<marcos> Hi, I was wondering if someone could give me some advice. In the widget spec, we want to define and "auto update" mechanism for updating widget packages over the air....
<marcos> the proposal looks like this <widget version="1.0 beta" id="htt://widgets.org/mywidget/"> <update uri="http://widgetupdates.com/mywidget/update.php?version=%version%"> </widget> Where, at runtime, %version% is replaced by the UA for the value of the version attribute.
<marcos> The question I have is, would it be better to to define%version% as an entity reference?
<marcos> i.e., &version; ?
<marcos> does it matter?
<Philip> Where the entity reference is not defined in the widget XML file itself, and comes from the environment outside the parser instead?
<marcos> yeah
<Philip> If so, that sounds like it'd be a bit of a pain for anyone using normal XML tools to process or generate those files
<marcos> yeah, true
<marcos> that would suck
<marcos> I'll stick with %version%
<Philip> Rather than %-substitution, you could use (a subset of) URI templates instead
<Philip> i.e. basically just use {version} instead
<marcos> ok, that could work too
<marcos> is there anything written about that?
<Philip> http://tools.ietf.org/html/draft-gregorio-uritemplate-02
<Philip> Oh
<Philip> http://tools.ietf.org/html/draft-gregorio-uritemplate-03
<marcos> great, I'll check those out.
<Philip> That might be a silly idea, but at least the syntax seems less likely to conflict with the usual existence of % in URIs
<hsivonen> marcos: entity references are trouble
<MikeSmith> fwiw, I agree with hsivonen .. named entity references are one of the ugliest legacies inherited from SGML.. wish we could kill them off entirely
<marcos> hsivonen, yeah. I see that now...
<marcos> MikeSmith, understood.
<Philip> (but sandboxing saved the day)
<Philip> (Uh, I think they might have meant to write file:// in that example)
<jmb> Philip: have you a link to that?
<Philip> jmb: http://crypto.stanford.edu/websec/chromium/
<jmb> ta
<Philip> (It seems they're ignoring the issue of the web security model (i.e. enforcing same-originality etc), and only trying to prevent web pages interfering with the user's own computer)
<hsivonen> what's the observing etiquette and schedule? is it poor form to request observer status on another WG on the same days the HTML WG meets (since that would mean either skipping some HTML stuff or not actually observing most of the time)?
<DanC> you're welcome to request observer status all over the place, hsivonen . TPAC scheduling is overconstrained, and creative compromises are expected
<DanC> it's quite common to request to observe several meetings at once
<hsivonen> DanC: OK. thanks.
<smedero> DanC: So I requested observer status through the survey form - but should I be following up with the chairs of the respective working groups directly?
<DanC> the norm is that the ball is now in the chairs' court, smedero . You should follow up only if you think a special case is worth making
<DanC> most chairs just send a bulk "sure, fine, come and observe" message to all requesters, I think
<DanC> I suppose some don't get around to it at all
<smedero> that's what I assumed... I just wanted to make sure I didn't drop the ball on that.
<ChrisWilson> grr
<scribe> scribe: Gregory_Rosmaita
<scribe> scribeNick: oedipus
<ChrisWilson> in light of DSinger's only being here for the first 10 minutes or so, I'll suggest we start with a/v accessibility discussion
<ChrisWilson> anything to add to the agenda?
<Laura> tp://lists.w3.org/Archives/Public/public-html/2008Aug/thread.html#msg697
<Laura> http://lists.w3.org/Archives/Public/public-html/2008Sep/thread.html#msg15
<Laura> I started a page in the Wiki for Multimedia Accessibilty <Audio> <Video>
<Laura> http://esw.w3.org/topic/HTML/MultimediaAccessibilty
<DanC> copy of use cases: http://lists.w3.org/Archives/Public/www-archive/2008Sep/0020.html
<Laura> Considerations:
<Laura> Multimedia presentations (rich media) usually involve both motion and the spoken word. This can present accessibility barriers to those who suffer either visual or audio impairments.
<Laura> The visual components of a multimedia presentation can't be directly accessed by visual impaired users. Likewise, users who are deaf or hard of hearing will not be able to directly access auditory information.
<Laura> Some users may simply not have the equipment, software or connection speed necessary to access multimedia files.
<Laura> Multi-media = multi-modal/multi-sensory so there are many permutations where one or more "modes" may not be available, so we need to try and address each mode as a separate entity as well as the default "combined" or multi-media asset. Separation of these is key.
<Laura> Recommendation:
<Laura> HTML 5 needs to supply support pieces that can be referenced from within <video> and <audio>. Ideally all the support pieces should be direct descendants (children) of the parent <video> or <audio>, but unique and separate. In a perfect world all would be supplied, but even when less than perfect a method to provide one or more support pieces should exist.
