See also: IRC log
<trackbot> Date: 21 January 2010
<kford> rrsagent make minutes
<kford> Scribe: Kim
<scribe> new members, introductions
<kford> Group all introducing themselves.
<kford> Mark do you want to type an intro?
<kford> zakim take up item 1
Kelly: 4.9 talks about giving the user some control over content that may produce accessibility -- deals with multimedia and a multitude of other things. Need someone willing to drive 4.9 and consolidate thinking and look for consistency.
<kford> Mark, have you had an opportunity to review this message? Any interest in owning 4.9 and driving it?
<mhakkinen> it could.
<mhakkinen> i mean, yes, assuming I make headway on my other actions.
Correction: ... over content that may reduce accessibility...
<kford> Several of your items are in this area already. Let's call you the owner and you make this one happen. Feel free to pull others in.
<kford> Agreed, Mark owns 4.9. Make it happen.
<jeanne> Guideline 3.1 Provide access to alternative content.
<jeanne> 3.1.1 Notification of Alternative Content: Provide a global option for the user to be notified of alternatives to rendered content (e.g., short text alternatives, long descriptions, captions).
<jeanne> 3.1.2 Configurable Default Rendering: Provide the user with the global option to set which type of alternative to render by default. If the alternative content has a different height and/or width, then the user agent will reflow the viewport. (Level A)
<jeanne> 3.1.3 Browse and Render: The user can browse the alternatives and render them according to the following (Level A):
<jeanne> * (a) text alternative @@Editors' Note: this criterion is under development@@
<jeanne> * (b) captions
<jeanne> * (c) audio descriptions
<jeanne> * (d) sign language video
<jeanne> * (e) full text alternative
<jeanne> * to replace...
<jeanne> * (a) synchronized alternatives for time-based media (e.g., captions, audio descriptions, sign language) can be rendered at the same time as their associated audio tracks and visual tracks, and @@Implied in 2.3 in UAAG10@@
<jeanne> * (b) non-synchronized alternatives (e.g., short text alternatives, long descriptions) can be rendered as replacements for the original rendered content. If the new item has different dimensions, then a user option controls whether the dimensions of the original content are used or the dimensions of the new content, which will cause the document to reflow accordingly.
<jeanne> 3.1.4 Available Programmatically: If an alternative is plain text (e.g., short text alternative), then it is available programmatically, even when not rendered. (Level A)
<jeanne> 3.1.5 Rendering Alternative (Enhanced): Provide the user with the global option to configure a cascade of types of alternatives to render by default, in case a preferred type is unavailable. If the alternative content has a different height and/or width, then the user agent will reflow the viewport. (Level AA)
Kelly: 3.1 a few different issues -- most notably w 3.1-3.3 -- numbering issue? Greg e-mail.
Greg: briefly, definition of alternative content, so raw HTML view isn't considered as alternative content for example
Great: I'm not sure 3.1.1 not sure achieves goal, 3.1.2 text error, 3.1.4 availability of alternative content -- facilitate programmatic access,we don't define plaintext
Greg: there are some cases where you don't want to automatically expand the containers instead do scrollingor provide some or else in interface like spreadsheet
Patrick: reflow of everything
else makes a lot of sense -- it's more about making sure it's
available, making it discoverable rather than notify actively
-- they can still be something that the user has to take an
action, just a matter of wording. If alternative text is too
long -- I would agree with that if there was a better way of
presenting that when the user requests that's a valid...
... approach. So definitely being careful with the wording. Needs to be provision when it's not just plain text.
Jeanne: I like discoverable
Greg: need to be a little bit more proactive
Jeanne: it does solve the problem of single-line text field -- to notify people. Use case single line text input field where the content is overflowing and currently no browser gives you a notification that there's more text to be found but having it be discoverable as opposed to a notification would cover that.
Greg: it's important for user to
know when there is scrolling that there is something to scroll
... today all browsers would fail -- no scrolling on single-line text boxes. Key question is what is the minimum requirement we want to have for the notifying or making discoverable the availability of altered content.
Patrick: discoverability -- visual feedback one possibility but make sure we don't hardcod it
Greg: visually indicate all elements that have alternative content available - I think we're pretty much talking just about visual here. Visually indicated, and all that are available at the same time
Bim: idea of having an indication
of alternate content -- would be the user agents responsibility
to make the access to that keyboard navigable
... concerned me long time that -- recommending people use captions instead of alt text
Patrick: captions rather than alt text -- hopefully that will be so presumably with better aria support -- also with, tied to better descriptions
Kelly: as much as possible we've
tried to be specification neutral in success criteria -- today
the agent may use aria to satisfy that criteria but five years
from now there may be some other technology. But we've tried to
do, companion document, techniques document for each success
criteria the intent of the success criteria -- if you're user
agent what do we really want you to do here,...
