W3C

- DRAFT -

Accessibility Guidelines Working Group Teleconference

08 Aug 2017

See also: IRC log

Attendees

Present
AWK, Rachael, Kathy, Joshue108, JF, shadi, MikeGower, david-macdonald, KimD, alastairc, Elledge, Makoto, Detlev, kirkwood, jasonjgw, JakeAbma, steverep, Melanie_Philipp, Glenda, bruce-bailey, Avneesh, jon_avila, Mike
Regrets
Chair
AWK
Scribe
Mike_Elledge

Contents


<AWK> Agend?

<KimD> *yes, I had the same issue, Andew

<Detlev> The webex link above points to the Thursday meeting...

<Detlev> Have trouble starting WebEx on the Mac... any ideas?

Looking for scribes

Alex will scribe next Thursday

<Detlev> damn

<AWK> Scribe: Mike_Elledge

meeting time (https://www.w3.org/2002/09/wbs/35422/mastiempo/results)

Awk: Idea that we can use extra time. Everyone responding to survey said okay to add half hour. Tuesday 11-1. Thursday 7 can, 3 conflict.
... other thoughts?
... Would like us to get through the next couple of weeks and see if we need the extra time. Another option just do Tuesday with extra time and see if we need Thursday.

<Detlev> I'm in - Mac users, use Firefox, not Safari

<bruce_bailey> +1 to two hour call Tuesday, especially since I have not been able to make Thursday calls

alex: Depends where we are in process. Don't want to do perpetually. It would add up. If we can make real progress, don't fill up time by talking. Only if we help productivity.
... Make judgement a week in advance.

awk: agree. Challenging to make judgement a week in advance. Ppl don't have calendars set. For month of August seems worthwhile. Survey set to go. Beneficial and well-structuerd enought.

jf: Don't know how got into queue. Having agenda with firm times will help. Have to be careful not to let it spill over. On board with adding time for next 6 weeks.

awk: Try for rest of August. Will reduce time if don't fill out schedule.

<Glenda> +1

awk: Start now or Thursday.

<Detlev> =1 for etending tues call now

<Detlev> =1

<Detlev> +1

awk: Start Thursday. Next week will split scribing in half too.

RESOLUTION: Extend the calls starting Thursday August 10. Split scribing for long call.
... Extend the calls starting Thursday August 10th. Split scribing for long call.

timeouts (https://www.w3.org/2002/09/wbs/35422/Final_prelockdown_set/results#xSC1)

<AWK> +AWK

Concurrent Input Mechanisms (https://www.w3.org/2002/09/wbs/35422/Final_prelockdown_set/results#xSC8)

awk: Reminder use + with name to sign in.
... Discussed last week after ppl had dropped off. Survey: 12 ready to go, 1 have issues. Jason: Make sure do work in other document. Agree. Alex: wordsmithing. Saw wordsmithing and raised you!
... have you seen?

<AWK> AWK's edits: Web content does not restrict use of input modalities available on a platform except if the restriction is necessary to ensure the security of the content, or if the restriction would invalidate an activity or override a user setting.

alex: just did.

awk: Believe says same thing as intended. Addresses potential confusion of meaning.

alex: My intention was use of essential. Essential well-known in wcag 2.0.

<david-macdonald> 2.2.1 Essential Exception: The time limit is essential and extending it would invalidate the activity;

awk: Use essential.

alex: use negative term now. would invalidate. not same language.

awk: does security fall under essential as well?

alex: don't know, ususally content not security.

dm: do have a precedent for invalidate. not close enough to say.

detleve: IRC

<AWK> AWK: Web content does not restrict use of input modalities available on a platform except if the restriction is essential, or is required to ensure the security of the content.

awk: suggestion, captures security as well as content.

dm: looks alright.

<alastairc> "except if" sounds odd, wouldn't it be: "except content where"

<Alex> Web content does not restrict use of input modalities available on a platform except if the restriction is essential, or is required to ensure the security of the content or respect user setting

<Detlev> you are very quiet

kathy: fine. add understanding to clarify the point. make sure Thursday callers okay with that.
... need to make sure lisa is okay with that.

alex: user setting for platform or content?

jf: could be user settings in future.

alex: which is it for? user agent or AT?

jf: user agent stands for user, AT, browser, don't really care so long as it is respected.

alex: if ambiguous hard to tell what failed.
... shouldn't it be about the content.