<Laura> Example: http://admissions.stanford.edu and look at the source code: there are params for the video, the caption, JW FLV also supports descriptive audio, and even little extras like the watermark. HTML 5 should have a similar approach in native support of any media asset
<Laura> Captions, transcripts, text descriptions, and audio descriptions are different from one another.
<Laura> A clean, semantic, explicit association to transcripts, text descriptions, captions, audio descriptions and/or streams that could be toggled on or off by the end user would be very beneficial. These items will get lost if they are outside the parent element. We need a way for authors to link items together, within the element.
<Laura> The embedded model ensures that the linkage is there, and if the author chooses to also provide an in-the-clear linkage to one or more of these support pieces than this is a win.
<Laura> Some kind of mechanism is needed...attributes on <video> could do it. Likewise for attributes on <audio>.
DanC: had trouble following thread - longdesc diversion; use cases help me a lot
JG: tried to get conversation on
higher level - not how to do, but ability to express
captions
... will look at use cases, as well
... Lachlan's point about user being part of rendering surface
-- sometimes yes, sometimes no~~~~~~; ~~~~ssed as
... media queries: adaptability of content; competing
approaches expressed on list
CW: need to do further reading on
topic before discussing on call (audio & video
thread)
... tons of posts on topic
CS: trying to discharge action items for PF
<ChrisWilson> any further discussion on this, or shall we move on?
HS: prefer discussion more focused - went off topic with longdesc and such
GJR: perhaps some wiki work is needed
HS: media query proposal good idea, but syntax needs work -- idea of querying user's indicated capabilities
<Laura> notes to Dan that I am having trouble hearing. Some static on my end of the call.
CW: interesting idea - privacy implications?
<Lachy> I'm here now, IRC only
HS: privacy problem in any case if person doesn't want to reveal disability to access video stream with certain properties
<Zakim> oedipus, you wanted to ask role of CC/PP
GJR: i am working on a CC/PP
profile for assistive technologies
... one possible approach to ensuring that content negotiation
done correctly
... will set up ESW wiki page for issue
<Lachy> ChrisWilson, no, I think i've said all I need to say on the mailing list
<hsivonen> clarification on the privacy point: a user who selects a media stream meant for people with a given disability end up revealing that (s)he may have that disability
CS: privacy issues about AT concerns?
<ChrisWilson> @Lachy ok thanks.
<hsivonen> but with MQ being able to query stuff all the time, the disability may be revealed without the user noticing that a selection mechanism ran
GJR: whatever is going to raise privacy concerns -- interested in CC/PP as well as XQuery
CS: use of CSS
<Zakim> hsivonen, you wanted to ask about CC/PP & XQuery
GJR: gathering participants to revive CSS-Reader
<DanC> (hmm... "need" is a strong word)
HS: CSS media selectors sufficient?
<DanC> "exposing an accessible _what_"?
<hsivonen> DanC, tree
<DanC> thanks
<DanC> HS: isn't the trend away from the AT accessing the DOM directly to the UA exposing an accessible tree?
<DanC> GR: ATs need to support XML vocabularies/applications
<DanC> HS: seems sufficient to convert to HTML server-side
GJR: HS' idea is a non-starter -- if people need to use an xml-derived language, they should be able to do so without having forced on them as text/html which is less robust
HS: underlying assumption of HTML5 work is HTML5 is common ground - arbitrary XML vocab mapped to semantics AT understands introducing new layer of interaction - new middleware between AT and markup - why not make that ground HTML
GJR: HTML5 does not exist as a specification - one class of readers to disclaimers might change are AT vendors -- they are going to wait until there is sufficient market penetration
<DanC> (the "just use HTML as the AT solution for specialized vocabularies" sounds familiar... that was the outcome in the HTML 2 timeframe for SGML accessiblity.)
GJR: one reason why at vendors putting eggs in ARIA basket
<hsivonen> ARIA is in flux as well
CW: adding features to HTML5 via ARIA aren't distant future-sollutions, but will be added over time
GJR: incremental adaptation without standards and coordination will not happen from AT side
CW: why for features added to HTML5 but not ARIA features added to current HTML implementations
<hsivonen> XQuery for AT capabilities doesn't exist as a REC today, either, right?
CW: agree that need standards and coordination here - have to have that on AT side, as well -- not sure why precludes one sollution or another
GJR: i don't think it precludes one solution or another, but i don't think we should proscribe one solution over another
DS: point of ARIA not to supersede native semantics of ML language; ATs will adapt to HTML5 and ARIA, but ARIA goes beyond the scope of HTML WG
CS: browser devs will do same things doing now - mapping HTML semantics to appropriate accessibility API so ARIA can be used anywhere
DS: not certain why conflict at all between HTML5 and ARIA?
