W3C

– DRAFT –
AGWG-2021-06-01

01 June 2021

Attendees

Present
.5, AWK, Azlan, Ben, ChrisLoiselle, Chuck, david-macdonald, Detlev, Fazio, Francis_Storr, GN, GN015, JakeAbma, jeanne, Jemma, Jennie, JF, jg, johnkirkwood, juliette_alexandria, JustineP, KarenHerr, Katie_Haritos-Shea, KimD, Laura_Carlson, Lauriat, MelanieP, mgarrish, MichaelC, morr, morr4, OliverK, Rain, sajkaj, sarahhorton, ToddLibby, Wilco
Regrets
Bruce Bailey, Jake Abma, Makoto Ueki, Steve Lee
Chair
Chuck
Scribe
ChrisLoiselle, mbgower

Meeting minutes

Chuck: We have a scribe for first hour

Chuck: We haven't turned on transcripts, so we proceed without for this call. We will turn on next meeting and apologize .
… Jake talked to WCAG 2. maintenance and we will talking that in second hour.
… do we have any new topics ?

Silver heartbeat publication

Chuck: discussion on Silver heartbeat.
… explainer is under development.

<Zakim> jeanne, you wanted to answer Melanie

Melanie: Are you adding any editors note on status and what is a heartbeat publication ? What is covered and that it doesn't cover all comments raised in FPWD.

Jeanne: This is part of status of document . I have seen a draft and it covers the points you raised.

Jeanne: It states it is a working draft, and that we have not addressed all comments, and that we are working on those comments as well as others, on the next draft.

<MichaelC> This is a Working Draft of W3C Accessibility Guidelines (WCAG) 3.0 by the Accessibility Guidelines Working Group together with the Silver Task Force and Silver Community Group. WCAG 3 was published on 21 January 2021. This version moves much of the introduction to the Explainer for W3C Accessibility Guidelines (WCAG) 3.0, adds a new proposed guideline, error prevention, and addresses editorial fixes from comments received on that draft. Not all comments

<MichaelC> on the previous draft have been processed yet. In particular, testing and conformance have received many comments which are being actively explored, but the group has not yet adopted updated content for those sections. The group will continue processing comments from the previous draft as well as on this draft.

Melanie: Thank you.

<AWK> +AWK

MichaelC: places in IRC the text for the status update text.

Jeanne: Reads off the text MichaelC pasted in to IRC.

Chuck: Any other questions on this topic?

AGWG member participation in the errors subgroup

Chuck: Call to participation in errors subgroup to advance the work, (same issue is for headings content group as well).

If you are interested, please let the chairs or task force leaders know.

Error prevention https://www.w3.org/2002/09/wbs/35422/wcag3-error-prevention/

<Chuck> https://www.w3.org/2002/09/wbs/35422/wcag3-error-prevention/results

Include Error Prevention in the Heartbeat publication

Chuck: Error prevention survey responses around WCAG 3 work.

Chuck: Editors wish to include error prevention in heartbeat publication. Talks to responses of survey.

<JF> +1 to AWK

AWK: The evaluation of criterion, methods and outcomes, is difficult. It is an active effort we are all working on. We should focus on this first, before adding more guidelines.

Chuck: I recall that we spoke to this prior in a conversation, regarding guidelines and conformance model. I believe the conversation was around craft guidelines to help test out conformance model.
… more data on where conformance model is falling short is the key, and more guidelines would help us understand with the data points. Both simultaneously being worked on was a key take away.

<Zakim> MichaelC, you wanted to ask if immature conformance makes this gl worse than the other ones already in the draft and to say we need gl to show the plans in context, especially with what´s in flux

MichaelC: I think we need some guidelines on the document to be able to see the structure and context , to be able to see we can offer guidance within this structure.

<Zakim> jeanne, you wanted to AWK

MichaelC: question on this guideline is whether or not we include it on this publication.

<Wilco> +1 to Jeanne

Jeanne: We have tried working on just guidelines and just conformance. From a project management stance, we need to move both together , as they validate each other. We need to migrate existing wcag 2.2 to silver to show that conformance works.

