HTML Weekly Teleconference

15 Jan 2009


See also: IRC log


Sam, +049251280aaaa, Julian, dsinger, Matt_May, Joshue, hsivonen, Mike, DanC, Lachy, ChrisWilson, +1.408.536.aacc, +1.519.538.aadd, MurrayM, LarryM, Shepazu, [Microsoft], smedero, Mike.a


Convene, take roll, review agenda

<DanC> (ideally, we would have updated those actions and sent out the contents of http://www.w3.org/html/wg/tracker/agenda at T-24hrs, but hey... we'll get there)

<trackbot> Date: 15 January 2009

<pimpbot> Title: Input for Agenda Planning for the HTML Weekly - HTML Weekly Tracker (at www.w3.org)

<pimpbot> Title: Re: {agenda} HTML WG telcon 2008-11-20 from Sam Ruby on 2009-01-13 (public-html-wg-announce@w3.org from January to March 2009) (at lists.w3.org)

<DanC> and this is sorted: http://www.w3.org/html/wg/tracker/agenda

<pimpbot> Title: Input for Agenda Planning for the HTML Weekly - HTML Weekly Tracker (at www.w3.org)

W3C publications heartbeat

ChrisWilson: let's talk about the heartbeat requirement first
... we need to get publication notes detailing changes from the last draft
... any objections?

<anne> Is it realistic to get a detailed changelog?

LarryM: LarryM: do we need a review first?

<Joshue> I would like to see the poll on @summary go ahead before the next iternation of the spec or the draft is published.

ChirsWilson: no, it this isn't a new document

<DanC> an update of html4-diff is much appreciated; i think the level of detail you typically come up with is fine, anne

LarryM: if you do an action with no significance, why do the action?

DanC: there is a lot of new forms material...

<ChrisWilson> Anne, how detailed are you considering pubnotes to be? I wasn't thinking checkin-by-checkin changelog; but the overview, e.g. webforms, would be important imo

<Joshue> I don't like the fact that @summary has been dropped. This was a unilateral decision and I think the wider group should consulted. Without this process @summary wioll have little to no chance of being reinstated.

LarryM: I don't have any more comments...

<masinter> yes

<DanC> Larry, at one point I tried to be sure every section had been reviewed by 2+ HTML WG reviewers: http://esw.w3.org/topic/HTML/SpecReviews

<ChrisWilson> Josh, is that related to current topic or a new topic?

<pimpbot> Title: HTML/SpecReviews - ESW Wiki (at esw.w3.org)

<Joshue> zaki, unmute me

<Lachy> which attribute is being discussed?

<hsivonen> Lachy, summary

Joshue: I think it important to resolve the summary issue before the next heartbeat document

<Lachy> summary was never in the spec

DanC: I don't believe that summary was dropped since the last draft

<anne> ChrisWilson, http://www.w3.org/TR/html5-diff/#changelog

<pimpbot> Title: HTML 5 differences from HTML 4 (at www.w3.org)

<DanC> issue-32?

<trackbot> ISSUE-32 -- how to provide a summary of a table, e.g. for unsighted navigation? -- OPEN

<trackbot> http://www.w3.org/html/wg/tracker/issues/32

<pimpbot> Title: ISSUE-32 - HTML Weekly Tracker (at www.w3.org)

<anne> ChrisWilson, that's about as much as I can commit to

DanC: does issue 32 need to be resolved before the next publication?

<hsivonen> Joshue, why should issue 32 block the heartbeat when none of the other open issues are blocking?

DanC: I symphatize for the issue, but don't believe that it need to be solved before publication

ChirsW: I would like to move forward towards publishing, Dan, do we need a poll?

<Joshue> Just giving my two cents.