CS: synchronization
DS: widgets will only be deployed on IE8x - not on IE5 or IE6 which are not going away soon -- sites will continue to use javascript libraries to backport content; rather than get IE6 out of the hands of users, we don't care what version of browser or what browser you are using, but here is our syntax; with script libraries, AT will understand things at level that HTML5 browser does
CS: if supports ARIA
... some vendors will do for older browsers
DS: only if AT supports it -- values and attributes into the DOM what UA should do
CS: also important for UA to map to OS
DS: if UA can do that, we are ok
GJR: point simply was that ARIA
for more than HTML5
... therefore needs to be janus-like
<Lachy> xquery is overkill, something like media queries would be syntactically better, but media queries itself is not suitable
HS: in end, need to have
something AT understands -- AT understands stuff in DOM or uses
API that AT understands
... arbitrary XML vocabularies are not accessible - mapping to
ARIA or HTML5 if UA knows about ARIA or HTML5
<DanC> (hmm... since the HTML 2 timeframe, I wonder what ebooks do... do they use HTML vocabulary for accessibility, I wonder?)
GJR: the solution is going to be middle-ware --
CW: mapping components from known to more simplistic
DS: HTML is going to be more
sparse language than GUI frameworks natively supported on
desktops;
... ARIA goes beyond what HTML5 does or plans to do -- why map
to something that is known to be less rich when have option to
map to something more rich just for the sake of a vocabulary
(in this case HTML5)
HS: HTML5 and ARIA being implemented so UA understands DOM and communicates to AT; just pointing out that arbitrary XML vocabularies
<DanC> (the first ebook standard I find uses XHTML + namespaces. <html xmlns="http://www.w3.org/1999/xhtml" xmlns:ops=" http://www.idpf.org/2007/ops"> http://www.openebook.org/2007/ops/OPS_2.0_final_spec.html )
GJR: available now is ARIA; what is being developed are expert handlers that build upon the ARIA model to provide meaningful read/write access to XML dialects in specialized knowledge domain
DS: CWilson right -- drifting away; don't need to introduce arbitrary XML into conversation; use of script, plus CSS plus markup to make
widgets
GJR: ARIA not for HTML5 per se, broader
DS: violent agreement
<ChrisWilson> overdue action items
<ChrisWilson> action-34?
<trackbot> ACTION-34 -- Lachlan Hunt to prepare "Web Developer's Guide to HTML5" for publication in some way, as discussed on 2007-11-28 phone conference -- due 2008-09-04 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/34
CW: action 34 - pretty old action on lachy
LH: have to put off again; been extremely busy
DC: approximate date?
LH: no set date just yet
CW: postpone discussion for TPAC?
LH: would be better
<scribe> ACTION: ChrisW - put "web developer's guide to html5" publication on agenda for TPAC face2face meetings [recorded in http://www.w3.org/2008/09/11-html-wg-minutes.html#action01]
<trackbot> Sorry, couldn't find user - ChrisW
<ChrisWilson> reset due date, will discuss moving forward at TPAC
<DanC> don't make a new action, pls
CW: 21 october due date - face2face is 23rd and 24th
<ChrisWilson> action-66?
<trackbot> ACTION-66 -- Joshue O Connor to joshue to collate information on what spec status is with respect to table@summary, research background on rationale for retaining table@summary as a valid attribute -- due 2008-08-29 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/66
<ChrisWilson> action-74?
<trackbot> ACTION-74 -- Michael(tm) Smith to raise on the list for discussion the issue of XSLT output=html (non)compliance in HTML5 -- due 2008-08-28 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/74
<DanC> yes
<DanC> (RRSAgent's action list is less relevant than trackers)
CW: status list from last week?
DC: raised for discussion
<smedero> issue-54 has been discussed over the last week
CW: can close it
<DanC> close action-74
<trackbot> ACTION-74 Raise on the list for discussion the issue of XSLT output=html (non)compliance in HTML5 closed
<Julian> pointer?
<ChrisWilson> action-75?
<trackbot> ACTION-75 -- Michael(tm) Smith to raise question to group about Yes, leave @profile out, No, re-add it -- and cite Hixie's summary of the discussion -- due 2008-08-28 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/75
<ChrisWilson> julian, were you asking for ptr to email to group on XSLT issue?
<Julian> yes
CW: action 75 - not done by MikeSmith - want to do a poll on that -- leave open for now
<Laura> What is the next step for @headers Action 72?
<ChrisWilson> Will look in a minute.
