W3C

- DRAFT -

Accessibility Guidelines Working Group Teleconference

28 May 2019

Attendees

Present
AWK, Chuck, Detlev, JF, JakeAbma, KimD, Laura, Lauriat, MichaelC, Rachael, alastairc, david-macdonald, johnkirkwood, jon_avila, mbgower, stevelee, Raf, shadi
Regrets
Fazio
Chair
SV_MEETING_CHAIR
Scribe
JakeAbma, Chuck

Contents


<AWK> +AWK

<JakeAbma> scribe: JakeAbma

TPAC registration https://www.w3.org/2019/09/TPAC/

AWK: early bird till 21st of June

Silver update – Migration plan overview

<KimD> KimD

Shawn: working on migrating content

<jeanne> https://www.w3.org/WAI/GL/task-forces/silver/wiki/WCAG_to_Silver_Migration_Map

Shawn: running through TOC
... been through all SC
... done some grouping, then identify user needs
... grouping may shift around a bit
... once user needs are identified, write tests, then platform specifics
... link is first pass of that
... we will ask some specifics to specialists for better understanding
... grouping results ready, not final though, some are repeated

Jeanne: had informal meeting at AccessU
... we wrote template to how to write content
... so make it easier for people to start writing
... what we did was: making template, users needs, identifying test, writing method, then roughly understanding like text and in the end, the Guideline

<jeanne> Content Writing Process https://docs.google.com/document/d/152RdsWpzq0QqdiSeW2dk0-RTp6QZkbcOZevbfa0tQMo/edit#heading=h.2ofg9ub2gkrq

Jeanne: this is the content writing process

<david-macdonald> can you open the doc to public, can't access

<stevelee> need access, thanks

Jeanne: need to make new copy, this one isn't working

<Lauriat> Copy viewable by all: https://docs.google.com/document/d/1eAqHf2dk-qhyk3YtwfHV0uIIrCejwCnlZnPuFzVhYx4/edit

Jeanne: we provide some instructions on how to write
... first map current techniques and ACT rules, identify needs
... see how existing tests can be included
... want to do some new type tests to see if the tests are OK, and after that the methods
... writing the methods will take up much more work
... at Step: 4 we're busy with, step 5 needs some more work soon
... next time we'll have content to show you!

<jeanne> Google drive folder with content drafts https://drive.google.com/drive/folders/15u5P_VwQUu--1NXSzu2Wwsj-QalDCt_W

stevelee: in what format is the template presented in?

Jeanne: at the moment in GOogle Docs, for abbility of comments
... all formats may be possible
... ow, you mean editors draft?
... we should make it publicly available
... we can go offline for discussing on how to work with COGA in formats they prefer

WCAG 2.1 techniques update: https://docs.google.com/spreadsheets/d/15idlBl1qQTNr2SIi4Drzk1Q1vnWAi5L26GCCv6mjD2g/edit#gid=0

ACT Rules updates proposal: https://github.com/w3c/wcag-act/issues/353#issue-428734105

Wilco: we've published CR a month ago

<Wilco> https://www.w3.org/TR/act-rules-format/

Wilco: today is final day for public review
... got some good feedback

<shadi> https://act-rules.github.io/rules/

Wilco: new name for community group: ACT R
... in contact with some companies to implement the rules
... couple of placed we see them embedded alongside the techniques

<Wilco> https://github.com/w3c/wcag-act/issues/353#issue-428734105

Wilco: also working on proposal for plan how process might look like, people / companies writing ACT rules
... broken down activities for embedding
... once we have enough implementation, the rule can be submitted
... got Rule support, Rule submission, Rule maintenance and Publication Approval
... we're pretty much ready

Shadi: we're incubating the rules in the community group
... discussing and agreeing on the rules
... ultimately AG group approval to authorize
... we need to still define quality requirements
... example is if ARIA rules comply with WCAG, and where do we agree on this

Cooper: we don't charter task forces, what we do need to know is deliverables

AWK: we're looking for feedback about: a: WG will publish the rules?
... if yes, is the process outlined the right one?
... will the WG be involved?
... ACT TF is specialized within the larger WG, works well in that regard
... worry about the rules getting too far
... might be the WG sees an issue in first part of rules while the TF already set second / trir

third part of the rule

Wilco: need mechanism for this

AWK: we need to strike right balance

Shadi: some rules might end up as "good practice" not specially full conformance
... idea to have monthly or quarterly heads up

<shadi> https://act-rules.github.io/rules/5f99a7

AWK: how is it looking right before CR?