<AWK> Possible new version, based on Alex's:

<AWK> Web content does not restrict use of input modalities available on a platform except where the restriction is essential, required to ensure the security of the content, or respect user settings

jf: wondering if we have examples where it is happening today. alsmost testing for negative, but still valid.
... not aware of any code that's restricting it today. could test it then.

awk: didn't get any examples. Kathy and Patrick worked on it and fixed. Ppl provided examples of how it could take place.

alex: philosophically, do we want criteria where we don't find a real world situation. applies to a lot of content. "don't break something we haven't seen anyone break in a while."

kathy: look at video that was posted? patrick showed the issue. but it has since been fixed. Have found it.

detlev: some sites are coded for touch, but would not work for mouth. there are ways to do that. keyboard on ios devices, may need a keyboard for some inputs.

jf: agree that don't want to create requirements for things not in real world. strobing is a requirement, but not seen often. doesn't mean it doesn't exist if don't have lots of examples. will only grwo. so let's be pre-emptive.

<david-macdonald> https://www.youtube.com/watch?v=f_2GKsI9TQU

<AWK> Web content does not restrict use of input modalities available on a platform except where the restriction is essential, required to ensure the security of the content, or respect user settings

mike: +1

alex: ok

<Detlev> +1 to that

<marcjohlic> +1 to AWK's wording

<Kathy> +1

<alastairc> +1

<kirkwood> +1

<Joshue108> +1 to AWK

awk: valuable even if not common?

<KimD> +1

dm: we could put in draft. ask for more examples before we put it in. extra burden with each sc. onus to provide real world examples.

<david-macdonald> +1

alex: david took words out of my mouth. need real examples so we're not making it up.

awk: put in editors draft?

mike e: +1

<Detlev> +1

<Joshue108> +1

<kirkwood> +1

awk: any objection to put in editor's next public working draft (npwf)?

<Glenda> +1

<Kathy> I will update Github

<AWK> Kathy will update SC text with "Web content does not restrict use of input modalities available on a platform except where the restriction is essential, required to ensure the security of the content, or respect user settings"

<Kathy> yes

RESOLUTION: Accept as amended.

Popup Visibility (https://www.w3.org/2002/09/wbs/35422/Final_prelockdown_set/results#xsc11)

<Joshue108> \0/

<Joshue108> https://github.com/w3c/wcag21/issues/75

awk: issue 75: 6 ready, 7 issues.

<JF> I've not yet submitted the survey, but I too have concerns/issues

alex: mostly text is not clear. unintended consequences of ambiguous language. didn't have opportunity to connect with steve. no conclusion yet.

steve: made changes. interference to visibility. editorial to keyboard focus, another? term popup still an issue for some ppl. only term very close is hoverbox.

<steverep> https://en.wikipedia.org/wiki/Hoverbox

<Detlev> hover box even after receiving kb focus?

steve: not really used much--to our adnvantage?

jf: modal dialog?

<AWK> AWK: don't think that this is talking about modals

<Joshue108> JOC: No - thats why, popover as a term was discussed last week.

steve: popup is generic, defines many things. tool tip, modal dialog, problematic.

<AWK> current definition for popup: content which becomes visible only when associated content, called the trigger, gains keyboard focus or pointer hover

Steve: is that covered.

jf: js alert a popup?

<Detlev> that was steve, not alex I believe

steve: only if appears in hover focus.

<Alex> Mike, that's not me talking

<Detlev> Mike it is Steve speaking I b

<KimD> * "Hoverbox" - I like that it's not used often. We can define/refine the word's meaning

<Glenda> +1 (agree with KimD)

steve: disadvantage for hover focus. can appear on focus only. generally would be calendar or number pad popup. less probelmatic to users. hover seems to dissuade ppl from popup definition.