<DanC> re action-74, there are lots of msgs, e.g. http://lists.w3.org/Archives/Public/public-html/2008Sep/0137.html
<smedero> julian, I groked that from the tracker agenda but you can view the issue itself to see there has been discussion: http://www.w3.org/html/wg/tracker/issues/54
<Julian> likely http://lists.w3.org/Archives/Public/public-html/2008Sep/0137.html
CW: action 74 - XSLT output=html
<ChrisWilson> action-29?
<trackbot> ACTION-29 -- Dan Connolly to follow up on the idea of a free-software-compatible license for a note on HTML authoring -- due 2008-09-11 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/29
CW: action 75 still open; action
29 status
... Dan?
DanC: some promising internal discussion, but don't know specifics yet - hoping to report today
GJR: discussed at HTC?
DanC: don't expect so
<DanC> (life is getting a little too meta when there's a question about what's the next step on an action; an action is supposed to _be_ a next step)
CW: move to?
DanC: give same due date as lachy
CW: done
... covers all of the overdue issues, i believe
<DanC> (ah... re my licensing action, there's an internal thingy due 2008-10-08 ; so I'll set my due date near that...)
CW: Laura asked about action
72
... Lachy, set deadline for TPAC for web dev doc
<Laura> Approx a thousand headers messages to date on list on action 72:
<Laura> http://w3.markmail.org/search/?q=headers+list%3Aorg.w3.public-html+date%3A20070501-20080912
<Laura> @headers has been an issue since May 1, 2007 with about 39 threads.
<Laura> http://esw.w3.org/topic/HTML/IssueTableHeaders#head-bac4baeb0cd0ea09b7f76ff9c409740257566408
<Laura> Latest @headers discussion thread, September 2008:
<Laura> http://lists.w3.org/Archives/Public/public-html/2008Sep/thread.html#msg114
CW: next steps on action 72 - action on CW about headers and header functionality; still need to shift through all info; either declare consensus or post a poll; collect issues into 1 email to send to list -- hope to get to it very soon
<ChrisWilson> @Laura - yes, I need to collate and distill, then ask a question to the group. Still assigned to me.
<Laura> Thank you.
<smedero> TPAC agenda: http://lists.w3.org/Archives/Public/public-html/2008Sep/0303.html
CW: MikeS sent out email on TPAC
agenda - only 1 response received so far
... my take on TPAC similar to MikeSmith's -- first steps are
looking through spec and ascertaining where spec is stable
enough to generate test cases and creating test suite
... figuring out where we are in terms of stability more
interesting in short term
... any suggestions for additions to TPAC meetings, plese send
to list
LH: too much time on tutorial slotted (3 hours)
CW: going to end up spending a lot more of our time figuring out what portions of tech are ready for test cases; easier for us to expand into test case writing area than contract schedule - take what time is needed
LH: ok
HS: face2face time better for discussion than writing tests
CW: absolutely
... comment to MikeS was time spent at last TPAC doing tutorial
on how to build test cases useful - can see us spending that
time, but this time with follow up -- identifying where we can
write test cases - what is solid and what is not
... don't want to spend a lot of time with everyone coding -
not point of TPAC meeting
<Lachy> DanC, I don't have any ideas yet, but I'll think about it and mail the list
CW: value in using test suite for getting handle on how stable spec is, which is what i really want to get out of f2f
<DanC> (I wish I'd made more progress on tests to date... sometimes I wonder where all the time went.)
<ChrisWilson> any other items for discussion today?
CW: additional items for discussion?
DanC: next week (18 september)
<ChrisWilson> move to adjourn?
<Julian> publication of spec?
CW: at Web 2.0 conference in New
York - scheduled to be doing something at this time slot - have
to check MikeS' availability
... have to move tuesday call too
<DanC> I'm out-of-office 25 Sep and 2 Oct
DanC: not available 25 september and 2 october
<ChrisWilson> adjourned
<ChrisWilson> Thanks again to oedipus for scribing
This is scribe.perl Revision: 1.133 of Date: 2008/01/18 18:48:51 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/interested/interested in/ Succeeded: s/lest/less/ Found Scribe: Gregory_Rosmaita Found ScribeNick: oedipus Default Present: +1.425.646.aaaa, smedero, +1.408.398.aabb, ChrisWilson, DanC, Julian, +1.218.349.aadd, Gregory_Rosmaita, hsivonen, Cynthia_Shelly Present: ChrisWilson Cynthia_Shelly DanC Gregory_Rosmaita Julian Lachlan_Hunt Laura_Carlson hsivonen smedero Doug_Schepers Regrets: Mike_Smith SteveF Joshue WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 Guessing minutes URL: http://www.w3.org/2008/09/11-html-wg-minutes.html People with action items: - chrisw developer guide put s web WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]