Weekly XHTML2 WG Teleconference -- 20 Aug 2008

alessio, have i pointed you to http://esw.w3.org/topic/PF/XTech/HTML5/RoleAttribute - need to synch with your work on @role for object

<scribe> scribe: Gregory_Rosmaita

<scribe> scribeNick: oedipus

Outstanding Reviews

SP: MarkB just got back from holiday, so will ping him about XSD review later this week
... important issue today is RDFa
... spec has been in CR - lots of implementations
... 4 implementations with everything
... only need 2 for every feature, we have 4; well done, shane
... have to decide to send RDFa on to PR status
... handled all CR comments; have discussion with Shane today on a single issue, hopefully after that discussion can move to PR

RDFa Transitioning to PR

SM: RDFa defines a module - that module defines collection of attributes
... RDFa also provides an instance of this against XHTML M12n
... problem: way we add attributes is add them to existing attribute collection, the common attribute collection
... handful of elements that don't use the common attribute collection
... is it ok to add attributes by hand in places where common isn't used
... should we be adding common to elements, or just metadata attributes

SP: had problem in past with @id - did we fix by putting "core" on elements

SM: no - put @id on all elements

SP: XHTML1.1+RDFa in short term, put on elements; when regenerate XHTML 1.1, we fix the issue

SM: not easy to fix - module defined wrong and needs to be fixed
... uses modules to fix so have to add to driver
... driver override of what is in module - don't use attribute list from module

SP: for long term, add it all

SM: what does it mean to add "common" to script - what would it mean on script module/element

SP: good question -- don't think does any harm; same argument with XHTML Basic - didn't want attributes that didn't match things that didn't happen - didn't want markup to catch events that won't get fired;
... if events don't happen, they don't happen - makes no difference if attributes there; because one CAN put attributes on SCRIPT, doesn't mean it is non-sense

SM: putting onBlur on SCRIPT doesn't mean anything

SP: right, so don't do it; in principle is allowed, but in princiiple could make script viewable and clickable; non-viewable in principle is has display none

SM: can't currently override with style

SP: in stylesheet could script { display:block; }

SM: in regards RDFa SP thinks fine to put meta attributes on everything: not on SYTLE, SCRIPT or PARAM
... think might need HREF on all of them

SP: resource, but not href everywhere

SM: changed and then ben had problems with it because resolved to put it everywhere in july 2007

SP: ben most against it

SM: creative commons said didn't care
... href everywhere in CR draft
... will put meta information attributes on SCRIPT, STYLE and PARAM and will ship
... any other comments?

GJR: yeah, we're done!

<alessio> <script> is correctly styled on FF, but not in IE6... I'll se on IE7

SM: ok to resolve to publish

SP: try and get MarkB to join IRC to vote
... can delay a few minutes until MarkB can join

SM: willing to table and move on until we can get more members

Modularization Transition

SP: ongoing - not yet reply from commentor - if don't come soon, will re-ping
... can prepare documents in any case
... rich going to join
... we would like to vote on RDFa today - just finished CR and want to transition to PR
... if you agree, we can record your position

RS: plus one to transition RDFa to PR

<ShaneM> +1

<Tina> +1

GJR: plus 1

<alessio> +1

<Steven> +1

proposed RESOLUTION: XHTML2 WG agrees transition RDFa from CR to PR

RESOLUTION: XHTML2 WG agrees transition RDFa from CR to PR

<Steven> http://www.w3.org/2006/07/SWD/RDFa/implementation-report/

GJR: implementation report URI, SP?

<ShaneM> ACTION: ShaneM to add RDFa attributes to script, style, and param elements in XHTML+RDFa implementation - leaving out was an oversight [recorded in http://www.w3.org/2008/08/20-xhtml-minutes.html#action01]

XHTML Media Type document

SP: SM requested we move ahead

SM: been sitting on it for while - did a lot of work then drifted away; what is plan on getting from editor's draft to Note

SP: question is how to get out there with idea that has had enough discussion to show been discussed - send to HTC for final discussion?

SM: a Note, so no approval cycle

SP: trying to reflect reality, so it is ok to deliver XHTML using text/html because IT JUST WORKS

SM: and here are some hints for making it work better

SP: right
... if just publish as note, could say "just documenting real world" but if say to HTC "planning to publish, has real world implications, any comments before transition to note?

<Tina> That DOES mean we are back to rubberstamping current practice.

SP: may get opinions either way - HTC good place to air - maximum hit for minimum investment

SM: in terms of process, i will take a pass to ensure that draft is clean; will notify list when i'm done, and SP can then notify HTC

SP: HTC meets next friday

<ShaneM> ACTION: ShaneM to ensure XHTMLMIME is ready to socialize, then inform the group (and Steven in particular) no later than 22 August. [recorded in http://www.w3.org/2008/08/20-xhtml-minutes.html#action02]

CURIE comment

SP: extended last call comments period; semantic web group said will send comments soon; surprise haven't received yet

Access Module

SP: did we look at any of the i18n issues last week?

