2010-02-11].

TVR: Take discussion offline

ISSUE-53: ACTION-231 & ACTION-232 conneg, generic resources

<DanC> close ACTION-232

<trackbot> ACTION-232 Follow-up to Hausenblas once there's a draft of HTTPbis which has advice on conneg closed

LM: Sent a note to requestor - should close the item

ISSUE-53: ACTION-231 & ACTION-232

LM: Propose to close the actions

HT: Why did we (re-) open this?
... Conneg text hasn't changed, has it?

LM: In editors draft

http://lists.w3.org/Archives/Public/ietf-http-wg/2010JanMar/0033.html

<DanC> I propose that HTTPbis changeset 745 section.4.p.5, along

<DanC> with the Nov 2006 finding, addresses our ISSUE-53, Generic Resources.

<DanC> http://trac.tools.ietf.org/wg/httpbis/trac/changeset/745

<jar> ht, description of http://www.w3.org/2001/tag/group/track/issues/53 seems pretty clear to me

<DanC> so RESOLVED.

HT: I'm happy with this

CLOSE ACTION-231

<trackbot> ACTION-231 Draft replacement for \"how to use conneg\" stuff in HTTP spec closed

ACTION-326: Polyglot documents

<DanC> action-326?

<trackbot> ACTION-326 -- Henry S. Thompson to track HTML WG progress on their bug 8154 on polyglot documents -- due 2010-01-21 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/326

DC: Done to my satisfaction

<DanC> "2010-01-12 15:11:08: The offending para has been removed: http://html5.org/tools/web-apps-tracker?from=4458&to=4459 [Henry S. Thompson] "

DC: offending para has been removed, you (HT) wrote on Jan.12

<DanC> close action-326

<trackbot> ACTION-326 track HTML WG progress on their bug 8154 on polyglot documents closed

LM: Haven't understood about the doctype - whether there were actually valid polyglot docs

HT: this was a very narrow issue

LM: We still have an issue around polyglot documents

<masinter> agree to close action

DC: Interested in XML well-formed

HT: There might be other issues, but not under this action

ISSUE-51 & ACTION-308: Propose updates to Authoritative Metadata and Self-Describing Web to acknowledge the reality of sniffing

JK: Ball is with the group

LM: Additional status - I sent review comments regarding the sniffing draft
... draft is inadequate

<masinter> http://tools.ietf.org/html/draft-abarth-mime-sniff-04

<DanC> tx

<masinter> dated January 26, 2010

<DanC> ("the issue"? which?)

HT: there is another action on sniffing not linked from the sniffing issue
... I sent changes to HTTPBis regarding sniffing

<DanC> action-370?

<trackbot> ACTION-370 -- Henry S. Thompson to hST to send a revised-as-amended version of http://lists.w3.org/Archives/Public/www-tag/2009Dec/0068.html to the HTTP bis list on behalf of the TAG -- due 2009-12-24 -- CLOSED

<trackbot> http://www.w3.org/2001/tag/group/track/actions/370

<ht> http://lists.w3.org/Archives/Public/ietf-http-wg/2009OctDec/0346.html

HT: At TAG request I sent http://lists.w3.org/Archives/Public/ietf-http-wg/2009OctDec/0346.html

<DanC> "Such 'sniffing' SHOULD NOT be done unless there is evidence that the

<DanC> specified media type is in error"

HT: Barth said OK

HT: However, change was rejected by editor

HT: We were asked whether we co-ordinated with HTML WG

<masinter> I am considering offering to rewrite barth-mime-sniff

<DanC> action-370?

<trackbot> ACTION-370 -- Henry S. Thompson to hST to send a revised-as-amended version of http://lists.w3.org/Archives/Public/www-tag/2009Dec/0068.html to the HTTP bis list on behalf of the TAG -- due 2009-12-24 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/370

DC: Can you pursue ACTION-370 HT?

