See also: IRC log
<janina> agenda: this
<Sean> ok. whats the passcode then
<kenny_j> its 2119
<janina> Shawn, Code is 2119#
<Sean> not working here
<janina> scribe: Shawn
<Sean> Issue 94 draft change proposal for html-issue-142 video (poster alt)
<Sean> Scribe: Sean
reviewed actions. no reported progress
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=10843
<janina> - Support user control over the visual presentation of caption text.
<janina> (RESOLVED NEEDSINFO) - John Foliot
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=10941
<janina> - Media elements need control-independent "pause" for presenting lengthy
<janina> descriptions/captions (RESOLVED REMIND) - Masatomo Kobayashi
sean sent email to silvia
discussed techniques, no inerface for author to interface to client side tts
but other than that the information is there for UA
Janina: there is a motion to taking up client side TTS in HTCG
as part of audio work
<janina> http://www.w3.org/2011/01/14-hcg-minutes.html
mkobayash: did you use normal pause?
sean: yes
mkobayash: user cannot press pause button
the technique uses the pause API triggered by caption
group: conclusion is that the informationis there
keeping open until we hear form silvia before closing the bug
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=11207
<janina> - Make track element additions technology neutral (RESOLVED FIXED) - Sean
<janina> Hayes
still section 10.3.2 close
if Ian was to delete this section then we could close the bug
have put a note to that effect
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=11395
<janina> - Use media queries to select appropriate <track> elements (RESOLVED
<janina> WONTFIX) - Frank Olivier
there is no need for the W3C to keep track
I checked with frank earlier today
eric to check with DSinger
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=11593
<janina> - The <track> @kind attribute should include the all of the identified
<janina> accessibility content types (NEW) - John Foliot
<scribe> ACTION: Eric_Carlson, get D Singer to adopt bug 11395 [recorded in http://www.w3.org/2011/01/19-html-a11y-minutes.html#action01]
<trackbot> Sorry, couldn't find user - Eric_Carlson,
<scribe> ACTION: Eric_Carlson get D Singer to adopt bug 11395 [recorded in http://www.w3.org/2011/01/19-html-a11y-minutes.html#action02]
<trackbot> Sorry, couldn't find user - Eric_Carlson
<janina> http://www.w3.org/Bugs/Public/show_bug.cgi?id=11593
<janina> - The <track> @kind attribute should include the all of the identified
<janina> accessibility content types (NEW) - John Foliot
in the run up to LC we were thinking about new track types so John filed a bug on it
however Silvias analysis is that they were all covered
Janina: we can close if that is indeed the case
check with silvia
janina: done with first
pass
... will need to return to some as a second pass
... confirm in email, ask john to manage
janina: is there anyone that
thinks that alt text needs to distinguish poster from
media
... seems that the poster behaviour functions as a static image
and thats what the alt is abouy
eric: john is not here
and his opinion is we need something as a separate discription
judy: whats the point of discussing now?
janina: bringing out the positions
eric: if you are saying that we need a separate description of the 1st frame from the video, i dont agree
janina: we dont need a description of the elemnet
this is what we traditionallly have done is describe images
so if poster is an image then alt and longdesc apply to it
or 1st frame if its functioning as an image
no alt description for the element
eric: I meant a description of the resource
eric; agree author needs alt or longdesc of the media content
eric: but not describe the 1st
frame separately
... if it needs to be destinct then it should be described as
part of the media resource
eric; seems to me that a description of a video describes what you see in the first frame or poster
janina: how does this differ from a static image
eric: video is different tform a static image
janina: sighted user knows its a video how?
eric: if default controls arent shown then it looks the same as an image
if the controls are there then you see them overlaid
janian: browser + AT can recognize the controls
eric; yes
janina: so see no need to describe that
prgrammatically ccan determine image from video, so whatever is being shown is all that needs describing - which is a static image
functionally no different than describing an image
eric: johns point is that we need
a new child element of the video containing the poster and a
description of the poster
... agree with Janina, use existing attribute on the element
itself
janina: true that alt as an attribute has some issues
for i18n
but thats a separate discussion on enhancing alt
eric: what about autoplay?
