W3C

- DRAFT -

Web Content Accessibility Guidelines Working Group Teleconference

19 Aug 2014

See also: IRC log

Attendees

Present
+31.30.239.aaaa, [FordMotor], +1.617.766.aabb, AWK, Wilco, Bruce_Bailey, David_MacDonald, Michael_Cooper, cstrobbe, Marc_Johlic, +1.703.637.aacc, Kenny, jon_avila, +44.172.172.aadd, alistair
Regrets
Sailesh, James_N, Loretta, Kathy_W.
Chair
AWK
Scribe
Bruce

Contents


<trackbot> Date: 19 August 2014

<AWK> Chair: AWK

<alistair> Running a couple of minutes late...

<AWK> Scribe: Bruce

<alistair> I'm having problems phoning in...

<alistair> Is anyone else having problems with the phone bridge

Feedback on Developers' Guide to Features of Web Accessibility Evaluation Tools 29th July 2014 https://www.w3.org/2002/09/wbs/35422/29thAugust2014/results

Please refresh on survey, should have six replies

Discussion if we turn over comment or other...

AWK: no burning disagreement

MC: not an approval review, so less formal

<Wilco> +1

AWK: question to dave 14/20

David: Did study for Canada, its ballpark figure from experience

David Grove came up with 18% in his report

MC: number is soft

hopefully, will increase over time

MC: with more processing power, AI, more machine testability is feasible

Wilco, +1 percentage can only go up

\

AWK: Is candaian report public?

MC: David will ask

AWK: Would be nice for citation to justify automatic evaluation being between 13-20 %

We have two studies, better than subjective expert approximations

David: Parts of study are pretty candid, esp. with regard to some products

May be able to post excerpts

AWK; Prolly take a while to get approval, Shadi can just go ahead

MC: Need to be clear that tools can help, but not do work

bruce: wrt Alistair comment

Alistair suggested new name change

Bruce asks if content just needs tuning

MC: Both Qs go to ERT, they can decide

AWK: Will make Shadi aware of our comments, take under advisement. Back to us if more questions or clarification or more harmonize response

MC: confirms this is already for public review with relatively short turn around

AWK: Consensus to share survey results

AWK sent email while on call

Survey from August 5th- https://www.w3.org/2002/09/wbs/35422/20140805/results (only #s 4 and 6).

https://www.w3.org/2002/09/wbs/35422/20140805/results#x2652

Questions for foreign language

4. Captions for video in foreign language

AWK: No need to debate too long, still may not have definitive answer

AWK summarizes question

David: Have folks seen thread from public email?
... Deaf advocates often feel strongly about issue

<cstrobbe> See thread at http://lists.w3.org/Archives/Public/w3c-wai-gl/2014JulSep/0163.html

Mixing translation with transaction is an error

Cleanest rule is to follow language of speaker

Providing multiple track can lead to difficult if subtle problems

AWK: Simplest is for captions to be in language of video
... Does not address needs of Deaf who are not speakers of language

David: Lack of subtitles (translation) issue for many people, not PWD in particular

AWK: Example, how do I make French video for English language audience?
... If I address needs of English speakers, have subtitles, but subtitles miss other auditory information

David: Trying to channel Loretta, others would miss things as well.

AWK: SC includes requirement for non-spoken information.

Marc: Discussion seems tricky
... English captions different than translation

Often English language captions just say [GERMAN] when language changes

English language film where someone speaks foreign language is pretty common

Caption should not be translated

cstrobbe: example of Japanese video, why would only English portion be subtitled?

Could just be languge in video

intent of video is important

Mike Elledge: It comes down to author intent

If no translation in default presentation, then no requirement to translate

This is for where film has portion of foreign language

Mike: Issue is more for fully dubbed video
... What about blind and audio description?

WCAG required audio description for prerecorded media

AWK: Different requirements for AD, and user profile

sub/Mike: Issue/Wilco: Issue/

Mark: Have to look at intent.