<DanC> action-370 due +2 weeks

<trackbot> ACTION-370 HST to send a revised-as-amended version of http://lists.w3.org/Archives/Public/www-tag/2009Dec/0068.html to the HTTP bis list on behalf of the TAG due date now +2 weeks

<DanC> (arbitrarily; feel free to choose another date)

LM: I don't like the sniffing document

<DanC> (henry, "the rest of us liked it" doesn't speak for me)

<ht> HST would need to look at the f2f minutes

<masinter> On Wed, 2010-01-20 at 15:17 -0800, Larry Masinter wrote:

<masinter> I reviewed draft-abarth-mime-sniff-03 and sent it to the authors and

<masinter> the IETF “apps-discuss”:

<masinter>

<masinter>

<masinter>

<masinter> http://www.ietf.org/mail-archive/web/apps-discuss/current/msg01250.html

<masinter>

<masinter>

<masinter>

<masinter> (ReferenceISSUE-24 and ACTION-308)

<masinter>

<masinter>http://www.w3.org/2001/tag/group/track/issues/24

<masinter>

<masinter> http://www.w3.org/2001/tag/group/track/actions/308

<masinter>

<DanC> ("improvement" can still lead to something I don't like. 1/2 ;-)

<Zakim> masinter, you wanted to comment on use of 'correct type'

<johnk__> http://www.w3.org/2001/tag/2009/09/24-minutes#item03

<johnk__> minutes from f2f where we created ACTION-308

<DanC> tx

LM: Notion that the file has a "correct type" is wrong

LM: You're making guesses about what the author intended

LM: language of "correctness" is wrong

<Zakim> DanC, you wanted to ask lmm about the status quo which relies on something other than what's in the content-type header

LM: if someone tells you it's text/plain and you guess something else, this is your peril

DC: community standard is that web content providers rely somewhat that the consumer will consult more than content-type

LM: Not sure that's true

DC: I know they didn't consider it, but if you took it away they'll be shocked

<jar> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/155 was closed 6 months ago... adam's 1st draft was july 13 ... not clear whether the httpwg talked about it (trying to understand whether there's new information for httpwg)

DC: Is it more cost-effective to specify what's going on, or to move the community away fromthis reliance?

DC: The latter seems expensive, perhaps not possible

LM: value of reverse engineering decays over time

DC: Would be happy to see an alternative draft

LM: Happy to propose alternatives

LM: Recommend that the TAG does not update our findings to reference the current sniffing draft

LM: needs to be opt-in mech as well as uniform and secure

LM: set of criteria need to be met

<DanC> (can anybody write down the criteria lmm said?)

LM: reluctant to recommend sniffing until we have a good algorithm

LM: "fine-grained opt-in"

LM: happy to review barth sniffing draft 4 and suggest any necessary follow-up to TAG

<DanC> ACTION: larry to review draft-barth-sniff-4 and send comments, cc TAG [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action02]

<trackbot> Created ACTION-386 - Review draft-barth-sniff-4 and send comments, cc TAG [on Larry Masinter - due 2010-02-11].

<DanC> ACTION-308?

<trackbot> ACTION-308 -- John Kemp to propose updates to Authoritative Metadata and Self-Describing Web to acknowledge the reality of sniffing -- due 2010-01-14 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/308

<DanC> close ACTION-308

<trackbot> ACTION-308 Propose updates to Authoritative Metadata and Self-Describing Web to acknowledge the reality of sniffing closed

JK: Would like the group to review http://lists.w3.org/Archives/Public/www-tag/2010Jan/0025.html

<johnk__> http://lists.w3.org/Archives/Public/www-tag/2010Jan/0028.html

HT: Will take a look

<DanC> ACTION: Henry to review JK/NM's stuff on sniffing, authoritative metadata, self-describing web, incl. http://lists.w3.org/Archives/Public/www-tag/2010Jan/0025.html [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action03]

<trackbot> Created ACTION-387 - Review JK/NM's stuff on sniffing, authoritative metadata, self-describing web, incl. http://lists.w3.org/Archives/Public/www-tag/2010Jan/0025.html [on Henry S. Thompson - due 2010-02-11].

<DanC> issue-24?

<trackbot> ISSUE-24 -- Can a specification include rules for overriding HTTPcontent type parameters? -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/issues/24

<DanC> ACTION-376?