DanC: no

ChrisW: I will send a mail out

<masinter> I would like the opportunity to review the draft in detail, and want to make sure that agreeing to publishing the working draft doesn't preclude raising issues

<ChrisWilson> action ChrisWilson send mail to WG saying we will issue new WD due tomorrow

<trackbot> Created ACTION-95 - Send mail to WG saying we will issue new WD due tomorrow [on Chris Wilson - due 2009-01-22].

<DanC> issue: HTML 5 spec update after 10 June 2008

<trackbot> Created ISSUE-65 - HTML 5 spec update after 10 June 2008 ; please complete additional details at http://www.w3.org/html/wg/tracker/issues/65/edit .

<anne> masinter, it never does

<anne> masinter, publishing is just sending out a note for wider review

<masinter> thanks, yes

ChrisW: publishing the working draft does not preclude raising issues
... I want to do pending review actions first

what list are we all looking at?

ISSUE-20 (table-headers): Improvements to the table-headers algorithm in the HTML 5 spec

<ChrisWilson> action-87?

<trackbot> ACTION-87 -- Michael(tm) Smith to ensure Ian Hickson follows up on semantics-tables messages -- due 2008-12-20 -- PENDINGREVIEW

<trackbot> http://www.w3.org/html/wg/tracker/actions/87

<pimpbot> Title: ACTION-87 - HTML Weekly Tracker (at www.w3.org)

<Joshue> regarding @summary please note the request from the PF to keep the attribute http://lists.w3.org/Archives/Public/public-html/2008Aug/0213.html

<pimpbot> Title: Re: Request for PFWG WAI review of summary for tabular data from Al Gilman on 2008-08-06 (public-html@w3.org from August 2008) (at lists.w3.org)

<DanC> so http://www.w3.org/TR/WCAG20-TECHS/H43.html is still non-conforming?

scribe is confused, issue-20 has an action which is due 2009-01-30

<pimpbot> Title: H43: Using id and headers attributes to associate data cells with header cells in data tables | Techniques for WCAG 2.0 (at www.w3.org)

<gsnedders> DanC: yes

DanC: would the validator flag the example?

Hsivonen: no

DanC: great!

<DanC> "HTML 5 draft allows @headers on td but not on th."

<DanC> is th/@headers allowed now?

Chris: I'll look into this, but I think we can close it

<anne> DanC, yes

<DanC> spiffy.

<anne> DanC, see http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#the-th-element

(many): yes

<pimpbot> Title: 4.9 Tabular data HTML 5 (at www.whatwg.org)

<masinter> did Matt_May liaison get on agenda?

ChrisWilson: status of 87 is now closed, I'd like to leave 72 open to remind me to review it...

ISSUE-63 (origin-req-scope): Origin header: in scope? required for this release?

<ChrisWilson> action-89?

<trackbot> ACTION-89 -- Michael(tm) Smith to make a proposal to the WebApps WG that we take this on as a work item there, with Adam Barth as the editor -- due 2009-01-06 -- PENDINGREVIEW

<trackbot> http://www.w3.org/html/wg/tracker/actions/89

<pimpbot> Title: ACTION-89 - HTML Weekly Tracker (at www.w3.org)

MikeSmith: consensus is that this doesn't belong in webapps

<DanC> it was in the editors' draft as of http://html5.org/tools/web-apps-tracker?from=2524&to=2525

<pimpbot> Title: (X)HTML5 Tracking (at html5.org)

MikeSmith: we can close 63

DanC: It was (previously) in the draft

<masinter> action is to bring proposal to IETF?

<trackbot> Sorry, couldn't find user - is

MikeSmith: it still is, the plan is to move it out

<anne> MikeSmith, really, did we discuss that? The Origin header is currently defined in a WebApps WG draft, after all...

DanC: I wouldn't mid a recorded decision that we aren't doing it

ChrisW: Is there something we need to do?

DanC: the consensus in the IETF liason call is that this belongs in the IETF

ChrisW: we can close this item

<masinter> liaison need to track?

DanC: I'd like a decision!

