Accessibility Guidelines Working Group Teleconference

30 May 2017

See also: IRC log


AWK, ChrisLoiselle, Joshue108, Kathy, jasonjgw, Detlev, davidmacdonald, Laura, Wilco, JakeAbma, MelanieP, wayne, Makoto, MichaelC, JF, MikeGower, Katie_Haritos-Shea, kirkwood, Mike, Elledge, KimDirks, steverep, marcjohlic
Lauriat, Mike_Pluke, Nurthen, EA_Draffan, bruce_bailey, pietro, Alastair


<Joshue108> trackbot, start meeting

<trackbot> Meeting: Accessibility Guidelines Working Group Teleconference

<trackbot> Date: 30 May 2017

<Joshue108> Chair: Joshue/Andrew

<scribe> scribe: Detlev

<ChrisLoiselle> I'll do June 27th as fallback scribe if need be too.

<JakeAbma> presnet+ JAkeAbma

ACT TF Update

Wilco: ACT TF has published FPWD, not so much feedback so far

<Wilco> https://wilcofiers.github.io/wai-act-quickref/

Wilco: couple of requests: TF request advice on rules repository - work on building it has started

<lisa> trouble joining

<Wilco> https://wilcofiers.github.io/act-rules/rules/ACT-R1.html

Wilco: trying to figure out how to integrate conformance rules into quickref
... link to draft of what a conformance rule should look like, using markdown formatting

<lisa> made it

Wilco: second thing is a list of test case repositories

<Wilco> https://www.w3.org/WAI/GL/task-forces/conformance-testing/wiki/Testing_Resources

Wilco: TF is collecting published snippeds of HTML code with good/bad a11y examples
... there is a buch there already like from openAJAX alliance Quail? Google Chrome, Ben2Web
... tell TF if it is missing anything - wants to build up a big repository so automated test tools can be compared to the repository

Josh: specificall for automated tools?

Wilco: not exclusively - includes best practices, also from training libraries, say from Deque Uiversity

Help: https://www.w3.org/2002/09/wbs/35422/COGA_help/results

<Joshue108> https://www.w3.org/2002/09/wbs/35422/COGA_help/results

Josh: look at feedback to help SC

Lisa: Was put together with Jeanne and Michael Cooper in relation to the concept of 'pillars'
... trying to get something in that gives you a feel for the topic

<Joshue108> https://rawgit.com/w3c/wcag21/provide-support_ISSUE-32/guidelines/sc/21/provide-support.html

Lisa: synergy would be if there is more details in the supplementary info
... improvement and improved understanding would come in WCAG 2.1
... not much work done so far on the definitions
... should become more precise and clearer
... if you don't have complex content then you don't need to do anything
... so comprehension support is only meant for complex contetn
... if you feel there should be a different presentation or note, tell Lisa
... should cover all sorts of situations related to complex situations - if necessary individual items in the list could be dropped - what is important to know now is if the general direction is OK
... awar ethat more work is needed

Josh: the main thrust is that support is provided for numerical complex info, long texts etc

Lisa: includes summaries, highlighte dkeywords etc.. is often badly done, but for more specifi advice you will have to go for the supplementary info

<lisa> maybe michael want to add to hat I said?

<Zakim> JF, you wanted to ask where the latest final language is?

<lisa> current text https://rawgit.com/w3c/wcag21/provide-support_ISSUE-32/guidelines/sc/21/provide-support.html

JF: Has been struggling with this SC - what is the most current text? SC tries to solve a lot of different things in one SC
... controls is very different from providing summaries
... complex interactions like custom controls is very different from tackling complex content - should that be split into two SCs?

<Wayne> q

Lisa: there may be not enough time for that

<Joshue108> +1 to JF

JF: Better produce something solid then rush with this one

Lisa: Tackling complex info is probably the most important - providing a list of 'one of the following' was an attempt to tell authors what the ycould do, leaving loopholes
... So it is menat for people who want to do the right thing, accepting the loopholes
... If we lose pulling them together it is difficult to address dyscalculia

Josh: Why is it not called out directly (Dyscalculia) or difficult to address?