<trackbot> ACTION-376 -- Daniel Appelquist to send to www-tag a pointer to and brief summary of Mobile Web Best Practices working group's "Guidelines for Web Content Transformation Proxies" and its implications for content sniffing : http://www.w3.org/TR/ct-guidelines/ -- due 2010-02-10 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/376

DC: Related to content-type override issue

DC: DKA - content transformation proxies?

ACTION-278: Draft changes to 2.7 of Metadata in URIs to cover the "Google Calendar" case

DKA: Later...

<DanC> action-278?

<trackbot> ACTION-278 -- Jonathan Rees to draft changes to 2.7 of Metadata in URIs to cover the "Google Calendar" case -- due 2010-02-04 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/278

JAR: Can continue my action 278

JAR: Change the due date

<jar> action-278 due 2010-02-09

<trackbot> ACTION-278 Draft changes to 2.7 of Metadata in URIs to cover the "Google Calendar" case due date now 2010-02-09

LM: Thinking about this a lot

LM: What distinguishes capability-based system and what is proposed here is...

LM: If you have a resource, you should have just one URI

<DanC> (indeed, there's a tension with aliasing)

LM: Capability URI is not usable for an "access control system"

LM: If you combine the key with the URI, you can't do lots of things (expire the key without expiring the URI for example)

LM: Confidential in the finding means something really quite strong

LM: Another use pattern where the information is not confidential, but not widely known

LM: I don't really care if people can read my calendar

LM: Not really confidential

DC: No sharp distinction between that and passwords

DC: Counting on you not to pass it (password) around

LM: I can change the password without changing the calendar URI

LM It's the address as well as the capability

DC: Large random numbers can be revoked

DC: Rethink "don't make aliases"

LM: That _is_ one of the conflicts

LM: The other is that infrastructure of the web assumes it's ok to make easily available URIs (in logs etc.)

TVR Not a useful question to answer

<Zakim> ht, you wanted to gloss larry's point as revocation is willfully breaking a URI

HT: if you put a large random number in a uri, it says that URI identifies a resource

HT: you shouldn't ever revoke that capability

HT:& so you can't easily say that a URI can be revoked

DC: 403/410 them, not 404

HT: It seems you're "cheating" - if you name a resource, and then remove access to the resource at that URI

DC: I'm persuaded that capability URIs are OK...

<DanC> ("actual access control method" is needlessly pejorative... closed-minded, even.)

<jar> lm: Three cases (a) public, (b) obscure, (c) confidential

LM: I see that use of capability URIs are for non-confidential cases

JAR: Not sure what Tyler thinks of Larry's distinction

JAR: Would like to write up the "unsubscribe" case

<raman> have a hard stop, need to bale.

DC: we did write that up

DC: GET/POST finding

ACTION-354: Client side storage APIs

<DanC> action-354?

<trackbot> ACTION-354 -- Ashok Malhotra to review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about -- due 2010-01-21 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/354

AM:AM: Two client-side storage API specs.

<DanC> TAG Action-354 Review client-side storage API’s

AM: should have a better story than "cookies"

AM: name/value pairs should be made more useful

AM: I only looked at two possible cookie storage replacements

DC: Just talking about different use-cases?

AM: there's also a caching API spec and a web storage spec.

AM: I've not yet looked at these

DC: Would like to know about all these APIs and how they compare

AM: Many documents seem to explore this client-side storage case

AM: reviewed 'index' API, 'web SQL' API

<jar> indexed api and web sql api

<johnk__> WebSQLDatabase and Indexed Database API

AM: WebSQL API is not really a spec...

AM: Based on SQLLite database

<johnk__> http://dev.w3.org/html5/webdatabase/

AM: Can look at the other ones, but with what goal?

DC: Is there room in webarch for all of these?

<DanC> Action-354: ashok to look at caching api, etc. as well

<trackbot> ACTION-354 Review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about notes added

<DanC> action-354: and web storage

<trackbot> ACTION-354 Review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about notes added

<DanC> action-354 due +2 weeks

<trackbot> ACTION-354 Review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about due date now +2 weeks

ISSUE-41 & ACTION-369: Shorter document on version indicators

<DanC> action-369?