... some examples, and then some resources. I think example you just outlined with aria is a good example of how you achieve the success criteria.
Greg: referring to IVR, call
trees things like that -- not sure whether the conflict of
alternative conflict applies enough today, but I imagine that
could. Choosing between having a tone indicating call versus
prerecorded voice indicating call. As in theory we are supposed
to be technology neutral -- hoping on getting peoples input
about whether we should address that here -- totally
... output Web browser
Kelly: accessibility nothing jumps immediately to mind, but certainly languages, phone based browser and its reading back everything and it has a verbal interface. There it might be applicable, but I don't have a good real-world example yet
Great: something that's not using HTML, we're all familiar with how alternative content works there. Using Skype or something like that -- and Internet base for web-based system for conference calling, for example. Pre-recorded message like to have been disconnected. Do we need to address, alternative content might be needed.
Kelly: a couple of things fall out of this
1. maybe this is an area where we want feedback on next draft
2. proposals for cleaning particular area out
Patrick: I can have a look at 3.1 in its entirety, bring it to the next meeting
<jeanne> ACTION: PatrickL to review 3.1 with proposal for a clean-up. [recorded in http://www.w3.org/2010/01/21-ua-minutes.html#action01]
<trackbot> Sorry, couldn't find user - PatrickL
<jeanne> ACTION: PL to review 3.1 with a proposal for clean-up due in 2 weeks [recorded in http://www.w3.org/2010/01/21-ua-minutes.html#action02]
<trackbot> Created ACTION-263 - Review 3.1 with a proposal for clean-up due in 2 weeks [on Patrick Lauke - due 2010-01-28].
<Greg> Re 3.1, Visual-output UA should provide the user option to have it “visually highlight all” content that has alternative content. However, for audio-output UA it would instead presumably audibly highlight content that has alternative content *when it’s being played/read*. That is, it’s indicated one-at-a-time while going through the document (or part of it) sequentially. The visual...
<Greg> ...equivalent woul
<Greg> d be to put notification on the status bar when the keyboard focus is moved onto an element that has alternative content, which would NOT satisfy the intent of this SC.
<kford> Simon's proposal on 22.
Greg: pasted in link that clarifies audio browsing and difference
Reviewing Simons proposal
<kford> Group reviewing Simon's proposal.
Kelly: I'd vote accept
<kford> KFord: From my perspective I'd vote to accept.
Patrick: only thing the jumps out
at me is it about if it's hidden content -- accordion style
expand contract with different sections and we say explodes
that even if it's hidden, and taking in practical terms how the
user agent would be able to understand, or if there some other
mechanism to you a would have to do to force the paring
container to be visible. Curious about how the UA
... do it in the first place.
Simon: voted at the time anyway so CSS display attributes can easily be switched on and off, and can be accessed just that the UA decides it's not going to be displayed when the rendering engine starts. Timestamp -- you can also search within that file, just not actually displayed at the time. Useful for locating areas of animations or videos you want to zip through or search within that.
Kelly: say I had a menu, said pick your color. When the user clicks, says red, blue, green. When I do a find for blue what would you expect to happen?
Simon: simple find, show blue, advanced find, find every instance
Kelly: what if UA could expand
because requires user interaction to expand?
... if displays everything no matter what actually creates other problems
Patrick: I can envision cases
where the changes more dramatic than simply flipping visibility
on and off -- triple nested constructs were UA may have a tough
time, finds and matches but then triggers a whole range of
different stylings because even though actual container
contains that word, parent it's not visible, might need more of
an algorithm to say go off the tree and make sure it's...
... visible throughout
Simon: just grab the surrounding text so you can see what the surrounding text is
Kelly: we have other guidelines like this -- something is done through script that the user agent can't detect. We have a qualification. I still think this is good to get to the cases where we want. Might want to tweak the examples.
<Greg> Re searching stuff not currently visible, AND re expanding containers to accommodate their content, many Web pages implement menus as "sprites" wherein all the menu items are separate regions within a single image, only one of which is visible through the viewport on the image; in order to display another menu item elsewhere on the page, the image is scrolled behind the viewport and the...
<Greg> ...viewport is moved on the screen. This would be problematic for both SC.
