Silver Task Force & Community Group

19 February 2021


AngelaAccessForAll, bruce_bailey, chrisloiselle, Chuck, Fazio, Francis_Storr, jeanne, Jemma, jennifer_strickland, Laura_Carlson, PeterKorn, Rachael, sajkaj, SuzanneTaylor
Azlan, Charles, Sarah, Todd, Wilco
jeanne, Shawn
Chuck, jeanne

Meeting minutes

<chrisloiselle> I will join on call after my current call is over, sorry for the delay on my end.

CFC on Extended Acknowledgements

Jeanne: Respond to the CFC on extended acknowledgements. Realized that it closed at 10am this morning. If you haven't voted, you can get in a response. It is now closed. Is Rachael present?

Rachael: I'm closing out the CFC as we speak.

Rachael: There will be an announcement after the meeting.

Jeanne: Thank you everyone. We can start filling out names for the next WD. Any q? before I go on?

Jeanne: We want to turn on zoom's auto captioning. Can Michael?

Rachael: May be able to claim host and do it yourself.

<RickBoardman> Host Wars ;)

Jeanne: Got it turned on.

Examples of Use Cases from Conformance Options

Jeanne: Any q on the CFC?

Jeanne: I've asked conformance options sub group to give us a preview of their work. They will be showing their working doc, we will only look at some examples to get a sense of what the work is doing.

Jeanne: And to get some feedback. Not specific language or wording, it's still being worked on. This is general feedback on the direction.

Janina: I pasted the URI for the working doc.

<PeterKorn> https://docs.google.com/document/d/1GyUYTnZp0HIMdsKqCiISCSCvL0su692dnW34P81kbbw/edit#

Janina: Our process first, then focus on a part of the larger doc where we'd like to take some time.

Janina: Back in OCT we articulated some over-arching principals that described WCAG 3 success. It's a really high bar, it's not an excuse for not making things inaccessible.

Janina: 2-7 are specific enough... we've worked on use cases that might illustrate some gaps or areas of discussion we might want to make sure get covered in the spec.

Janina: That would make that high bar happen. #2 says "you should be able to do what you want do with minimum difficulty". #3 prioritize

Janina: We have 2 use cases from 2 and 3. Don't worry about the wording, we'll clean that up later on.

Janina: We found use cases to be very interesting.

Janina: Simple use case for #2.

Janina: Slightly more complex use case for #3.

Janina: Peter can read out the use cases. If memory serves, if you want to follow, this should be the top of page 2.

Peter: I want to add one other piece of framing. This sub-group came at the issue in a slightly different way. Reaching the same conclusions, from the viewpoint of a goal to make the largest most complex and dynamic sites accessible

Peter: In a way that's achievable. Thus far we've broadly seen the approaches taken in Silver are well aligned. There are 1 or 2 things we are still working on.

Peter: May lead to expansions or changes.

Peter: First use case at top of page 2. <reads>

Peter: In this case we have a page with a small amount of content, split between two visual headings. Not marked up properly.

Peter: Screen reader users wouldn't be able to skip to 2nd heading. They would have to perceive all the content up to the 2nd heading. They wouldn't know that the 2nd heading is a heading.

Peter: This was another way of illustrating what we've been talking about of an image w/o alt text in a footer, we have our existing alt text scoring mechanism to do...

<jeanne> Principle 2 Use case [discussed 4Feb21]: A web page with a small amount of content (e.g. 200 words), split between two headings. The headings are only displayed visually, but not marked up as headings. Screen reader users will not have the ability to skip directly to the 2nd heading, and so have to listen to / read the contents under the first heading before they discover the second. They

<jeanne> likewise wouldn’t be able to discern that the text comprising the second heading is a heading, and not part of the contents underneath it.

<jeanne> [Jeanne: this is not currently covered specifically by Silver, but would be covered by scoring within the guidelines (the heading guideline would need to be updated for this use case)]

Peter: But hasn't yet attached itself... we haven't built it in our scoring mechanism for headings that would fit into this use case.

Peter: This working google doc.. we are using it, it doesn't have all the accessibility support that all the sub team uses. We ask that if you want to make contributions to this google doc, do so only as comments in the margin, not as edits to the text.

Peter: The mechanism for tracking text changes is not accessible.

<Chuck> Peter: Let me bring up to hand the fpwd.

CA: You said that our current draft doesn't cover it because I thought it did

<PeterKorn> https://w3c.github.io/silver/guidelines/#structured-content

PK: In the FPWD, it isn't an issue of the structure, but it isn't covered in the guideline scoring system

<Chuck> Peter: I should say, it isn't an issue of the structure, but I believe it's not covered in the rating system for the guideline itself.