<trackbot> ACTION-369 -- Larry Masinter to write a shorter document on version indicators -- due 2010-02-04 -- PENDINGREVIEW

<trackbot> http://www.w3.org/2001/tag/group/track/actions/369

DC: Larry wrote a short document on version identifiers, and I reviewed it

LM: Have not incorporated your comments

LM: Suggest we postpone

LM: This is related to polyglot docs

<DanC> subject of my review msg was something like "can't get behind DOCTYPE-based proposal"

LM: would like conforming xhtml to be conforming html when a doctype is present

LM: (scribe: regarding quirks mode, I missed this mostly)

<masinter> trying to speak to the polyglot issue

HT: all kinds of things wrong with the section about doctypes

DC: W3C validator will take a document without a system identifier...

<masinter> I'm asking for help with http://www.w3.org/html/wg/tracker/actions/171

<DanC> ACTION-364?

<trackbot> ACTION-364 -- Dan Connolly to ask HTML WG team contacts to make a change proposal re issue-53 mediatypereg informed by HT's analysis and today's discussion -- due 2010-02-09 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/364

<DanC> ACTION-334?

<trackbot> ACTION-334 -- Henry S. Thompson to start an email thread regarding the treatment of pre-HTML5 versions in the media type registration text of HTML5 -- due 2009-12-02 -- CLOSED

<trackbot> http://www.w3.org/2001/tag/group/track/actions/334

HT: difference between browser behaviour and "meaning" of HTML documents

<DanC> action-364 due +1 week

<trackbot> ACTION-364 Ask HTML WG team contacts to make a change proposal re issue-53 mediatypereg informed by HT's analysis and today's discussion due date now +1 week

<DC: can you review 0015 JAR?

<DanC> ACTION: JAR to take a look at LMM's doctype/versioning proposal http://lists.w3.org/Archives/Public/public-html/2010Jan/0015.html [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action04]

<trackbot> Created ACTION-388 - Take a look at LMM's doctype/versioning proposal http://lists.w3.org/Archives/Public/public-html/2010Jan/0015.html [on Jonathan Rees - due 2010-02-11].

misc action review

<DanC> action-354?

<trackbot> ACTION-354 -- Ashok Malhotra to review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about -- due 2010-02-18 -- OPEN

<trackbot> http://www.w3.org/2001/tag/group/track/actions/354

AM: There's a note from mnot asking "why are they doing this?"

DC: Do we want to talk about resource/representation?

<Ashok> ACTION-354: Discuss MNot note when we next discuss this action

<trackbot> ACTION-354 Review client side storage apis (web simple storage etc.), looking for architectural issues or other critical problems... or interesting design features the TAG should know about notes added

LM: Yes

JAR: can talk more in email...

DC: shall we adjourn?

<DanC> ACTION: Larry to take Dan's proposal on resource/representation and turn it into a change proposal [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action05]

<trackbot> Created ACTION-389 - Take Dan's proposal on resource/representation and turn it into a change proposal [on Larry Masinter - due 2010-02-11].

<DanC> close ACTION-378 (edit)

<DanC> close ACTION-378

<trackbot> ACTION-378 Draft suggested text re resource/representation in HTML 5 for discussion with LMM and JAR closed

resource/representation

LM: One more thing....

HTML Microdata publication news

LM: HTML WG is considering publishing microdata and RDFa FPWDs

LM: Vocabularies have "popped back in"

<johnk__> ADJOURN

<jar> Looking for an action on LMM to draft an html5 change request with DanC's work as input...

<jar> oh i see it now.

Summary of Action Items

[NEW] ACTION: DanC to take approval of minutes 28 Jan offline [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action01]
[NEW] ACTION: Henry to review JK/NM's stuff on sniffing, authoritative metadata, self-describing web, incl. http://lists.w3.org/Archives/Public/www-tag/2010Jan/0025.html [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action03]
[NEW] ACTION: JAR to take a look at LMM's doctype/versioning proposal http://lists.w3.org/Archives/Public/public-html/2010Jan/0015.html [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action04]
[NEW] ACTION: larry to review draft-barth-sniff-4 and send comments, cc TAG [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action02]
[NEW] ACTION: Larry to take Dan's proposal on resource/representation and turn it into a change proposal [recorded in http://www.w3.org/2010/02/04-tagmem-minutes.html#action05]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/02/16 19:26:28 $