<Joshue108> JOC: Hoverpop would be nice.

steve: don't know how we would reword.

jas: not telling what we want it to, since not well-accepted term. wouldn't worry to use it in a draft. haven't heard any objections so far. putting in draft could inspire ppl to come up with different term.

<allanj> +1 to put in draft, get wider feedback/review

<david-macdonald> I think "hover box" is the way to go with a definition for it provided, to ensure we have what we need.

alex: couple of things. popup won't work. popup blocker wouldn't apply. need to invent term. lots of questions if trigger condition is problem and how it is a problem.

<AWK> AWK: If we are trying to suggest device independence I would discourage "hover box"

detlev: transient content. hover box has disadvantage of multi-centric?

steve: condition on the trigger is there to ppl with vision problems have to re-focus all teh time. Not expexting it. moving focus difficult if magnifying. may have have to pan screen if can't turn it off or if it doesnt appear near trigger.
... under magnification outside of view have to move pointer to other part of screen, or have to scroll.
... could have either or condition.

dm: what if button?

steve: rare if it appears over button. not covering re-styling.

dm: if popup obscures image, would want to press escape.

steve: most apps don't have that programmed in.

awk: don't have objectin to using popup, but hear what alex is saying. could find a new term, whether transient content or whatever, let's just call it "x". What precsiely are we trying to solve?
... feels like one scenario when going to menu, try to move mouse on content that has come up, go outside of trigger area and not to content, it will go away. Have to be very careful--drives me crazy.
... not sure that text covers that.

steve: in my experience problem when you don't have interactive components in content, then author makes it that only the trigger will cause hover. If move away it disappears.

<Detlev> maybe Jon Avila can add his view what we want to focus on?

steve: try to follow a narrow path to popuup, don't know how to describe w/o getting into pixels.

awk: example?

steve: examples page.
... wiki editor. comments button.

awk: "add your reaction". don't think of magnification situation where user can't see all the content.

steve: if using mouse could move it off and have it disappear.

awk: other use cases?

alex: hovering at "add your reaction".

<allanj> see also - https://www.w3.org/WAI/GL/low-vision-a11y-tf/wiki/Metadata_On_Hover#Perkins

<jon_avila> The member tooltip is an example that has more text.

awk: tooltip is pretty clsoe by to trigger, but sometime AT will read text on hover, will start to lose what's in tooltip, when move toward text it goes away.

steve: if you move your pointer towoard text goes away, but not if toward smiley face.

<jon_avila> Another example is on a screen that horizontally scrolls and the tooltip appears cutoff but when you scroll to see the rest of the tooltip the tooltip goes away

steve: want it to be clsoe enough not to trigger, but not so far away that it goes away when you zoom.

awk: for example, lengthy definition of a term can zoom in and not be able to see all. move to see rest and it disappears.

<Zakim> allanj, you wanted to say example

steve: if have lots of AT large pointers and accents, want to move to see it all.

<allanj> example https://www.w3.org/WAI/GL/low-vision-a11y-tf/wiki/Metadata_On_Hover#Cursor_Overlapping_Tooltip_Text_on_Hover_Example

jim: put link that says perkins above. hover over name and to goes away have to go very quickly to get to link that's in the middle.
... another example as well above.

awk: may be a platform issue, so need exception for situation where user agent controls hover.

jim: tooltip with actual information, ppl objected to <title> attribute.

alex: when you hover over any link, browser edge and chrome shows destination in lower left hand side of window; won't work in this case. browser dictates it.

awk: would be a problem for users, but not covered in this sc.

alex: it is a predicament.

steve: two schools of thought. Content therefore have to accept it. Or not content, generated by agent so should address it.

<jon_avila> Also hover content that is off the edge of the screen

ja: other issue is when hover is cut-off. when try to scroll to see it it goes away.

awk: do magnification tools compensate for this? a way to pan within an area without triggering?

steve: not using mag software. see it in github. tooltip off right of screen.

<Alex> Mike, that was steve talking :)

steve: not sure, it's not the default. dk if there's a keyboard short-cut or if it's programmed in. wouldn't be heavily used if it was. can check.

awk: wouldn't matter for the solution anyway, not everyone using mag tools.