<Chuck> Peter: Going into structured content we have the rating section. <reads from>

Peter: This page would get a rating of zero. This is different from what we have images of text or text alternatives where the rating system itself brings in a notion of whether the text alternative or critical or not.

<PeterKorn> Rating 0 25% or less of expected headings are present and describe the content contained in the section OR there is a critical error in the process

Rick: I'm new to the group. I have some feedback on Symantecs of the use case. The language around accomplishing what they want with minimum difficulty.

Rick: Suprised that we don't have language for people without that difficulty.

Peter: This is a working document. This is not yet read to be proposed as recommended language. We are trying to... this effort started when we were trying to look at what are the kinds of challenges that really big complex programmatically generated sites have.

Peter: We know we need to wordsmith it, there's a note attached to the phrase, knowing we need to get back to that. We are hoping we can focus our share and request for feedback on what we are trying to do with the use cases.

Peter: Want to avoid wordsmithing at this time.

Rick: What would be helpful would be to include a table of contents. This helps to navigate and to refer to sections.

Peter: Yes, it's a working doc for a small group of people. Is not felt to be ready for publication or wide sharing.

Rick: OK, I'll be patient :-)

<jennifer_strickland> Rick, I think putting a comment in the document as others do makes sense. :)

<laura> G docs do have a document outline feature.

David: I feel like this goes well with Maturity Modeling. We could somehow tie in scoring, somebody won't be completely accessible. How can you be compliant but still have defects that need to be fixed.

David: Jake can talk about the EU standard. It starts with a claim, then there are some questions in the conformance claim. Lays out a roadmap of these things. It's ok to not be completely accessible as long as you continue to advance.

David: I feel like this goes well with the work we are doing and with the evolution of the standards in the EU and broader.

Jeanne: Structured content guideline is our prototype guideline, doesn't have a sub-group working on it. I don't think in the rush to get everything ready for publication, I didn't spend sufficient time that what was in there was accurately reflecting the direction.

Jeanne: There's a lot of things in the scoring that got decided at the last minute. Not everything got updated, particularly on headings, because no group was directly responsible.

Peter: Part of what we are trying to do in the sub-group, and you can see on first page under timeline, is to report on next month, the extent to which the use cases are already well addressed by the language in the editors draft of WCAG 3.

Peter: And things we think aren't quite covered yet. Could be minor, like tweaks to the rating language...

Peter: My reading of the rules in ratings of fpwd would score this use case "zero".

Peter: Which may not be what was intended. As well as where we might need more substantial changes to accommodate the use case. We might disagree if a use case is allowed under bronze or not.

Peter: We'll have a body of use cases to help us get concrete.

Chuck: +1 to Peter's goals, groups goals.

<jeanne> +1 to the value of use cases

Peter: At bottom of second page we have the second idea, which was that assessments for conformance should be modeled around what the site is for.

<jeanne> Principe 3 Use Case example: The City of Whoville’s website includes a calendar of upcoming events, so that citizens of Whoville can learn about what will be happening in the future, and decide whether they want to participate in any given event. As a few Whoville citizens travel, and some events take place virtually, as a convenience there is also a tool on the calendar site for converting

<jeanne> event times into different timezones. Generalized timezone conversion isn’t a primary use of the site, and an assessment of the accessibility of the Whoville calendar might assess the accessibility of the primary calendar information (what the events are, and when they take place in the timezone that Whoville is situated in) to be at a different, and higher, priority than the accessibility

<jeanne> of the timezone conversion convenience feature (which in any case is commonly available elsewhere).

Peter: We are leveraging what the expected user interactions are for the site. The use case example is the fictional city of who-ville. They have a calendar of events.

<Fazio> Wireframes can help prove the claim of what the site is for also

Peter: Citizens can know what's happening and if they want to take part. Because a few citizens travel, and some events are virtual, the city added a tool for converting into timezones.

Peter: Generalized use of timezones isn't the primary use case of the site.

Peter: The events information is of higher priority than the timezone convenience feature.

Peter: Might make the assumption that people who travel are familiar with the process.

Peter: This use case isn't attempting to declare what bronze, silver, gold should be assigned. These may be two different levels of importance, and our conformance assessment should recognize those levels.

Janina: People who travel are most likely something on hand to indicate time where they live. So not just converting random timezones. Most likely to know about who-ville and where they are physically present.

<RickBoardman> Suggestion: switch to comment access only except for core authors

<jeanne> CA: There may be two concepts: That the time zone is not accessible, could an individual de-scope this feature outright?