Wilco: had 6/7 commenters, mostly superficial comments

Shadi: several people do implementations, looks good

MG: don't take it too far before we get a bit more involved

WCAG 2.1 techniques update: https://docs.google.com/spreadsheets/d/15idlBl1qQTNr2SIi4Drzk1Q1vnWAi5L26GCCv6mjD2g/edit#gid=0

<Chuck> Scribe: Chuck

Andrew: Great scribe work!
... WCAG 2.1 techniques and where we are at. We have accepted some since we started monitoring.
... The status is still largely where it is. The main question is are there any... anybody taken up new technique for success or failure?

<no response>

Andrew: With this we do need people to take up some.
... Looking over one in particular, one that's orientation we could use another for content on hover or focus.

<JakeAbma> just finished a new SC, will do doorslam coming week

Andrew: We could use another for character key shortcuts, there's one drafted for each success and failure.
... Pointer gestures will be discussed today.
... there are a few in motion now. Is there anyone on the call that has time or interest and doesn't know what to work on?

DM: I've got 4...
... I should be able to do one a week.

Andrew: What SC are those?

DM: 1.4.12 failure, 1.4.11, 2.5.2 failure.
... I've got 2 in the queue.
... 1.4.12 and 1.4.11.

<Zakim> JF, you wanted to sheepishly admit that I have a task to create another Failure Technique for SC 1.3.5 (that essentially shows/explains that the metadata is applied *only* to user

JF: I've been dragging this one around. I've an action to create failure technique for 1.3.5
... Metadata applies only to user and not names and addresses, etc. Show and tell technique.
... Only the name input directly related to the user applies.
... I owe that. Not sure if you are tracking it. Add it and assign it to me.

Andrew: It's a failure ... form that asks for your name and your spouses name or child's name. It shouldn't apply the attribute to the spouse or child.

JF: Essentially correct. That's one use case. booking airline tickets for the entire family. There may be different family names.
... Another use case... HR form with names of all employees in office. Your name, but all other names would be different.
... Autocomplete shouldn't apply there.
... Auto-complete, where the data that's going to be captured relates directly to the end user. Lot's of confusion around this.
... I need to write up a technique, may be a success technique may be a failure technique. I have to show code examples.

Andrew: Is Marc on the call?

mbgower: No.

Andrew: Was going to ask about news for row 12.
... I added it, does this look good?

JF: Yes, this describes it well enough.

Jake: Couple of q. I have time to start some techniques this coming week. I see one for....
... At reflow I see approvals, while those techniques for reflow are already published I think you need to update those.

Andrew: The first worksheet may not ... be up to date. Feel free to write published.

Jake: Line 30 and 38 are published already. I see some other ones AAA for target size. Which one is more important?
... AAA or M/yellow?
... Row 124 or 27...
... Not sure while we have failure and sufficient while they are opposite, but we can make them. I can start with 124. or
... I can start with #5.

Andrew: I would rather you see #5.
... Given that the... column f priority is a bit confusing. For orientation that one is high priority as a SC as oppposed to target size
... Which is medium, while it's in a state of complete incompletion, it's a lower priority than the AA one.

Jake: I'll do #5. #30 is published also.

Andrew: I think I had that as approval. I didn't know if it was published or if it was approved but not yet published. Haven't looked yet.
... Your telling me it's published I can change it.

Jake: Not sure...

Andrew: If wg said yes it's good, but not yet pushed out on the server, it's not published yet.

<JakeAbma> https://www.w3.org/WAI/WCAG21/working-examples/css-fitting-images/

<JakeAbma> https://www.w3.org/WAI/WCAG21/Techniques/css/C37#examples

Jake: This is the one <above>
... Seems published to me.