ChrisW: I'll do that

Larry: do we need to track this?

DanC: I'm content that this is being tracked

ChrisW: I propose closing the issue and action

DanC: I think it has had enough attention

<masinter> closing it sounds good to me

DanC: Sam?

Sam: should there be an action to remove it from the spec?

Doug: would it be prudent to leave it in the spec pending some action?

Larry: I think it would be imprudent to keep it in the spec...

Doug: I'm fine either way..

larry: If we are deferring to the IETF, we are saying we aren't doing it.

<DanC> (rubys, we've experimented with hixie carrying tracker actions, and the current status is that he doesn't; somebody else takes an action to work with hixie...)

MikeSmith: my action is done

ChrisW: Is this actually referenced in the HTML spec?

DanC: it was

Julian: the spec currently has this text

<Julian> it's in http://dev.w3.org/html5/spec/Overview.html#origin

<pimpbot> Title: HTML 5 (at dev.w3.org)

ChrisW: We need an action

Henri: I can take this action

<trackbot> ACTION-96 -- Henri Sivonen to to ensure editor removes Origin header: from spec -- due 2009-01-22 -- OPEN

<trackbot> http://www.w3.org/html/wg/tracker/actions/96

<pimpbot> Title: ACTION-96 - HTML Weekly Tracker (at www.w3.org)

<DanC> feel free to give a different ETA, hsivonen

<DanC> a la: action-96 due 15 Feb 2009

<anne> I'm not sure I agree this is the right course of action. It only affects HTML forms... Didn't we establish this last time this was discussed?

<anne> Also, the WebApps WG are the ones currently defining the Origin header...

<MikeSmith> trackbot, comment action-96 see http://krijnhoetmer.nl/irc-logs/whatwg/20081211#l-45 : Hixie: "splitting off the protocol part of websocket, the content-sniffing section, the uri section, and a brief definition of the Origin header, and submitting them as four tentative IDs"

<trackbot> ACTION-96 to ensure editor removes Origin header: from spec notes added

ISSUE-31 (missing-alt): What to do when a reasonable text equivalent is unknown/unavailable?

Matt: what's the goal here?

DanC: I just want the overall plan for having it fixed in general, no need to worry about trying to get it fixed by the next draft

Matt: I can come back next week with status

<ChrisWilson> whoops, didn't mean to hit enter yet.

<ChrisWilson> just a sec, Henri

<ChrisWilson> thx

Matt: this is a topic I don't expect the HTML spec to solve...

<Joshue> Just for the record http://esw.w3.org/topic/HTML/IssueAltAttribute

<pimpbot> Title: HTML/IssueAltAttribute - ESW Wiki (at esw.w3.org)

<DanC> (taking a peek at http://esw.w3.org/topic/HTML/IssueAltAttribute to see how it does with NPOV...)

<pimpbot> Title: HTML/IssueAltAttribute - ESW Wiki (at esw.w3.org)

<smedero> This was the last editor's summary on @alt: http://lists.w3.org/Archives/Public/public-html/2008Aug/0759.html

<pimpbot> Title: The alt="" attribute from Ian Hickson on 2008-08-26 (public-html@w3.org from August 2008) (at lists.w3.org)

Matt: the current draft doesn't close the door on optional alts, and I'm unsure how to proceed from here.

<DanC> (re HTML/IssueAltAttribute "latest published" is at risk of going stale ...)

hsivonon: what should a piece of software do if a user is uncooperative? what are the sequences of bytes that a validator must flag as nonforming?

<Joshue> @Dan, yeah it probably needs updating

<pimpbot> Joshue: Huh?

hsivonen: my position is that a data format should allow a signal for an authoring tool to indicate that a user did not provide the data, and that should be conforming

<DanC> Josue, so far, I don't see any critical NPOV problems in that wiki summary, though; we seem to be getting better at using the wiki

<DanC> (it would be nifty if the arguments hsivonen and matt are talking about were easy to find in that wiki summary)

<Joshue> Dan, yeah, it is a pretty good overview of these myriad issues and related offshoots.

cynthia: what worked well in the past is the alt attribute is required, but an empty string is ok, why does that need to change?