<Zakim> AWK, you wanted to say that additional guidelines can be in the editor's draft, but shouldn't be in the heartbeat

AWK: I get that. I don't think it is at a point of maturity to say it should be included in a public working draft. Issues were raised on conformance model, but not a lot has been made on addressing those. Our published work should showcase progress around issues raised.

<JF> +1 to Wilco

Wilco: We have editor's drafts. When we publish a new draft, it should have new answers. Public should have another review.

<Zakim> MichaelC, you wanted to say heartbeats are part of not getting shut down

Chuck: My thought was that even though it wasn't an editor's draft version, it was a chance to get feedback on conformance model. I don't think I hear objection to add errors to heartbeat, but not in a heartbeat.

MichaelC: I think publishing heartbeats is important. What is in them for content, is less critical. Progress is important to showcase.

<Wilco> why can editors drafts not show progress?

MichaelC: drafts are drafts and may change due to being a draft. Better to showcase progress than not.

JF: Plus 1 to AWK and Wilco. It is introducing another scoring feature, of credits. That is a concern.

<KimD> +1 to Michael and Jeanne

JF: To AWK's point, migration to WCAG 3 and conformance model with not knowing full picture, leads to issue around timing .

Chuck: To JF's point, better set of activity is to migrate wcag 2.2 content into the structure of work?

JF: I think it is a requirement we want to include in the specification, but I'm struggling how to score it. What is a credit, and how that impacts final score? How do we put that under same structure within current conformance model , etc.

<Zakim> jeanne, you wanted to say that we also need to include new material to see that Conformance works

JF: Perhaps migration of current SCs is simpler approach.

<Zakim> MichaelC, you wanted to say having a singular focus for the heartbeat has value

<JF> The question is, will the conformance model support the *existing* requirements

MichaelC: A primary reason to publish the heartbeat is important, I agree with JF's point.

SarahH: This isn't a new requirement, it is a reframing of a requirement, i.e. labels and instructions. It is grouped in a different guideline, speaking specifically to error prevention. The requirement is very much based on labels and instructions.
… it builds on it , i.e. grey areas of placeholder text and does it qualify as a label or instruction. The method attempts to improve the grey area of placeholder and testing within new methodology. This is a migration project, with enhancements.
… I can speak to the scoring. This is a prototype of using credits.

<laura> Wondering if we could publish without the Scoring piece?

Chuck: Talks to LauraC's comment in response of survey.

<Zakim> MichaelC, you wanted to deconfound strategic and content reasons

Chuck: Laura agrees with AWK's comment.

MichaelC: Mix of strategic and content reasons on publishing errors in heartbeat...do we have a priority of content or strategy?

Chuck: I agree with your comments raised.

Wilco: What is the advantage of a heartbeat over the editors draft?

<laura> Wondering if we could publish errors without the errors scoring piece?

MichaelC: Format matters to some.

<Chuck> proposed poll: Is our priority to focus on "strategic" or focus on "content"?

Jeanne: Talks to Laura's comments in IRC around publishing errors without scoring?

<JF> yes

Chuck: Is our priority to focus on strategic or focus on content?

<Chuck> content

<sajkaj> Define "strategic?"

<jeanne> Strategic means publishing content. We have to make the project move forward.

Chuck: Can you place in either content or strategic in your IRC response?

<Ben> content - as it will hopefully bring in more comments from the public

MichaelC: Content = actual things that people might review, whether it is mature or not. Strategy = longevity of project, organizational progress being made.

KatieH: I believe strategy is what silver is working on. Whether or not we have scoring , errors is important and should be talked about. A lot is missing from WCAG 2 and we want to handle those pieces in WCAG 3. Strategy in my mind, is clear conformance model.

Chuck: We are reviewing this guidance for this particular heartbeat, we aren't losing any of the work by not including it in the current heartbeat.

<Zakim> JF, you wanted to ask whether this new requirement could also fit into the existing WCAZG 2.x structure?

JF: I think both Sarah and Katie make very interesting points. We want to include it in big picture. Could this work also be included in a WCAG 2.x structure?