<allanj> github uses CSS to display the aria-label

awk: just talking about when things are hovered over or focus; not clicking? like when menu popups options.

steve: yes in general.

<Alex> Mike, that's steve

<Alex> not Alex

steve: managing focus is a way to get around some of these issues. like mega menu, can put focus on component so you can pan around without worrying about it going away.
... goes away and you try to see subitem one, then try to see content on subitem two and it goes away.

dm: does same thing on focus as on hover will fail people.

steve: in many cases would cover it. if get content on hover with clickable components, will put focus to move it away, disappears.

dm: in general if hover over popup it will stay. move off of popup to make it go away.

awk: From survey: some issues: 1) using word popup. 2) make it clearer what it's trying to solve. 3) wordsmithing. What should we do.

<AWK> Josh: Suggests "reducce popup interference" as a good phrase

jo: found it useful to think in terms of popup interference. relates to megamenus being triggered accidentally too.

jason: what happens with rationale text. when ppl are deciding on proposal would be helpful to have rationale. useful also if tightening sc that we already have.
... good rationale is helpful for review.

<allanj> -1 hoverbox

<Detlev> I prefer transient content

awk: change popup to hover box? Hover box refers to pointer-specific functionality, wouldn't be right term for keyboard instance.

<david-macdonald> transient popup

<Joshue108> dont like the box in hover

<JF> -1 to hoverbox

<JakeAbma> -1 hoverbox

<Rachael> -1 hoverbox

<Joshue108> Something like popover was better..

<alastairc> I don't like the 'hover' in 'hoverbox'. Or box.

<Kathy> no preference as long as it is defined well

<Joshue108> lol

mike: hover content?

<Detlev> straw poll on transient content, perhaps?

<Joshue108> transoveronhoverentity?

<alastairc> NB: I'm not going to object though, I'd rather see it in the draft.

awk: address remaining issues. not insurmountable. but need clarification before goes in editor's draft. What do you think.

<Joshue108> I wont object,,

dm: put in editor's draft, clean up later.

<Kathy> I would like the definitions clarified first

<Joshue108> I think Jason is right about the definition being important

<steverep> +1 to David, cleanup later

<Glenda> +1 to put in editor’s draft (and wordsmith as we get public feedback)

<Joshue108> +1 to getting it in

jf: wordsmithing on list. general principle has acceptance.

<bruce_bailey> +1 to including in next draft

dm: require that it disappear...

<KimD> Agree, more wordsmithing issue than content/concept

<Alex> more wordsmithing

<jon_avila> can we put a note about the term pop telling people we will be coming up with a different term?

<JF> +1 to attempting to wordsmith on-list, and revisit on Thursday

<allanj> +1 in now

<steverep> Would anyone to object to a CFC now? I can add an editor's note.

<Kathy> can we update the definitions

awk: editor's draft, requires cfc. or take it up thursday. should be able to get in w/o lots of extra tiem by 22nd.

<Detlev> +1 include with some wordsmithing and a good def

<david-macdonald> +1 in now

<marcjohlic> +1 to add now - let's put a stake in the ground on a term and add it..

<Joshue108> +1 in now

<KimD> I'd be ok with putting in as is, with note as Jon said

awk: no consensus.
... let's do wordsmithing so we can wrap up thursday.

<Kathy> the definitions need to be updated

steve: when can we put in stake? other than popup? nothing out there matches it better.

<kirkwood> +1 to put in editor draft define term popup

jf: so long as have clear definition okay. must define popup better.

<david-macdonald> "transient popup" distinguishes it from triggered popup on ENTER

<Detlev> agree with speaker before that popup is used by browsers (blocker) and shouldn't be used

<jon_avila> +1 to Michael

<allanj> definition: content which becomes visible only when associated content, called the trigger, gains keyboard focus or pointer hover

<Glenda> +1 put it in now with editor note about definition. Stop taking up working group time on this one.

<Joshue108> +1 to Michael so I'd accept hoverbox as a more semantically accurate term

<Detlev> ok +1 to Michael

mc: think about timelien. throw it in for path. in past have put notes that will be working on it. put in now. ppl can work on it. not take more time.