<Steven> PR8044-48

SP: 47 and 44 substantive ones

<Steven> http://lists.w3.org/Archives/Public/public-xhtml2/2008Aug/0006.html

scribe's note: in 13 august 2008 minutes - SP stated "principle is to have keys all in one set and diff sets in diff modalities"

<Tina> We discussed this last week, yes?

SM: suggestion can be read 2 ways; 1) additional control that would allow author to indicate the annotation or 2) remove text about UA annotating key
... if UA can recognize...

SP: suggest we say we recognize concern and that is why couch discussion in terms of "MAY" and "e.g." and think that current text covers this case

GJR: plus one from me

SP: "likely to be problemmatic" but we say "depends on UA" - we recommend use of access key in label; if UA can, can do in such and such a matter; not in conflict with commentor's comments; covers cases like his
... text fine as is - covers richardI's cases

GJR: agree

SM: agree

<alessio> agree

SP: happy to write reply myself

<alessio> :)

<Steven> ACTION: Steven to reply to Richard Ishida's comment 8044 [recorded in http://www.w3.org/2008/08/20-xhtml-minutes.html#action03]

SM: turn to 47
... discussed at length last week

http://www.w3.org/2008/08/13-xhtml-minutes.html#item05

SP: didn't come to conclusion, but felt "not our problem"

SM: changed wording a little

SP: can reply changed wording to clarify; please take a look and tell us if suffices

SM: used "abstraction" - TH, GJR and i discussed in hour after call on IRC

<ShaneM> The key attribute represents an abstraction. The use of the name "key" for this attribute is historical and does not mean that there is any association with a specific "key" on a keyboard, per se. It is up to the user agent to provide a mechanism for mapping the document character set value(s) of the attribute to the input methods available to the user agent. For instance, on some systems a user may have to press an "alt" or "cmd" key in addition to the access key

GJR: notes that has outstaning request to WAI/PF and UAAG to review new draft and to propose what to do when key bound to element not in label for element

http://lists.w3.org/Archives/Public/w3c-wai-ua/2008JulSep/0088.html

http://lists.w3.org/Archives/Public/w3c-wai-ua/2008JulSep/0089.html

SP: easy to find out what is being used - UA has necessary info

GJR: programmatically not a problem; discoverability is the issue
... discoverability for those with disabilities who don't use assistive tech

SP: different access keys for different UAs - might have 2 sets of accesskeys - 1 for full-blown computer, one for mobile device - keep text and offer different bindings

GJR: will discuss at this week's UAAG telecon (tomorrow, 21 august 2008)

SP: other comments from RI editorial - don't need to discuss/thrash out on call

[fyi] porting Access Module to HTML5 - http://esw.w3.org/topic/PF/XTech/HTML5/AccesskeyRequirements

SM: concern is do we want to say anything about accesskey from HTML4; "meant to be used instead of that - in MLs where both are present/supported, Access Module is preferred"

GJR: position i have advocated in UAAG - access module preferred

SP: a bit like @style versus STYLE

SM: just adding something to event loop

SP: good way to approach - difference btw @style and STYLE - one overrides the other

SM: what if 3 access elements with identical key attributes?

AC: good question

SP: so what happens if multiple bindings of same key to different elements; 1) disallow (don't have to say what happens); 2) not recommended, but here is what happens when you do it

SM: think we have to take the latter tack; is allowed

AC: agree

SP: can live with that

SM: 2 targetids and 1 targetrole in example - give people rich enough environment by providing ability to provide multiple toles or IDREFs

SP: media query impllications?
... attribute not currently on STYLE is in XHTML2 in common collection; don't want to say "not allowed"
... perhaps call them "active bindings" if host language has way to switch off, can have bindings to same key, but not active at same time

<Steven> http://www.w3.org/TR/xhtml2/mod-mediaAttribute.html#s_mediaAttributemodule

SM: don't define active, so trying to figure out definition

SP: same wording used in XHTML2 for @media module

SM: wouldn't mind brining media into this element

SP: already there in XHTML2
... make up-front now

GJR: plus 1 to doing it now

SP: plus 1

<alessio> +1

<ShaneM> ACTION: Shane to add @media to the access module - steal text from XHTML 2. [recorded in http://www.w3.org/2008/08/20-xhtml-minutes.html#action04]

<Tina> +1

SP: now can speak of "active access elements"

Wrap Up

SP: last minute comments?

ADJOURNED

<Steven> Thanks yet again Gregory

no problem

<Steven> and happy birthday again

thanks!

RESOLUTION: add @media to the access module - using text from XHTML 2

- DRAFT -

Weekly XHTML2 WG Teleconference

20 Aug 2008

Attendees

Contents