<Zakim> jeanne, you wanted to say that the long term strategy is publish content as it is developed

<Fazio> I've been wondering that too lately JF

JF: I think that would advance the work , but not under the WCAG 3 framework, which has the conformance questions around scoring.

Jeanne: Dividing into strategic and content is not the way to look at this. Publishing shows progress. It would be helpful to be included, so we can keep working drafts and editors draft in sync.

Jeanne: I feel that whatever work gets done, we need to have it included. We need work in guidelines and working WCAG 2.2 over to WCAG 3 framework.

Jeanne: Please let us move forward and receive feedback.

<jeanne> https://w3c.github.io/silver/guidelines/#error-prevention

<Jemma> https://w3c.github.io/silver/guidelines/

Laura: Can't unmute, we will get back to her when we can.

Chuck: Talks to DavidMcD's comments: This is a global issues: We have at least 7 categories of Cognitive functionality. Have we provided the same level of granularity for the other categories of disability.
… There are multiple types of vision loss, multiple types of color blindness, and dexterity disabilities. I think we have to review the functional categories and provide a comparable level of granularity for each category.

<Zakim> jeanne, you wanted to answer DmD

There might be some more critical errors to provide such as: -Failure to provide error messages in tex

The plain language description of atomic tests says "like the way we test today" And then for the holistic tests they say "test with a screen reader"

I think that most professional testers today are constantly checking things with various screen readers, at least I am. If we are going to add "like we do today" to atomic and think we'd have to say the same thing in the holistic tests regarding checking with AT.

Jeanne: There are 5 top level categories, essential , sensory have 15 needs, etc.

Jeanne: I don't think it will be able to be done the way David is proposing to address all users with disabilities.

SarahH: This is error prevention, so before user is making an error. This is related but separate from error messaging and notification.

<jeanne> Statistics on Functional Needs:

<jeanne> Essential - 1

<jeanne> Sensory - 15

<jeanne> Vision & Visual - 8

<jeanne> Hearing & Auditory - 4

<jeanne> Sensory Intersectional - 3

<jeanne> Physical - 14

<jeanne> Mobility - 3

<jeanne> Motor - 6

<jeanne> Intersections - 3

<jeanne> Speech -2

<jeanne> Cognitive - 23

<jeanne> Attention - 3

<jeanne> Language & Communication - 8

<jeanne> Language - 3

<jeanne> Memory - 3

<jeanne> Executive - 3

<jeanne> Mental Health - 1

<jeanne> Cognitive & Sensory - 2

<jeanne> Independence - 2

Melanie: I agree with DavidMcD. on cognitive areas. Voicing support of David.

Chuck: Talks to editorial changes on Stefan's comments

Chuck: Talks to Oliver's comments, more editorial of nature.

Oliver: Listing of examples , if lists should be complete, or just examples was the main comment. Also talked to data validation, but around content, yes.

Chuck: Rain, you agreed but have a comment?

<jeanne> Top level categories of Functional needs and the number of specific needs in each Top Level Category: Essential - 1

<jeanne> Sensory - 15

<jeanne> Physical - 14

<jeanne> Cognitive - 23

<jeanne> Independence - 2

Rain: Comment is around date formats, there is a better way around date formats and error prevention. Flexible formatting would be beneficial. Separating dates without the word format in phrasing.

Chuck: Comment inspired me. Seems you looked at error prevention and took a fresh approach, which was helpful.

Rain: Error prevention could also include clearer input fields.

<johnkirkwood> +1 to Rain

SarahH: I think that is a great comment, Rain. Moderated form submission is another topic. I.e. keeping people from making mistakes.

Chuck: Makoto, agreed that it should be included. Concrete examples are presented and allow for comments from people in future.

Chuck: Regarding Gundula's comments, do you want to talk to specifics?

Gundula: Documents are mentioned twice. Platform vs. technology. Mentioning that once, perhaps in technology. Also talks to examples being used. Web is not a technology, rather mention HTLM and ARIA right away.

Gundula: Is automatic validation being looked at? Telephone numbers ? Grouping of numbers of accounts vs. comparing to their source related to input mask. I.e. cut and paste and removing due to formatting.