<dsinger> the discussion was that the empty string then becomes ambiguous: it's also used to say that an image is, for example, purely decorative

<dsinger> the UA cannot then tell the difference betwen "alt desirable but unavalable" and "alt wasn't needed"

cynthia: I agree with Matt that descriptions as to what is a valid alt tag belongs in WAI...

<anne> cyns, why? alt="" is not just an accessibility technique

<anne> cyns, it's important for e.g. text browsers as well

Doug: Henri? would you agree that if all you are validating is HTML, then you shouldn't flag a missing HTML, but a validator that also also was aware of WCAG would validate more?

hsivonen: yes

<cyns> slight modification of summary of my statement: what is *good* or *appropriate* alt in a scenario is not a language issue, but an authoring issue. should not be in the language spec.

hsivonen: it may be a different class of message

<pimpbot> Title: IRC logs: freenode / #whatwg / 20081211 (at krijnhoetmer.nl)

<anne> cyns, the language should tell authors how to write it, no?

<anne> cyns, the language spec...

doug: fine, this may be a communication issue; not requiring it in the language doesn't mean that it can't be flagged

hsivonen: right

<cyns> anne, the language should tell the author what is valid, but not necessarily what content is equivalent.

<takkaria> I'm not sure what quite what use defining a language is if you don't define how to use it in certain situations

matt: DreamWeaver gives you options, but if you don't make a decision, no alt tag is generated, and I consider that a "pass"

<jgraham> FWIW I guess not putting information about alt usage in HTML 5 will jsut mean fewer authors are exposed to that information

matt: not letting you save non-conforming documents is a non-starter in the marketplace

<anne> cyns, why not? the author needs to know how to write the language properly

<masinter> this is a case where defining what the language *means* independently of how authors should author and browsers should interpret is a good idea

<anne> cyns, the language specification should tell the author how to properly include an image, imo

<anne> masinter, I don't really see why that matters

matt: having the validators be the gatekeeper has provided significant value in the past

<masinter> alt="" means something different than 'no alt'.

<dsinger> the assumption that HTML generation is coincident with a person who knows what the non-text content is like, is not tenable

<masinter> different requirements on interpreters and generators

<cyns> anne, examples, yes, but not every possible scenario. fine in an authoring guide, but it and WCAG should be single-sourced in that case

<anne> masinter, sure, that's all in the specification

<anne> cyns, but WCAG is for accessibility, doesn't cover e.g. search engines or text browsers

<takkaria> how can you talk about waht a language means independently of how people should understand it?

<DanC> (pointer to relevant part of ATAG docs, anyone?)

hsivonen: authoring tools should do what DreamWeaver does, and believe that that ATAG 2.0 should direct tools to do what DW does

<dsinger> not all HTML is made by interactive tools...

<shepazu> so, if the government requires that content for government web content follow WCAG guidelines, then there will be market pressure for an authoring tool to force/encourage authors to add alt text, regardless of whether it is is for HTML or WCAG validity

<masinter> there are different constraints on the content and on the tools for generating content

hsivonen: we shouldn't make HTML5 require that some streams that can be generated using ATAG guides be considered non-conforming

<masinter> different constraints on visual display browsers and screen readers

<dsinger> I agree with Julian; weare going over old ground that took hours on the mailing list

<Joshue> +1 to Henri, I agree there should not be rubbish values inserted into content just to satisfy a validator

<DanC> (julian, if you can point to a succinct summary, then that might save us time, but otherwise, yes, some redundant discussion is natural)

masinter: if you try to do everything in one specification, it will be hard to read and contain a lot of information that is irrelevant to many classes of users.