Lisa: Giving users some way out (alternative words, keywords) can prevent dyscalculia
... It is a burden since it makes you think about new user groups and requirements - if it is dropped it does not surface in WCAG 2.1

DMD: Maybe one SC with two sections
... Discussion of wording suggestions in Survey

<Zakim> Wayne, you wanted to say

Wayne: Unsure how th ecode level semantics would work (?)
... how is th eauthor going to include the semantic markup for this

Lisa: The semantics that is being made (ARIA for Coga) offers ways of including alternative wordings, alternatvie for directions etc., but that would be at Technique level

Wayne: Where is that?

<MichaelC> https://w3c.github.io/personalization-semantics/

Lisa: (looking)

<Zakim> Joshue, you wanted to say that this is useful and may be good split up

Josh: Suggestions to break this SC up seem relate dto testability - looks like a triple A thing right now, but a good idea, should be firmed up

JF: Concern 1: Building SC with loopholes

Concern 2: ARIA for coga is very much draft proposal currently - authors need something that is fir for use today

Lisa: nothing in the SC relies on ARIA for coga

<Zakim> laura, you wanted to say: If we want to combine per John’s suggestion…maybe combine the 2 first bullets as the both deal with summary? And then explain in the understanding doc

Laura: If it should be shorter the first two bullets might be lumped together (both concerne dwith summaries)

Lisa: This would make it more confusing - you have a chart for complex numerical info and a summary for long texts so these are different things
... would make it shorter but less clear

Wayne: Will have to look at it in detail, find it a bit confusing - there is ambiguity in the ways browsers handle (?)

Josh: walk through concerns in survey

<Joshue108> +1 to AAA for this.

Jason: Details are in written comments - the essence is that while all techniques ar evaluable, they are not broadly applicable to the web as a whole, so they seem to belong to level AAA
... more work would be needed to make it feasible to be applied to the web in general and then be positioned on level A or AA

<Wayne> There are many ambiguities within W3C standards. How layout is accomplished at the borwser level varies across browsers and there is no contradiction with the HTML or CSS standards. The ambiguity here is that this example is confusing.

Jason: issues arise when this is applied to web content, would be very different in different contexts and with different types of content - so the applicability would need to be specified a lot more

Lisa: There are clear definitions, like texts above 300 words would be defined as long, needing a summary
... clear that a better definition for complex numerical info is still needed

<Ryladog> I would prefer to see this split up into several distinct SC, and not at Level AAA. I would prefer to see those that can be, placed at Level AA.

Lisa: why would text length definitions not be enough?

Jason: There are internationalization concerns; what is the summary supposed to achieve?

Lisa: There can be bad summaries but that is a separate issue - but summary is widely applicable

Josh: scope needs to be narrowed - lots of useful commehnts to work with

Lisa: please update your comments

orientation https://www.w3.org/2002/09/wbs/35422/MATF_orientation/results

change of content: https://www.w3.org/2002/09/wbs/35422/SCreview_May_17/results#xq10

DMD "owns this guy" :)

DMD: Comments were addressed

<Joshue108> https://rawgit.com/w3c/wcag21/change-of-content_ISSUE-2/guidelines/sc/21/change-of-content.html

DMD: editorial comments have been addressed, SC has been updated

Josh: Look at changes, feel free to get in touch with David

Michael: comment "screen orientation" instead of just "orientation"

DMD: will change that

<Joshue108> https://www.w3.org/2002/09/wbs/35422/SCreview_May_17/results

Alex: Which one are we talking about ?

MichaelC: change of content

AWK: Put in some scenarios to test out the SC
... when submitting a form and you go elsewhere then th eSC would not apply
... if you put things in the cart you get a live regions update but would you need another update "your cart has been updated"

DMD: if the cart content increases, a notification would be fine (also without live region)

AWK: what is the requirement for "programmatic notification"?