Capitalization vs. non capitalization of letters used.
… instructions that disappear are not consistent , should that be dropped ?

<Chuck> proposed RESOLUTION: Include Error Prevention in the June heartbeat

<jeanne> +1

<Chuck> +1

<Lauriat> +1

<Fazio> 0

<OliverK> +1

<JF> -1

<Ben> +1

<KimD> +1 to including

<Rain> +1

<johnkirkwood> +1

<Azlan> +1

<JustineP> 0

<Jemma> +1

<AWK> -1

<Wilco> 0

<MelanieP> 0

<Detlev> 0

<Ryladog> 0

<laura> -1 Could agree if we publish Errors without the Current Scoring. Add: "Scoring under development."

<AWK> yes

Chuck: We've already discussed the points, but specific to JF and AWK and objecting to CfC?

Chuck: To Laura?

Chuck: If we were to remove scoring section, does that change your position, AWK and JF?

JF: concern is around the guideline , would be opposed to CfC at this point.

AWK: I think it helps prevent compounding of problem. Core issue is around current state and work of conformance model.

<laura> yes

Chuck: In removing scoring and stating "Scoring under development" could you live with it?

JF: I see the value of the work. Strategy should gain more focus and newer content to specification is not as much as a focus.

Wilco: There doesn't seem to be a lot of enthusiasm, there is reluctance.

<AWK> need to drop for another call

<ChrisLoiselle> thanks, Mike.

Michael: If we want to capture some concerns, we don't need a CFC to do that.

<Zakim> Lauriat, you wanted to ask about how to move forward to avoid similar lack of enthusiasm in publishing future drafts

Chuck: Is there something you think we'd get out of it with a CFC more than captured in this call?

Lauriat: This is existing guidance. We are going to have this same situation. They are going to look different as we migrate them.

<KimD> +100 to SL

Lauriat: I'm concerned we're getting into philosophical debates. We have to build it up and validate things -- change them as they don't work. This is not carved in stone. I want to avoid this hesitation in the future.

Chuck: We have individuals who express concerns about publishing the migrating effort.

<Zakim> jeanne, you wanted to say that we get asymc

Chuck: How to we handle this when we hit the same stumbling block?

Jeanne: I believe it would be valuable to have a CFC since we get non-synchronous participation -- from folks other than on this call.

<Lauriat> +1 to the async point, definitely

Michael: I'm not going to object if you want to do it. Even with the objections we have, it's a foregone conclusion.

<Wilco> +1 to Michael

<JF> +1 to Michael

<johnkirkwood> +1

Micahel: I'm not sure the CFC approach is how we get asynchronous participation.

Chuck: I don't think Shawn's concerns were addressed.

Wilco: I think Silver is doing work, and by the time it gets to the WG there is a sense it is ready to be published.

Wilco: By the WG does not view it this way.

<JF> +1 to Melanie

Melan: This is the kind of the situation I think we need an answer in order to avoid this situation in the future. Issue 376 tackles this.

<Chuck> issue 376 <- https://github.com/w3c/silver/issues/376

<Zakim> Chuck, you wanted to say that bringing the groups together would help with that

Chuck: There is a process that has been crafted to migrate content.

Chuck: This deliberative process is a part of this.

Chuck: Getting more volunteers to help would get more investment.

Chuck: We are actively working on unifying the groups, so that it becomes more of a holistic process.

WCAG 2.x issue resolutions https://www.w3.org/2002/09/wbs/35422/WCAG22-Misc-items/

Silver folks are welcome to depart or stay on as we move to 2.2

Question 1 - 1. Target size understanding review #1770

<Chuck> https://www.w3.org/2002/09/wbs/35422/WCAG22-Misc-items/results

Chuck: AWK had to leave. I will read out his comments.

<ChrisLoiselle> yes

<Zakim> mbgower, you wanted to say I have pushed a new commit that includes most of these comments

<ChrisLoiselle> Mikeg: I went through the comments. I attempted to include changes for all of these.

<ChrisLoiselle> If people refresh the PR, they should see these included.