Example of Japanese video, non-spoken audio part needs to have captions in Japanese

<Wilco> +1

WCAG cannot require translation, could not be a failure

AWK: Agree, but WCAG does not clearly address issue of dubbed content

Response could be captions need to be in language of video

And if subtitles are provided, they need to be for broadest audience possible

Marc: Issue is foreign language video for English speaking audience

If we remind folks about Japanese captions, need to remind about Japanese AD

<marcjohlic> +1

cstrobbe: If the video is dubbed, the language of the captions need to follow the language of the dubbing

If language on film changes, captions reflects language heard

AWK: Offers to rewrite based on feed back

Will come back next week

RESOLUTION: Leave open based on conversation

<marcjohlic> http://www.w3.org/TR/2014/NOTE-WCAG20-TECHS-20140408/G93

<marcjohlic> http://www.w3.org/TR/2014/NOTE-WCAG20-TECHS-20140408/G87

<AWK> http://www.w3.org/TR/WCAG20-TECHS/G87.html

Wilco: Raises issue with accuracy/completeness in G93/87

<David> +1 on accuracy check

Wilco: Failure tests need to include accuracy

Other common captioning expectations (speaker identification, non obstruction) also absent from test procedure

AWK: Do we recall debate?

David: Accuracy important and implied if not explict

Techniques like Google automated captions not sufficient

Jon: Don't for important sound effects (F8)

http://www.w3.org/TR/2014/NOTE-WCAG20-TECHS-20140408/F8

F8: Failure of Success Criterion 1.2.2 due to captions omitting some dialogue or important sound effects

AWK: Don't want to have success conflicting conflict with Failure

Could resolve by adding F8 as related technique for each?

<David> We should probably be consistent in test procedure...

Can we put intent into test procedure

<Zakim> Bruce_Bailey, you wanted to ask if best practices in SC ?

Bruce: thinks inaccuracy is a failure

Obscuring faces is not a failure, just a bad practice

Jon: These are mapped to both live and pre-recorded

Would be better if expectations for live were easier than prerecorded

Live captions will have more errors

AWK: Good live captions different than good recorded captions

Hard to capture language

Wilco: Gives example of two speaker, other text on screen

<jon_avila> not obscure a crawler?

So bad placement could be failure

AWK: Context makes difference, example sports program

Marc: Trying to separate point, F8 seems good to capture bad quality

Missing speaker identification and obscuring other text on-screen not reflected as failure

AWK: Change test procedure, rather than check captions are visible

something like check that caption meet WCAG definition for captions

David: Check that captions capture the content?

People just turning on YouTube captions need to be clearly insufffient

Also, testing live content not really an issue, since its live

By the time people are evaluating, as a matter of practice, captions are prerecorded so expectation can and should be high

AWK: Need to be able to fail practice for live video, example CNN or other live news

David: Language like "represent the content" may sufficiently flexible to reflect industry norms and practices

Phrasing as simple as "caption accurately reflect content" gets at difference between live and prerecordd

Marc: Still want to get at obscuring piece
... add "without obscuring important information"

<marcjohlic> Check that captions (of all dialogue and important sounds) are visible and accurately represent the content without obscuring any important information.

+1

<marcjohlic> alt: Check that captions (of all dialogue and important sounds) are visible and accurately represent the content and do not obscure any important information.

AWK polls for any objection, none heard

AWK: Need to figure out where to make these changes

<AWK> Possible places for change: g87, g93, SM11, SM12, F8

<AWK> others

<AWK> ?

<scribe> ACTION: marcjohlic to create wiki page documenting all places for this edit [recorded in http://www.w3.org/2014/08/19-wai-wcag-minutes.html#action01]

<trackbot> Error finding 'marcjohlic'. You can review and register nicknames at <http://www.w3.org/WAI/GL/track/users>.

<marcjohlic> mjohlic

<scribe> ACTION: mjohlic to create wiki page documenting all places for this edit [recorded in http://www.w3.org/2014/08/19-wai-wcag-minutes.html#action02]

<trackbot> Created ACTION-269 - Create wiki page documenting all places for this edit [on Marc Johlic - due 2014-08-26].

6. Improvements to Understanding and Techniques

https://www.w3.org/2002/09/wbs/35422/20140805/results#xq6

AWK: Material is challenging for multiple consumers

This survey item is offering a chance for group members to offer thoughts and suggestions about ways to improve these documents, or to offer thoughts about what problems they experience personally or hear about from others. It is also appropriate to indicate what you like about the documents.