janina: can keep that from happening
eric: even if it starts playing on any user action. what does it mean if the description only cvers the static part and the AT gets to it after playback has started
janina: we should try to avoid that
eric: other controls on the page e.g. an element, can start the video playing
so thinking that doesnt make sense to just describe the static representation, should include description of the media and the first frame (if significant)
eric: we cant prevent it, but may be an edge case
image on another part of the screen which starts the video on hover
any script can trigger media element
janina: yes - a problem
wcag requirement is that you can disable that
or stop it
eric: hard to do
iOS has this behaviour
needs an explicit user gesture to start video
must be in a user initiated event handler
but hard to do on desktop, big change
would break content
Judy: I'm worried we might not have anything exlict on this in wcag
possible we have something adequately generic
and a technique would take care of it
eric: best we can do is an authoring guideline
janina: thinking we cant prevent it
judy: except if we have the ability to feed in UA guidelines
eric: my point is that its possible for a UA to prevent autoplay, not honor page authors trequest
but its difficult prevent any script
when is it OK, thats the issue
janina: its a common behaviour, and people like it; but for some its an accessibility issue
eric: but whats to prevent a script running>
kenny: difficult for engineers, but could filter out based on events
eric: wont prevent it
they are attached to media loading
very difficult to determine automatically when its ok and when its not
on iOS script has to be triggered from user gesture
but its hard to implement; and breaks a lot of content
lots of bug reports on iphone because of this
causes authors to write new scriptsa for the phone
wont get buy in on such a change
janina: maybe its a new user requirment
judy: wcag doesnt say anything directly relevant to preventing autoplay
the relevant place is UAAG, and has a clause on separate selection form activation
as a user option
partial relevance
and another one about event handlers
maybe needs to revisited if we have feedback on autoplay
kenny: Q: is there a possibility of dealing with it in browser to give screen reader user a means to stop the video on request
judy: both probably, but we dont have both covered
eric; technically possible
but custom controls wuld have to have a case to deal with it
on page with default controls tho could work
janina: strayed form the poster issue? but identified an important issue
need to articulate it better
janina: that media does not start automatically unless user asks for it
not bringing focus on another element etc. setting for specific users
eric: this is a problem that cant be solved cleanly
we will prevent people from writing custom controls
kenny: so if browsers give a keystrok to stop, it should be possible now that video is native
may be a more robust solution
and its only one new gesture to learn
eric; could work.
sean: but if the video starts then the poster is no longer visible
eric: so the description needs ot be of the whole resource
so its always relevant
janina: ok, hadnt thought of that
kenny: alt text for the video must describe the video, but not frame by frame
janina: description for poster would need to dissapear
would make sense if the video is paused on different frame
wouldnt
janina: more fruitful conversation tan expected
summary: 1 mechanism or multiple mechanisms
point of disagreement
point 2: jf and janina have discussed offline, he's cncernd that the poster may be irrlevant to what the movie is
if thats the case then its the case for the sighted user too
expecting there is some thing which tells you what the video is, separate from any poster image
like a title
judy: a name is the most consistent thing
eric: majority of web content just has the first frame of the file
so not that helpful to sighted user either
janina: should try to solve bad authoring
should not
but john may have a different perspective
have we missed anything?
[silence]
CfP for media API went out. deadline feb 21
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/form/from/ Found Scribe: Shawn Found Scribe: Sean Inferring ScribeNick: Sean Scribes: Shawn, Sean Default Present: Janina, Kenny_Johar, Judy, Eric_Carlson, Sean_Hayes Present: Janina Kenny_Johar Judy Eric_Carlson Sean_Hayes Got date from IRC log name: 19 Jan 2011 Guessing minutes URL: http://www.w3.org/2011/01/19-html-a11y-minutes.html People with action items: d eric_carlson get singer WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]