<ChrisLoiselle> Mikeg: I attempted to address Wilco's. Someone commented on heuristics. I didn't tackle that specifically.

<ChrisLoiselle> Mikeg: All others I attempted to address.

<ChrisLoiselle> Mikeg: That is a new comment around touch, I haven't seen that but will incorporate it.

<ChrisLoiselle> Mikeg: Existing understand talked about touch. I wasn't trying to change understanding doc.

<ChrisLoiselle> mikeg: I didn't bring up touch. I'm not adverse to it, but didn't make those changes. It is another issue.

<ChrisLoiselle> I don't want to edit text on a PR that I didn't alter.

<Chuck> proposed RESOLUTION: Accept amended PR 1770.

<Detlev> +1

<Ben> +1

<Chuck> +1

+1

<laura> +1

<JustineP> +1

<Wilco> +1

<Ryladog> +1

<OliverK> +1

<morr4> +1

<GN015> +1

<jg> +1

<ChrisLoiselle> thanks, Mike!

<johnkirkwood> +1

<ChrisLoiselle> yes, I'm here.

<ChrisLoiselle> :)

Resolution: Accept amended PR 1770.

Resolution: Accept amended PR 1770

Resolution: Accept the draft response for issue #930

Chuck: This was raised last week.

Chuck: Based on discussion a draft response was made. 7 agreed with response.

Rain: Looking at it further, I came to realize a couple of problems. I have seen more than edge cases where changes to more than one affect it.

Rain: My second concern was that the way the response was worded, it feels like we are prioritizing the needs of the testers above the needs of the individuals we are trying to make the web better for.

Rain: I would like us to consider rewriting the response, if we are not going to make the Normative change. M

Rain: "guide testing" instead of whatever the wording was.

<Chuck> ach Ch

Rain: Where I see this is a problem is where someone is using extensions, say for dyslexia.

Rain: using a font would break Medium.

Rain: I've seen instances like that fairly frequently, where someone uses an extension intended to help somebody

David: Would that Medium fail if all four of those settings were made?

David: Or would it correct itself?

Rain: I don't know the answer to that.

David: What I'm suggesting is that if the tester had set all four of those, it would still have failed, so it would have failed the SC.

<laura> If it is tested by setting all of them, that is going to provide the most change, the most likely way to break a layout. If I recall correctly that is what the LVTF had in mind when we wrote the SC. I was the SC manager.

<GN015> q*

David: The testing would be to set all of them. We could do it one at a time, but if you do all of them, the other three are not going to correct that other. i don't see how you alter the pass fail.

Chuck: Just to emphasize, Rain was going in two different directions. She has sympathy with the change, and had offered an alternative response that de-emphasizes the tester

Gundula: I'm trying to wrap my mind around that. if the line height is hard programmed, the text might not overlap, but still the line spacing cannot be as high as expected as the test, so testing for all four at once should also fail.

Ben: I was plus one'ing Rain. I'm okay with it going out, but I'll echo her concerns.

<Chuck> proposed RESOLUTION: Accept Rain's amended draft response for issue #930

<Chuck> +1

<johnkirkwood> +1

<Ben> +1

<david-macdonald> +1

<morr4> +1

+1

<Rain> +1

<Ryladog> +1

<laura> +1

Resolution: Accept Rain's amended draft response for issue #930

Question 3 - Update benefits of 2.5.1 pointer gestures #747

<GN015> +1

Chuck: Nine responses. Everyone agreed.

<Chuck> proposed RESOLUTION: Accept PR 1569 to address issue 747

<Ben> +1

Resolution: Accept PR 1569 to address issue 747

<Chuck> +1

<Detlev> +1

<Ryladog> +1

<johnkirkwood> +1

<JustineP> +1

Question 4 - Text spacing: does it require compatibility with one or all methods? #975

Chuck: 9 people agreed, 2 with minor adjustments

Gundula: I tend to understand "that people can override" that there is the possibility. The new wording suggests that 'as soon as people chose to override the author must react'

<ChrisLoiselle> mgower: I don't think it talks to actively

Chuck: I'm having problems parsing Gundula's interpretation, how it's written.

Chuck: Can you help us understand?