Issue from last week is that you cannot tell if technique is advisory or requirement

The issue from last week is that you cannot tell if technique is advisory or requirement

When you are only looking at a particular technique

Also, cannot tell if a technique is part of an AND or OR processs

People don't understand how important test techniques are

AWK summaries comments from survey, some detailed, so quick

sub /detailed, so quick/detailed, some quick/

<jon_avila> I still see a lot of people who think that if you don't meet a sufficient techniques it is a failure of a SC.

<David> Agree...

Bruce clarifies that editorial work important to consumption

AWK: New current drafts trying to clarify that techniques are informative, not requirements

<scribe> New boiler plate elevates disclaimer, but template ripe for discussion

AWK invites discussion

AWK: How should we approach this? What would you do with unlimited control and money?

<jon_avila> perhaps organization by media type, forms, tables, etc.

David: Better search capacity. People remember a couple words, sentence, tag, but can't find it again

Wilco: If you don't know title, can't really expect to find technique

Tags in technique could be very useful

<jon_avila> +1

<Mike_Elledge> +1

<AWK> http://www.w3.org/WAI/GL/2014/WD-WCAG20-TECHS-20140724/G83.html

<David> +1

Example is one that was rewritten recently

First heading is "Important Information about Techniques"

AWK: Is there a better way to do address this issue?

Alistair: would be great to have techniques related to tag as developer

So in IMG different techniques could be available in authoring tool

Describes how could work in practice as a developer

All part of an IDE, tool extract information from techniques documents

Would take a lot of rejigger for tools to scrape our docs

Jon: People have nav bar blindness, just moving makes little difference

Nice for aggregation of resources, listserve, blogs, forums

Not getting these as submitted techniques

AWK: Should people be able to comment on techniques?

Presentation is very plain, might be turning some folks off

Might be good to pull in people with great design skills

Great design prolly hard for volunteers, but WG members may have this skill set

David: Reminds us that Shawn Henry address many presentation issues back in 2007

David clarifies that this was WCAG proper and How to Meet only

AWK: So, how to proceed? Need specific proposals on what to change.

Could have survey for outside of working group, call for suggestions

Calls for volunteers on how to solicit feedback

Q to MC about process

MC: Wording and expectation setting could be tricky

AWK: Looking for brainstorm wrt public feedback

David: WET folks have expertise

Might even do a little mock up

AWK: Need to figure out what we want

Don't want design work until we have more concrete ideas

Not even up for wire frames at this point

What do people like?

What don't they like?

What issues are people facing?

Marc will ask some suggestions / feedback from a couple of design folks on my end.

AWK: Ask around, come back to group, feedback to AWK.
... Out of time most unfortunately

Alistar has been kept waiting

Alistair: Responsive design is bad trigger word

Thinks he has addressed issue

addressed issues raised before.

Alistair and AWK concur that processing by working group takes took long

Alistair: First survey went through without comment, request new survey from scratch

Old title and re-directs cause confussion

AWK: We will get to this soon, but prolly not next week

Technique will have to wait for March edition

<Mike_Elledge> bye all!

<AWK> trackbot, end meeting

Summary of Action Items

[NEW] ACTION: marcjohlic to create wiki page documenting all places for this edit [recorded in http://www.w3.org/2014/08/19-wai-wcag-minutes.html#action01]
[NEW] ACTION: mjohlic to create wiki page documenting all places for this edit [recorded in http://www.w3.org/2014/08/19-wai-wcag-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014/08/19 16:34:19 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/accuratecy/accuracy/
No ScribeNick specified.  Guessing ScribeNick: bbailey
Found Scribe: Bruce
Default Present: +31.30.239.aaaa, [FordMotor], +1.617.766.aabb, AWK, Wilco, Bruce_Bailey, David_MacDonald, Michael_Cooper, cstrobbe, Marc_Johlic, +1.703.637.aacc, Kenny, jon_avila, +44.172.172.aadd, alistair
Present: +31.30.239.aaaa [FordMotor] +1.617.766.aabb AWK Wilco Bruce_Bailey David_MacDonald Michael_Cooper cstrobbe Marc_Johlic +1.703.637.aacc Kenny jon_avila +44.172.172.aadd alistair
Regrets: Sailesh James_N Loretta Kathy_W.
Found Date: 19 Aug 2014
Guessing minutes URL: http://www.w3.org/2014/08/19-wai-wcag-minutes.html
People with action items: marcjohlic mjohlic

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]