Andrew: Changed!

<Zakim> mbgower, you wanted to say the spreadsheet isn't totally up to date?

MG: The discussion that just happened is what's happening to me to. Even when working on editors draft, looks aug 18.
... did we lose continuity in our version control?
... They aren't showing up in editor's draft or published. My comment is it would be good to understand how frequently these happened...
... And things were adopted.

Andrew: Label and name?

MG: Yep.

Andrew: I've got on here overview tab 3 label and name techniques. I'll go through and check where things are at.
... I'll make sure this is accurate.

<MichaelC> http://w3c.github.io/wcag/understanding/ shows updated 22 May 2019

Andrew: And MC, as far as date on editors draft, we need to update the date in the master branch, is that right?

<MichaelC> https://www.w3.org/WAI/WCAG21/Understanding/

MC: Editor's draft says <date>, it's possible that date updating isn't working but it is, not sure what you are looking at.

MG: May be a caching issue, but came up that way on 2 different computers.

Andrew: If you still see that and don't see current date ping us, we'll sort it out.

MG: w3c.github.io...

MC: slash understanding...?

MG: <mentions different url>

MC: You are looking at older version. If you look under understaning that's more current, date is on cover page.

MG: <url> the 3rd link is latest editors draft. Which is the standard everyone would go through.

MC: Last time we updated guidelines, not understanding.

MG: I'm not the only person that will follow this path. No obvious way to get there.

MC: You did get there.

MG: OK.

Andrew: You click on 1.4.10 reflow, then click on understanding reflow, that is taking you into the w3.gi... <url>
... Which that one is may 22nd.

MC: I know that a person not familure wouldn't know, but that IS the most recent version.
... We can discuss things about putting dates on every understanding page, not constrained on structure. We can do that if helpful.
... But you are getting to the latest info.

MG: OK, links to techniques seem out of date.

Andrew: May be that the techniques were accepted and that we forgot to update the references?

MG: I made a PR for that. I guess I'm saying it gets messy. Now I understand why guildelines didn't make sense to me.

Andrew: We'll get that updated. Let's move on. Get to survey, will help us progress on techniques.

ACT Rules updates proposal: https://github.com/w3c/wcag-act/issues/353#issue-428734105

New techniques & issues survey: https://www.w3.org/2002/09/wbs/35422/Tech_Und_survey/results

Andrew: On this.... <NO!>
... We haven't had a ton of people responding.
... First question.

Pointers gestures clarification path base gestures.

Andrew: From last week we had discussion around what is covered by the concept of path based gestures.
... Which comes into play with regards to pointer gestures SC. We wanted to have a survey so folk could weigh in.
... We've got some feedback from people. We are trying to get to a resolution on whether... around slider.
... Is a slider where you use the mouse to move the thumb around, does this violate the sc or not?
... Some came back with "maybe", some came back with difinitive answers.
... Patrick Luck prefers option 3. For this pointer gesture sc that the only thing that would be acceptable without an alternative
... is a pointer action that actually is free form, hand writing for example.
... That seems to be what Patrick and Detlev and Jake are behind.
... Then MG and myself both responded to the survey with a different opinion.
... Anyone who's ... Detlev on the call?

MG: He couldn't make it today.

Andrew: So Detlev wrote the proposal, he definitely supports only freeform dragging or drawing is excluded.
... If you need to drag an item through a maze shaped path for some reason (game or activity) where it's essential would be ok.
... Anything else would not be ok. On the call... slider such that it only matters where you start and end the dragging.
... Doesn't matter what path you take... that's why we have the survey.
... Anybody else want to weigh in?

<AWK> Success Criterion 2.5.1 Pointer Gestures (Level A): All functionality that uses multipoint or path-based gestures for operation can be operated with a single pointer without a path-based gesture, unless a multipoint or path-based gesture is essential.

Jake: I have a q. For you or MG. Why you do not want #3 included? It almost seems like we have a AA sc for #1 and #2,
... And an AAA for #3. Why do you or MG want to leave out the one for #3? I don't understand the reason for this moment.
... I think Detlev also does not understand.

Andrew: What do you mean leave it out?