Gundula: It's the word "can" versus "when". It's a timing question.

Chuck: No one else expressed that reservation. Is this a change you can live with?

Gundula: Further down there is an explicit statement on the duties of the author. So I can live with the writing, although I don't like it.

yes

<Chuck> proposed RESOLUTION: Accept amended PR 1687 to address issue 975

Chuck: This is with Michael's amendment I'm talking about.

+1 (and I'll review to try to understand Gundula's point)

<Chuck> +1

<Ben> +1

<Rain> +1

<johnkirkwood> +1

<Ryladog> +1

<GN015> + .5

<OliverK> +1

<jg> +1

<morr4> +1

<laura> +1

Resolution: Accept amended PR 1687 to address issue 975

Question 5 - Error Identification slightly unfocused/mixes up various concepts #977

Chuck: I'm going to go through the comments in order.

Gundula: To keep an instruction that was dropped on the error message being as clear as possible

Oliver: The source is important

Chuck: AWK had some concerns but is no longer on the call [reads out AWK's comments]

Justine: I restract my comment. I'm fine with the updates

Nope, comments can stand

Question 6 - Can text buttons fail 1.1.1 and 3.3.2? #985

<Chuck> NOTE: Ask Patrick to make some updates to this which reflect the comments, and review later.

<ChrisLoiselle> mgower: Do you want me to make a note on your comment?

<ChrisLoiselle> Chuck: I've done it.

<ChrisLoiselle> on it.

<ChrisLoiselle> mgower: my comments stand, Patrick's update work. Input type of submit is an input. It should apply to an SC that applies to a SC around labeling inputs.

Gundula: Interactive controls should be understandable by users. Why is this restricted to data entry?

Chuck: You're comfortable with the other changes?

Gundula: Yes.

<ChrisLoiselle> mgower: To Oliver, my apologies. Could you type in what you said? My keyboard stopped working, sorry.

<OliverK> Agree to Gundula. Would drop the sentence about the link.

<ChrisLoiselle> mgower: I understand Gundula's comments. Without altering requirement, it needs to be constrained as currently stated.

<ChrisLoiselle> Chuck: scope change ?

<OliverK> Interactive components should have clear names and if these have a context relation this would also be appreciated to be added to the name, so the user understands what is being collapsed/expanded.

<Chuck> NOTE: Ask Patrick to make some updates to this which reflect the comments, and review later.

<ChrisLoiselle> mgower: actually wording has content requires user input (I believe) thus would need to looked at more clearly.

Question 7 - "Changes of color" applied to motion definition #1535

Chuck: Jake raised that 'change of color' has affected our motion definition

Chuck reads out the suggested rewording.

Chuck: We have a mix of agree, disagree and something else

Gundula: A move transition can happen in place.

Gundula: Color changes usually are used.

Oliver: I thought the sentence in the middle might be confusing with a double negative. Maybe rephrased in a positive way?

<ChrisLoiselle> mgower: talks to Oliver's point as a double negative, but hard to get around without it.

<OliverK> thank you :-)

Question 8 - Carousels fail 2.2.1: Timing Adjustable? #1658

<ToddLibby> have to drop for another call

<Fazio> +1 gundula

Gundula: I feel that the 'content repeats' is not sufficient.

Gundula: I suggest it applies to carousels.

<Detlev> *1 Chuck

<johnkirkwood> well done!

<ChrisLoiselle> it is working now for me

<ChrisLoiselle> it wasn't working earlier

<Rain> rrsagent: generate minutes

<ChrisLoiselle> I'll send out minutes MikeG.

Summary of resolutions

  1. Accept amended PR 1770.
  2. Accept amended PR 1770
  3. Accept the draft response for issue #930
  4. Accept Rain's amended draft response for issue #930
  5. Accept PR 1569 to address issue 747
  6. Accept amended PR 1687 to address issue 975
Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).

Diagnostics

Succeeded: s/issue 376/issue 376 <- https://github.com/w3c/silver/issues/376

Maybe present: David, Gundula, Justine, KatieH, Laura, Melan, Melanie, Micahel, Michael, Oliver, SarahH