Patrick: not guarantee -- might be instances where can't
Kelly: move advanced fine to Triple-A, otherwise as is?
no objections to that
Correction: "advanced find"
<jeanne> ACTION: JS to update the document with new text for 4.6. See minutes of meeting and http://lists.w3.org/Archives/Public/w3c-wai-ua/2010JanMar/0009.html [recorded in http://www.w3.org/2010/01/21-ua-minutes.html#action03]
<trackbot> Created ACTION-264 - Update the document with new text for 4.6. See minutes of meeting and http://lists.w3.org/Archives/Public/w3c-wai-ua/2010JanMar/0009.html [on Jeanne Spellman - due 2010-01-28].
Kelly: process -- if you have a proposal that is pretty solid, send to Jeanne, will put on a survey so the other members can respond
<kford> Survey link: http://www.w3.org/2002/09/wbs/36791/20100112/
Greg: we were looking for a way
to do something that hasn't those far been in the document --
trying to come up with a new term that we could define that
would include both elements in the user interface and elements
of the rendered content. Propose the term recognized
... the term recognize right now is only used for content, not user interface elements -- we are open to a different term
Patrick: defined in brackets rather than new term?
Greg: distinguish things rendered and things it recognizes -- doesn't recognize its flashing, for instance
Patrick: something along the lines of the user agent should never display any content that flashes... then in brackets specifying that this applies to both the user interface and the content of the document or page-- something along those lines. In brackets specifying both UA and content
Bim: What if of the flashing is in a movie.
display but does not recognize
Greg: another term that has a
special meeting -- can't use the word content if we mean UA and
... a distinction if you imagine that the media player can detect, does video that turn into recognizable that every other user agent is required to detect and suppress?
<Greg> Adapting from Patrick's suggestion, how about something like "In its default configuration, the user agent does not display any user interface elements or recognized content that flashes..."
Patrick: recognized is more applicable because you can say something's recognized by a general rule but then there's an edge case and should be recognized but is not -- get out clause because such an edge case...
Bim: I think recognizable is content that is capable of recognition as opposed to content that has been recognized -- otherwise it's a bit of a get out
Patrick: example Blink tags -- if we don't look for an actively it's not recognized
Greg: the question of recognized versus recognizable as a general one that would apply in many cases throughout this document
Kelly: defined in the
... user interface certainly should understand all of its own elements that it's creating
... I could live with separating them out
Greg: can't use display because its visual, can't use render because that's content, how about present
<Greg> No, I was wrong, display is fine in this specific context.
<Greg> New version: 4.4.1 Three Flashes or Below Threshold: In its default configuration, the user agent does does not display any user interface elements or recognized content that flashes more than three times in any one second period, unless the flash is below the general flash and red flash thresholds. (Level A)
<Greg> 4.4.1 Three Flashes or Below Threshold: In its default configuration, the user agent does not display any user interface elements or recognized content that flashes more than three times in any one second period, unless the flash is below the general flash and red flash thresholds. (Level A)
<Greg> 4.4.2 Three Flashes: In its default configuration, the user agent does not display any user interface elements or recognized content that flashes more than three times in any one second period (regardless of whether not the flash is below the general flash and red flash thresholds). (Level AAA)
Patrick: we definitely want it to be default configuration
<jeanne> ACTION: JS to update document with new text for 4.4 [recorded in http://www.w3.org/2010/01/21-ua-minutes.html#action04]
<trackbot> Created ACTION-265 - Update document with new text for 4.4 [on Jeanne Spellman - due 2010-01-28].
Kelly: we want to float out there default configuration
Bim: default would be essential for the UI because someone could get ill
Kelly: Next issue, similar, has
to do with audio and stopping and starting.
... Mark proposed perceived instead of recognized
<jeanne> ACTION: JS to write status and announcements for next Working Draft. [recorded in http://www.w3.org/2010/01/21-ua-minutes.html#action05]
<trackbot> Created ACTION-266 - Write status and announcements for next Working Draft. [on Jeanne Spellman - due 2010-01-28].
Kelly: face-to-face meeting in February in Austin, let us know who can and can't attend by next week, and we should be able to work out the details for remote participation
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) No ScribeNick specified. Guessing ScribeNick: KimPatch Found Scribe: Kim Default Present: sharper, Jeanne, Greg, +0190476aaaa, +035840024aabb, KimPatch, kford, Patrick, Bim Present: sharper Jeanne Greg +0190476aaaa +035840024aabb KimPatch kford Patrick Bim Kelly Kim Mark Regrets: Jim Jan Found Date: 21 Jan 2010 Guessing minutes URL: http://www.w3.org/2010/01/21-ua-minutes.html People with action items: js patrickl pl[End of scribe.perl diagnostic output]