DMD: AT has to be able to determine that things have changed - we might add language to improve understanding

AWK: if you scroll around page and new content is added is this just a live region issue or doe si tnee da separate announcement?

DMD: brief one: like "new content added"

AWK: if it updates frequently in differnet places how would that be handled?

DMD: there was an exception "more than 5 times a minute" which was taken out due to sme responses

AWK: So then you would not have to do anything?

... other people will ask the same question, maybe it needs exceptions, or narrowing the SC to user-activated things
... but that would nit serve the community as well as covering any changes

<Alex_Li> can you define what is an action taken?

DMD: the alternative would be to reintroduce frequent update exception
... would a big clock not fail 2.2.2? Probably not?

Josh: How would cange of content relate to client-side error validation?

<AWK> AWK; also the clock is essential for a greenwich mean time page

DMD: Yes, would produce message "errors on page, please review" or similar

Katie: We have been looking for that - the exception for frequent changes seems useful

<Joshue108> https://www.w3.org/2002/09/wbs/35422/SCreview_May_17/results

DMD: Is in the survey
... reading Detlev's concern in survey
... reading MichaelCs concern in survey

Katie: Can we come up with a number, like 10 per minute?

<Zakim> Joshue, you wanted to ask if Prog notif is suitable here - would notification be sufficient as prog deter is in the qualifiying statements

Josh: wondered if 'programmatic' should be dropped from notification

DMD: is needed

<gowerm> aria-live roles: timer https://www.w3.org/WAI/PF/aria-1.1/roles#timer

DMD: Cna be widened, better explained, but generally looks good

Josh: things it is good to go

<Zakim> MichaelC, you wanted to say s/programmatically determinable/programmatically determined/

<gowerm> marquee https://www.w3.org/WAI/PF/aria-1.1/roles#marquee

MichaelC: "programmatically determinable" should be "programmatically determined"

gowerm: Pasted things that can be applied to live region
... marquee - migh tbe useful
... time addresses AWK's page scenario

timer not time

Alex: There is already a realtime exception in 2.2.2 - why not copy that?

DMD (weighing relevance of exception)

gowerm: caution: in situation where info is added as in this IRC the exception should apply

DMD (editing)

Josh: goes in the right direction, issue scan be ironed out

Jason: Will look at revised versions, not clear what is required when DOM is changed - should require to distinguish between significant and insignificant changes,

<gowerm> David, I think you can have a technique that flags with role="timer"

Jason: if there are several updates occuring (atomic) only one update might be needed, needs to be clarified what needs to be done on top of updating the DOM in the browser, that the changes are important enough for the user to know about
... some update might be part of the same change so that needs to be taken into account

DMD: distinction between essential (prinary purpose of the page) and inessential is already there

atomic vs. non-atomic: if true, it presents the whole changed region to the user ("you have 15 sec left" instead of just "15"

scribe: we can include something "where the change has been sufficiently descriptive" (?)
... weary to put atomic / non-atomic in it

Josh: should go into techniques

Jason: When authors provide notification od each change they might thing they have satisfied that - that's a danger
... programmatic notification that changes are significant

<Wayne> Does notification cause visual loss of place in the document? Can you turn notification off?

DMD: no requirement for a visual notification, where it exists it has to be programmatic also

There is a link to the coga SC Feedback

Josh: Good issue that notifications may not be visible / noticeable when magnification is used

DMD: focus in this here is programmatic notification
... notifications (noted by Wayne) may also be obstructive visually

<gowerm> David, possible exception "The change of content results from auto-updating content that is an essential part of an activity"

DMD: no requirement here to move the focus to th enew information

Alex: (reads SC text - confirmed): what means "action taken"?

DMD: a user action

Alex: any action of the end user?