<jeanne> PK: That is a potential approach. If an entity that is mandating conformance isn't going

Peter: That is a potential approach, where it runs into difficulty, is if the entity that mandates conformance isn't... that a conversation about de-scoping may be difficult to have.

Peter: If you have a law that your entire site must conform, but this part isn't important, the argument may be raised that it was important enough to put there.

Peter: There needs to be a mechanism that supports this kind of granular analysis, we run into same issues as WCAG 2. Someone may not be comfortable with descoping some content.

Peter: And the problem of 3rd party content and a desire to bring 3rd party content into a more intentioned way of making conformance statements.

Peter: Now you can make a claim of partial conformance, which is not a claim of full conformance. Laws that say you must conform, then you can't have partially conforming 3rd party content.

Jeanne: We set up a process in the point system of critical errors, I think this could also be handled there. We haven't written guidelines that are directly applicable to this example. It's a good use case to save for when we do write guidelines about it.

<Fazio> I've mentioned this before the ADA allows for removal of "readily achievable" barriers. Which kind of allows for partial conformance as long as you have a plan

Jeanne: Timezone conversion isn't in the process the user is trying to accomplish, and it's not flagged as critical to the guideline, it wouldn't fail. I could see how guidelines could be written...

Jeanne: That would accommodate this use case. It's important to look at. It's a classic example of a developer throwing in a feature that may not have been tested for accessibility.

Janina: It's also asking how we scope the process. If we go by the feature, conformance claims could be unintelligable.

Susan: I think it's great to have this as example. I'm comfortable with the idea of a critical path and critical errors in the working draft. Some of the discussion is concerning, in that "if it's a law", then we have a hidden way in the guideline that allows us to not conform.

<Fazio> +1 even though well-intentioned because 100% conformance is unlikely

Susan: Whatever we do, the critical path is more important. This would need to be simple for the user. It's not a sweeping "this feature need not be accessible".

Susan: We actually don't know if a person has a kind of disability or what they have access to.

susan: That does not mean that out in the wild that the guidelines would be interpreted in such an authentic way.

Susan: There shouldn't be any features that are allowed to be wholely inaccessible.

<Fazio> +1

<laura_> +1

Susan: We don't want to promote such a message. And we don't want to encourage assumptions. That's a big way that users with disabilities get left out.

<Fazio> that happens a lot

<jeanne> +1

Susan: Such as "a blind person isn't interested in art". We need to be very careful.

Jeanne: Thank you.

<Fazio> Again why I like the Maturity Model approach

Peter: I agree, this is hard. This is tricky. Coming back to the first principal, this should be a high bar and achievable bar that doesn't bless inaccessibility.

Peter: As david points out with maturity models, part of the goal should be to help guide website authors in how to get better, in what to prioritize.

Peter: "here's a higher medal you can get, and here's how you get outstanding". As we saw from webaim study, perhaps 1 or 2% of home pages of top million websites can past just the automated tests.

Peter: We have an immense task of moving industry towards accessibility. I worry about saying the "next bar" is perfection, we won't be more successful with WCAG 3.

Jeanne: There in lies the difficulty. Susan do you want to respond?

Susan: The comment is that there's some websites, including those that claim wcag 2, many of them aren't accessible, I'm not sure that's relevant. What I'm concerned is that if you say "it's compliant"....

Susan: Lots of sites are following wcag 2 and claiming conformance. Those sites have bugs, and will always have bugs. I hope we are moving away from "do we have to or can we avoid?"

Susan: This happens with alternative text. Lots of time spent on evaluating images.

Susan: Specialists spend time putting images into categories, so that when it goes to a vendor, it's a different price scale.

Susan: It doesn't help the user at all. It's an industry specific thing. Sometimes you have the better specialists making the categories, and you have vendors writing the alt text quickly. That's backwards.

Susan: I think we want to avoid spending time on deciding if a widget is important.

Jeanne: That's a really good use case. Could you write up a use case?

Susan: Sure!

Peter: That is completely backward... if the amount of energy to decide if something should be made accessible approaches the energy to just fix it, we've set up our incentives wrong.

Jeanne: I'm hoping we get from these use cases these consequences. That's why I'm enthused about the sub-group doing them. We can start to see the unanticipated consequences through these use cases.

Jeanne: I agree with Peter, we do not want to set the incentives up incorrectly.

Peter: ...fixing medium or large sites that already exists. Such as adding audio descriptions to movies. The cost to creating a new studio is lots of 100K of dollars.

<Fazio> +1

Peter: If you look at a library that's a million hours long, times $19 a minute, that's prohibitive.

<jeanne> +1

Peter: If you are creating a book for the first time, it's different from a cost perspective than remediating an existing library.