Jake: Why not go all the way except for the signature or other freeform? How can we make this clear to users of WCAG where the line is?
... And the reason why it stops at some point. Where you and MG draws the line, or Detlev and myself draw the line?
... What's your reason?

MG: First of all, the actuall wording says path base gestures. We had an official response in the wg by an outside agency.
... Path base gestures would have same definition as keyboard, we didn't do that. We voted to do that, never happened.
... If we are not following that, we aren't...
... If you want to redefine path based gestures the way it is defined in 2.1.1 keyboard.
... If the wg wants to do that, we are on shaky ground.
... The pointer gestures is talking about single point usage. We already have this wierd conflict going on.
... We do have clear guidance on what we were going to say before.

<JF> +1 to Mike - we did agree to that quite some time ago

Andrew: Jumping in, agree completely with MG. There is a clear point that we need to do a better job of what that is.
... That it is related to just start and end points or if it's the path inbetween.
... One of the things that's a concern to me, it seems to be about if sliders are an issue or not. It should not be aboug...
... Any given slider, it's about the implementation of the slider. Is it implemented in a way that meets the sc or not?
... I see JF agreeing with Mike. Anyone else in queue?
... I'm trying to find the link you sent MG.

MG: I can find fast. One sec.

DM: Jumping in.
... So a bit of history on path based gestures. It really was about drawing programs and couldn't replicate with keyboard.
... Everything had to be operatable with keyboard except for drawing program, keyboard. Could have any # of middle points.

<mbgower> https://github.com/w3c/wcag21/issues/478

DM: Otherwise we would have done an etch-a-sketch which would have been difficult. That's where it came from.

Andrew: Thanks.
... MG has a link to issue 478. And ...
... This is where we talked about changing single untimed pointer gesture to untimed single pointer gesture.

DM: From Nov 2017. It's a little while ago. Important point is last part of response.

Andrew: Yep.
... We said there would be a new definition.
... We published in June but we did not do that.
... So... we can't just add it in now. We could add in 2.2.

<mbgower> "user's movement" to me is at the crux of this discussion

Andrew: We don't have that definition in. Q for other people in the call. Do we feel this provides sufficient history about where we've been?
... To convince people that we don't want to take option #3?

Jake: Is the definition is not there, how can we make a decision? It will not be more clear. You will still have people on both sides.

Andrew: We have to make some sort of decision?

Jake: Postponed?
... Definition is a little awkward. Because of the fact that the definition was not there, maybe we ended up with this discussion.
... For me I've never seen this one before. So you can make a decision without the definition. But I don't think people
... will pick it up automatically.

MG: You don't think the rewriting covers this?

Jake: I think that NOW we lack the definition. There's some awkwardness with single pointer where it's single pointer gesture.
... where it contains "path based" in it where it's not defined.
... From a holistic approach. If we have this definition, it might fit in your proposal. Maybe if and when it's there.
... It may be oppposed to the sc. We may not cover the intent of these. This feels not correct. I can imagine it's hard to explain.
... This slider is about the end points. The intent is to help people to have an alternative with a motion or gesture.
... We still don't help people out with sliders.

<Zakim> mbgower, you wanted to say that having to make a specific motion or gesture is included

Jake: Is that clear? I think you are right if we have this definition. We still exclude some custom widgets and components we would like to cover.

Andrew: MG...

MG: For people who require specific motion or gesture IS included.
... Originally the wording Detlev had is he excluded drag and drop. Gestures that require direction...
... Are defined as what path based is.
... Now making clear where direction is involved. Those things are included. Even the slider is included if the definition requires specific motion along the x axis.

Andrew: And I think Jake that we can make a decision about what the WG feels the text says.
... It's helpful if we have a definition we approved. but we have lots of things that don't have definitions that we still have thoughts about.
... In this case we've got... a response back to a question which clarifies what we were thinking about even though we don't have a definition.

<AWK> Success Criterion 2.5.1 Pointer Gestures (Level A): All functionality that uses multipoint or path-based gestures for operation can be operated with a single pointer without a path-based gesture, unless a multipoint or path-based gesture is essential.