DMD: If the content changes - that would not happen as a result of scrolling or moving the cursor

<Ryladog> +1 to adding "user" in front of action

Alex: at least chang eto "user action" - also be cleare about defining what constitutes an action

DMD: "user action that changes the content" - so the notification confirms that

Josh: what about changes of content without user action?

<Joshue108> https://rawgit.com/w3c/wcag21/change-of-content_ISSUE-2/guidelines/sc/21/change-of-content.html

DMD: ..OR "that conveys information" - so it is not only user-initiated

8discussion of details of wording of the SC

<Zakim> AWK, you wanted to suggest change from "The user has been advised of the behavior before using the control" to "The user has been advised of the behavior before or as a result of

<gowerm> "The change of content results from auto-updating content that is an essential part of an activity"

DMD: updated SC text

<Wayne> +1 to include now, leave editorial changes to David's discression.

gowerm: pasted in wording suggrstion above

<davidmacdonald> https://rawgit.com/w3c/wcag21/change-of-content_ISSUE-2/guidelines/sc/21/change-of-content.html

<Mike_Elledge> Good luck, Wayne!

<AWK> "The user has been advised of the change of content before or as a result of using a control"

AWK: goal is to infom the user about important changes - worried about second bullet
... would allow a web page that pops up a dialogue annoncng the change rather then using aria-live on a car, or similar
... "advised of the behaviour" seems strange

DM (editing, committing changes)

Steve: excited and scared by this SC as SR user (because there is a lot of misuse of ARIA) - the SC does not delineate how polite that notification needs to be - is there any way to deal with that?

DMD: the spec says 'use polite unless the house burns' - so the spec is pretty clear

Josh: a matter of Techniques

Steve: fears it is buried in Techniques, would like to see it up front in the SC

<AWK_> "programmatic notification is provided for each ESSENTIAL change of content.."?

DMD: you can turn off live-regions

<AWK_> or adding "essential" inside of "conveys information"?

Katie: if we include an exception, include times / frequency

<Ryladog> Add to exception that "or it can be turned off"

goverm: some techniques refer to leve regions; would like to get it out and get public comments

<marcjohlic2> +1

<gowerm> +1

<JF> +1

<davidmacdonald> +1


<AWK_> +1

<ChrisLoiselle> +1

<Ryladog> Add to exception that "or notification can be turned off"

<KimDirks> +1

<Joshue108> +1

<laura> +1

<Ryladog> +1

<JakeAbma> +1

<kirkwood> +1

<Alex_Li> 0

<Makoto> +1

<Kathy> +1

RESOLUTION: The working group has accepted this SC for inclusion in the Editor's draft

Josh: Will put our a CfC to the list

<Mike_Elledge> bye all!

<Mike_Elledge> +1

<AWK_> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

  1. The working group has accepted this SC for inclusion in the Editor's draft
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/05/30 16:31:26 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/prodie/produce/
Succeeded: s/sexception/exception/
Succeeded: s/Jofh/Josh/
Default Present: AWK, ChrisLoiselle, Joshue108, Kathy, jasonjgw, Detlev, Laura, Wilco, JakeAbma, MelanieP, wayne, Makoto, davidmacdonald, MichaelC, JF, MikeGower, Katie_Haritos-Shea, kirkwood, Elledge, KimDirks, steverep, marcjohlic
Present: AWK ChrisLoiselle Joshue108 Kathy jasonjgw Detlev davidmacdonald Laura Wilco JakeAbma MelanieP wayne Makoto MichaelC JF MikeGower Katie_Haritos-Shea kirkwood Mike Elledge KimDirks steverep marcjohlic
Regrets: Lauriat Mike_Pluke Nurthen EA_Draffan bruce_bailey pietro Alastair
Found Scribe: Detlev
Inferring ScribeNick: Detlev
Found Date: 30 May 2017
Guessing minutes URL: http://www.w3.org/2017/05/30-ag-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]