<Fazio> sounds like a good use case

Peter: It's under 3rd party. If you look under 3rd party examples, we've got a discussion going. Under "5" if you want to read along.

Peter: We are piecing them out and making separate use cases. "Licensed content from 3rd party libraries".

Peter: vs fresh content.

Peter: With 3rd parties that don't have expertise. This is all absolutely real world.

Peter: We need to review the cases separately, and then fold back together.

<Zakim> jeanne, you wanted to say (after the detailed discussion) that I have been seeing new use cases from the Github Issues and have been assigning them to the subgroup.

Jeanne: I have been going through the issues that have been coming in from the public, with comments on fpwd, I've spotted a couple of good potential use cases, I've assigned them to the sub-group to be reviewed.

Peter: Great.

Janina: We have issues assigned to us!

Peter: We look forward to reviewing those.

<Fazio> we will use this doc to guide our development of scoring, and wcag 2.x reconciling in the maturity model

Peter: Jeanne you had opened this discussion with a question to the group. Does this approach feel useful and helpful? Is there any feedback on what would make it better still?

Jeanne: Do others have comments, suggestions?

David: I think we can use this to guide us for scoring, and when we import more WCAG 2.x sc.

Peter: I hope that it won't be this specific doc, but a cleaner version.

David: Whatever is available at the time, we'll definitely keep in touch.

More on Bronze Silver Gold

Jeanne: Thank you very much to conformance options group. Great work! Excited how it will help make the doc and process stronger and better!

Jeanne: I want to find the flaws early.

Janina: Do we have a specific label for issues?

Jeanne: I'll put it on your wiki page.

Janina: I'll put link to wiki into irc.

<sajkaj> https://docs.google.com/document/d/1GyUYTnZp0HIMdsKqCiISCSCvL0su692dnW34P81kbbw/https://www.w3.org/WAI/GL/task-forces/silver/wiki/Substantial_Conformance

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

Jeanne: That's two links that got concatenated.

Jeanne: We only have 7 minutes, I don't want to restart the conversation on bronze/silver/gold, we will talk about it on Tueday.

requests for people to work on proposals for Issues

Jeanne: I'll take up last one. I've been going through more issues, and doing triage. There are a number of issues that need proposals and need resources.

Jeanne: For things that can be assigned to a sub-group, I've done that. But there are some issues that apply across the document or don't really apply to a specific targeted topic.

<jeanne> https://github.com/w3c/silver/labels/action%3A%20assign%20to%20editors

Jeanne: "The whole structure needs to be taken apart and put back together". There are issues being assigned to editors.

Jeanne: I'm looking for people that would pick out one and actually doing some writing to address the concern. Probably a few people who have usability experience to improve layout and structure.

<Fazio> I can give some usability time

<RickBoardman> Happy to help!

Jeanne: When I look at it, made sense at the time, but too much detail frontloaded, and too hard to find the meat of the guidelines.

Jeanne: Thank you David and Rick.

<Francis_Storr> I can help with that

<jennifer_strickland> Jennifer raises her hand to help

Jeanne: Jennifer, you interested in helping work on the structure, or an issue?

<RickBoardman> I can bring a beginners' mind (AKA clueless newbie ;)

<Fazio> Got Team UX!

Jennifer: The usability of the structure.

<Fazio> Go not got

Jeanne: I'll set something up. We'll hopefully have a big improvement in the next working draft. Again, I ask everyone, unless you are active in a sub-group, if you would take a look at the list and see if there are things...

Jeanne: That you can work on and write proposals. If you look at our wiki page, processing comments...

<jeanne> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Main_Page#Processing_Comments_on_WCAG_3

Jeanne: you will see there's a link to a folder in google drive where you can put proposals. I'd like to ask to keep things organized, if you could do a small document where you say what issue you'll address, that you make a google doc and put it in there, or if you want to do a github pr, just create a new branch.

Jeanne: Put your initials in the branch. If you are comfortable with that. Use the tool you are most comfortable with. I want to keep the proposals together.

<jeanne> https://drive.google.com/drive/folders/1WQ5wfAQ30cfS3VfwspmvyN-mzhhyY9l-

Jeanne: That's where I'd like you to put proposals. Any questions or volunteers?

Jeanne: Any issues to wrap up?

David: Very productive meeting.

Jeanne: Thank you Peter and Janina!

<laura_> bye

Minutes manually created (not a transcript), formatted by scribe.perl version 127 (Wed Dec 30 17:39:58 2020 UTC).


Maybe present: CA, David, Janina, Jennifer, Peter, PK, Rick, Susan