Andrew: It would be good to clarify today if we can if people feel like in this sc that path based means just related to end points or not.
... Here's sc text <reads>.
... Given what we talked about, are there people who feel that this path based gesture here....
... Are there people who feel the wg's intent to only address free form gestures or whether it was to allow gestures where the endpoints are the only points that matter.

DM: I think we intended to include swipe, that's a path based gesture, a particular path to make it work. Custom swipes is what we are trying to deal with.

<mbgower> +1 to David's comment. Custom swipes.

DM: Making sure people with shaky hands could do that functionality.

<mbgower> https://github.com/w3c/wcag21/issues/61

Andrew: OK, MG agrees.

MG: I pasted in the original issue back when the original wording was there. This comes down to any form of dragging should be included.
... If it's included then the conversation can stop, if not then we need to talk about what dragging we don't want to have included.
... How many people think all forms of dragging are included?.

DM: Drag-and-drop examples?

MG: Any .....

DM: You'd have to have a pointer you could fall back for.

MG: You'd have to have a single click operation. Using no movement.

<AWK> Question: Was the WG intent with 2.5.1 to allow any gesture as long as the result of the gesture only depends on the start and end points?

Andrew: Asking question just related to start and end points? <reads>
... Which means drag and drop is ok as long as you didn't have to follow a specific path when following it. Just start and end points.

DM: I see what you're saying.
... I don't think that was the intent but it sounds interesting.

Andrew: You don't think that was the intent?

<mbgower> "Was the WG intent with 2.5.1 to allow any gesture as long as the result of the gesture only depends on the start and end points?" Yes, and we have multiple votes backing that up.

<AWK> +1 to that being the intent

DM: I don't remember us discussing this. but I think this is an interesting thought.

Andrew: OK. David says no, MG says yes.
... This does not mean that you think the sc is perfect or that there isn't more to be done. It's about what we were trying to do.
... Most people are not saying anything. Not sure? Don't care? Aren't following?

Jake: Just one... what I'm trying to understand, besides the definition... but if say we do not cover drag and drop or some other
... forms which do not align with the sc. Then we end up with a version where certain user interactions will not be possible for certain people...
... With motor challenges, or at that isn't able to perform the movement. If we do not demand from the guidelines to have a single point alternative...
... Then we don't have coverage for these people. Why do we want to exclude those movements or those cases right between...
... The specific path based or the end points? Why would we leave that part open?
... Because it doesn't follow the normative text we don't have? Or is there another reason?

MG: We already had a vote on drag-and-drop being out of scope. We had a vote on an official vote on a response.
... We had this discussion a lot of times. If we could go back in time and say path based is not in the language then we could have that discussion.
... There's multiple comments and issues that have had comments and discussions, not a new discussion.

Jake: At least to not have the enormous elaborated understanding doc. Cover what we discussed in the past.
... Took a while to read the comments. Maybe discuss it for 2.2 AA or AAA sc. To cover those other cases.
... Those other cases will still be an accessibility fail?

<Zakim> JF, you wanted to ask if we can add some exceptions here?

Andrew: MG proposed for 2.2.

JF: For the most part I agree with Mike. Lot's of discussion. I hear Jake's concerns as well.
... The editorial and understanding has a lot of words but no simple examples. Is there any value in us looking at the understanding doc...
... And adding a table about what's in scope and not in scope?

<JakeAbma> https://github.com/w3c/wcag/issues/744

JF: If I'm to understand your concern, there's a lack of clarity about what's in scope. Maybe we can clean up the understanding doc with a table.

Jake: I added an issue, I'll make a proposal. You can read about it in issue 744.

Andrew: The pull request we have here are updating understanding content to clear things up.
... This first question... the original pull request, option 2 was doing something similar but different, option 3 was more substantial.
... I think that what you are saying Jake is true. We are trying to get clarity, but we aren't in agreement on what clarity means.
... MG pointed out what the wg was thinking in the past. Maybe current group can confirm if we are just talking about the end points.
... Then we can zero in on specific clarity.
... Does that make sense?

JF: I think you said it better, if that's the case I agree.