<Glenda> Public Working draft is not a FINAL

<jon_avila> I didn't hear that we were still talking about what's in and out

<allanj> perhaps we need a wider view of what's in/out

alex: still talking about what is in and out. far cry from saying "let's move forward".

<jon_avila> I formerly disagree with Alex

<Glenda> Public Working Draft is to get public feedback

steve: what is in and what is out?

<jon_avila> deferring it will kill it.

steve: have to be clear what we are asking people to do

<JF> +1 to Alex

<Zakim> bruce_bailey, you wanted to ask JF where definition fails?

bb: not undestanding what alex and john f can't live with. popup poor, not perfect, but what can't you live with.

<alastairc> https://rawgit.com/w3c/wcag21/popup-interference_ISSUE-75/guidelines/#dfn-popup

jf: don't see popup definition.

<allanj> definition https://rawgit.com/w3c/wcag21/popup-interference_ISSUE-75/guidelines/terms/21/popup.html

<alastairc> content which becomes visible only when associated content, called the trigger, gains keyboard focus or pointer hover

bb: it's in draft.

jf: sorry not seeing it in actual draft.

<Joshue108> https://rawgit.com/w3c/wcag21/popup-interference_ISSUE-75/guidelines/terms/21/popup.html

bb: look into github.
... works as is.

steve: posted question about it this morning.

<Zakim> JF, you wanted to ask why we cannot continue to fine-tune this over the next 48 hours, and revisit on thursday

alex: trigger point, if you have very large trigger, how is it a trigger.

jf: we're discussing for 20 minutes plus. still needs further discussion. don't see risk for letting it simmer for 48 hours.

<AWK> +1 JF point on working for 48 hours

<jon_avila> would be open to allow hover to cover trigger area that doesn't contain content. That is it's ok to cover empty space of trigger

<alastairc> Suggested def: "An element that appears when the mouse is placed over a triggering element, or the keyboard focus is on the triggering element."

jo: fine with that. have you been on call from last week?
... agree with michael c. my conclusion is that we have reasonable definition, term not perfect, but would like to see it go in. in case of use case went over it quite well with mouseover.

jf: burned a lot of time haven't reached a resolution. two more days is reasonable.

awk: waiting another couple of days okay, willing to look at definitions. as long as people can look at before hand.

<david-macdonald> I agree that there are some situations where we don't want to forbid the popup from overlapping the trigger

awk: some valid concerns: tooltip covering large area.
... let's work on this, take it up on Thursday.
... this, personalization, time-outs on Thursday. Please look at CFCs.

jf: moving to two hours on thursday.

trackbot, end meeting

Summary of Action Items

Summary of Resolutions

  1. Extend the calls starting Thursday August 10. Split scribing for long call.
  2. Accept as amended.
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/08/08 16:35:41 $

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/alex: is that/Steve: is that/
Succeeded: s/alex: popup is generic/steve: popup is generic/
Succeeded: s/alex: only if/steve: only if/
Succeeded: s/alex:not/steve:not/
Succeeded: s/alex: yes in general/steve: yes in general/
Succeeded: s/alex: what is in and what is out?/steve: what is in and what is out?/
Default Present: AWK, Rachael, Kathy, Joshue108, JF, shadi, MikeGower, david-macdonald, KimD, alastairc, Elledge, Makoto, Detlev, kirkwood, jasonjgw, JakeAbma, steverep, Melanie_Philipp, Glenda, bruce-bailey, Avneesh, jon_avila
Present: AWK Rachael Kathy Joshue108 JF shadi MikeGower david-macdonald KimD alastairc Elledge Makoto Detlev kirkwood jasonjgw JakeAbma steverep Melanie_Philipp Glenda bruce-bailey Avneesh jon_avila Mike
Found Scribe: Mike_Elledge
Inferring ScribeNick: Mike_Elledge
Found Date: 08 Aug 2017
Guessing minutes URL: http://www.w3.org/2017/08/08-ag-minutes.html
People with action items: 

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


[End of scribe.perl diagnostic output]