<dsinger> Dan, I fear that no-one is satisfied (not even the editor)

matt: the WAI coordination group is planning on discussing this.
... I can report back next week on this and on summary

<DanC> dsinger, I don't expect the editor to commit text he's not satisfied with... without an explicit note. is there one?

chris: I'll update the status

<trackbot> Created ACTION-98 - Discuss missing-alt with the WAI CG and report back [on Matthew May - due 2009-01-22].

<Joshue> summary attribute http://www.w3.org/html/wg/tracker/issues/32

<pimpbot> Title: ISSUE-32 - HTML Weekly Tracker (at www.w3.org)

<masinter> and action 90 and action 92 should both be closed

<DanC> close action-90

<trackbot> ACTION-90 Ask Matt May if he can help represent WAI WGs in the HTML WG closed

ISSUE-37 (html-svg-mathml): Integration of SVG and MathML into text/html

<ChrisWilson> action-94?

<trackbot> ACTION-94 -- Doug Schepers to report back on SVG WG's integration proposal re: issue-37 -- due 2009-01-22 -- OPEN

<trackbot> http://www.w3.org/html/wg/tracker/actions/94

<pimpbot> Title: ACTION-94 - HTML Weekly Tracker (at www.w3.org)

doug: I've moved the action back to next week

ISSUE-54 (doctype-legacy-compat): tools that can't generate <!DOCTYPE html>

<ChrisWilson> action-91

rubys: proposed on the list, awaiting feedback from the editor

<DanC> well, the "report on feedback " part isn't done

ChrisW: what's the current status

rubys: we have a proposed change, and waiting on the editor to make the change

DanC: sam, you've seen all the feedback you think you need to see?

rubys: yes

ChrisW: what is limited quirks?

<DanC> +1 empty string, i.e. <!DOCTYPE html "">

<Julian> +1 as well

<anne> ChrisWilson, limited quirks is a new name for "almost standards mode" (because it's now part of to be standard)

<DanC> +0 legacy-compat

<Lachy> no, it would by <!DOCTYPE html PUBLIC "">

<ChrisWilson> thanks Lachy, I was going to say that.

<Julian> +1 on empty string, +0 on legacy-compat

<hsivonen> sigh, a moment ago we were so close to declaring consensus on "legacy-compat"

<DanC> sorry, lachy, I'm perhaps going too fast

<ChrisWilson> Henri, no I don't think we were

<Lachy> I object to using the empty string version

<hsivonen> I'm +1 on "legacy-compat" and -1 on ""

Lachlan: does not support "", prefers xstl-compat or legacy-compat

<ChrisWilson> takkaria, I disagree

rubys: I've yet to hear anybody argue against legacy-compat

<shepazu> how about "processor-compat"?

<gsnedders> I'm +1 on "legacy-compat" and -1 on "" too

<masinter> is the objection that the word 'legacy' is pejorative?

<Lachy> the only problem with legacy-compat is that it's not entirely clear that it's meant for compat with legacy generator tools, like XSLT, rather than legacy consumers like browsers

chriswilson: I'm not happy with legacy-compat, it implies that something is wrong.

DanC: "" is better?

<hsivonen> masinter, the string is pejorative on purpose to make people prefer <!DOCTYPE html>

<Julian> +1 on the reasons ChrisW is giving.

<anne> ChrisWilson, but something is wrong...

ChrisW: yes
... I don't understand the goal of being perjorative on purpose

<masinter> if you want to be pejorative, do it on your blog, not in the spec

<masinter> pejorative

<masinter> byeee

ChrisW: requests Sam to reply to the previous thread, and keep the action open.

Sam: OK

ChrisW: I move that we adjourn

numerous seconds

meeting adjourned...

<pimpbot> Title: HTML Weekly Teleconference -- 15 Jan 2009 (at www.w3.org)

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.134 (CVS log)
$Date: 2009/01/16 17:51:14 $