Andrew: I think so.
... So do people agree that the original intent of the wg was to interpret path based gestures as relying just on start and end points?

<laura> A Proposed WG answer for 2.5.1 Pointer Gestures from Wiki: Path-based gestures are any gestures where the user engages with a screen (e.g. touches or causes a mouse-down), moves the pointer, and then releases the finger or mouse at another location. The WG will still determine whether we need a definition.

<laura> https://www.w3.org/WAI/GL/wiki/Draft_Response_issue_721#2.5.1_Pointer_Gestures

Andrew: Or anyone disagree?

DM: re-ask?

Andrew: write it down.

DM: Swipe to right for instance that doesn't have an end point if you start from a point and go in a direction. Would that be in scope?

JF: That feels like a keypress operation. It's not but it feels that way.

<AWK> Does anyone disagree that the original intent of the WG was that path-based gestures meant that only the start and end points mattered in a gesture?

DM: Yes. We have swipe in the language.

<mbgower> A swipe is direction reliant and therefore path-based

DM: What are people's thoughts?

Andrew: David if you have a swipe where in iOs there's a mail app, you can swipe to get rid of an email from the list of emails.
... If you can start the swipe and move the mouse all over, and if you end up at the rigth spot that wouldn't count.
... But if you have to stay in the same horizontal band, you have to start on the left hand, you have to move right,
... It's not just relying on start and end points it would apply.

<Zakim> mbgower, you wanted to say I have never heard anyone suggest that swipe is not in scope. Swipe, by definition, is directional

DM: So swipes are in scope.

MG: No one ever suggested that about swipes.

DM: Drag and drop would not be in scope because it does not rely on the path.

MG: Exactly.

DM: I'm on same page.

Andrew: Does anyone disagree <reads>?

DM: one last thing.
... In case where there isn't a start and end, there's just a direction, then that's still in scope?
... A swipe doesn't have a start and end point. That's not getting excluded?

Andrew: I'm having a hard time hearing you. Swipe is still covered.

MG: The way Alastair defined it, swipe starts with a press movement and a release. Which means it does have a start and end point.

DM: I'm thinking with coordinates for start and end points. Everything has a start and end point.

MG: I'm trying to find language that Alastair put in.

Chuck: Time check. Near end.

MG: The language Alastair has.

<mbgower> https://github.com/w3c/wcag/pull/714/files

Chuck: Paste it in.

Andrew: Getting down to end of call.

DM: It works. Where's it on page?

MG: First green line. First...
... I'll read out line. Line 14 in replacement. <reads>

<AWK> A <strong>path-based gesture</strong> involves an interaction where the user engages a pointer with the display (down event), carries out a directional movement in a pre-determined direction before disengaging the pointer (up event). The direction, speed, and also the delta between start and end point may each be evaluated to determine

Andrew: From what I've heard, I feel like we can resolve this that it doesn't include gestures with....
... We can decide on which pull request to use, but we have come to a conclusion.

RESOLUTION: The WG intended that path-based gestures in 2.5.1 does not include gestures that only depend on the start and end points.

trackbot, end meeting

Summary of Action Items

Summary of Resolutions

  1. The WG intended that path-based gestures in 2.5.1 does not include gestures that only depend on the start and end points.
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2019/05/28 17:03:49 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
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/Shidi//
Succeeded: s/<someone>/mbgower/
Succeeded: s/MC/MG/
Default Present: AWK, alastairc, jon_avila, Fazio, MichaelC, Chuck, Laura, Detlev, david-macdonald, stevelee, johnkirkwood, KimD, mbgower, JakeAbma, Rachael, JF, Lauriat, Raf, shadi
Present: AWK Chuck Detlev JF JakeAbma KimD Laura Lauriat MichaelC Rachael alastairc david-macdonald johnkirkwood jon_avila mbgower stevelee Raf shadi
Regrets: Fazio
Found Scribe: JakeAbma
Inferring ScribeNick: JakeAbma
Found Scribe: Chuck
Inferring ScribeNick: Chuck
Scribes: JakeAbma, Chuck
ScribeNicks: JakeAbma, Chuck

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 28 May 2019
People with action items: 

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


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]