– DRAFT –
AG teleconference

Introductions

Introductions & Announcements

alastairc: Is anyone new to the group?
… in terms of announcements, We will have a face to face at CSUN on Monday. We have a room provided by CSUN and we have sponsorship for catering.

Rachael: It will be hybrid.

Chaals: Will we nee do to register for CSUN?

MichaelC: You do not need to attend CSUN to attend our meeting.

Shadi: Just to confirm that sponsorship has been covered.

Culture check-in https://www.w3.org/2002/09/wbs/35422/culture_check_in

Chris WIlson: From google, know many here, mainly observing

<alastairc> Results (people's comments) https://www.w3.org/2002/09/wbs/35422/culture_check_in/results

chaals: Last year, chairs asked me to run a few meetings to run meetings to better understand culture. This is a check in.
… Survey responses were varied.

<chaals> survey results

<AWK> +AWK

chaals: one request was more screen sharing. I am not going to do so today.

Response to how it was going. More people said good than bad. Not everyone was pleased. I did note that all people who said bad, said it had improved.
… trending in the right direction.

When asked whether we are improving, all indicated improvement.
… I'm not sure if we want to visit that. I also received private feedback. Some was similar to overall picture from survey.
… the private feedback was more negative. If the survey looks good, people are shy about complaining. The big points that came out in private were:
… One of the things that happens is that people walk away. That is a problem to talk about.
… One was a comment about the way stuff gets done where everyone wordsmiths every change. I've also noted this. It would be my feedback if I had answered it. I would say the culture has been improving in a reasonable way.
… in the way that decisions are made, it seems things still take a very long time. A suggestion made in that regard was having an editor responsible for writing down the ideas that the group agrees to and then having the group suggest editorial changes to the editor's work.
… another comment was that when people behave badly, it was suggested that there are not consequences to bad behavior.
… Final point was that intersectionality and how that plays out in this group is a big point. It cuts across a lot of pieces. When we're trying to solve a problem, how important is it to solve the problem before sharing a draft. There are all sorts of considerations about what will work in the real world. The other question is whether we are trying to solve all the political problems of the world.
… these combine to create accessibility barriers or are we trying to solve technical accessibility problems?
… that leads to some misaligned expectations.
… people start talking at cross purposes. There is a point I do want to call out from a response in the Survey from Wilco. "Empathy about understanding other people and if other people disagree, that we do not understand their reason rather than they are wrong." Understanding the disagreement in a way that leads to a resolution rather than one side going home.
… that is my attempt to balance the survey results with the private feedback (about half a dozen who sent private feedback). Since just over a dozen who answered, I think we need to talk about how we manage participation and make sure we are getting input from people who disagree with the consensus or trend on a specific issue. That doesn't mean we should agree with every minority opinion because the group needs to work as a group.
… but people need to believe that they have been listened to and that the group has considered before making a consensus decision.
… Chaals notes that it is hard to be on the disagreeing side.
… Important to avoid making people feel bad about disagreeing.
… comments?

<Glenda> +1 to what Chaals has said. Thank you.

<Zakim> bruce_bailey, you wanted to thank wilco for his coment

<Wilco> Thanks Bruce

<jenStrickland> +1

alastairc: Give people a chance to review and get into queue if they have comments. From a chair's perspective, we can review the survey responses. We will ask Chaals to provide us an anonymized summary of the private feedback so we can review and bring suggestions back to the group. Also we ask that if anyone else reviews the feedback and had suggestions, that they send it to us.

bruce_bailey: Thank you to Wilco for his comments. Excellent feedback.

<kirkwood> +1 to thank you to Wilco

chaals: Also appreciate the feedback from Wilco from initial meeting as it is a topic hard to engage.
… Jen, you were also someone who said we need to make things better.

<bruce_bailey> i appreciated jen's detailed feedback as well

jenStrickland: I had so much to say. It took me so long to find a door into W3C and AGWG and my onboarding than many. I felt that I was a pain in the butt to a lot of folks because I had to ask a lot of questions because I couldn't figure out where things are. I was trying to figure out ways that people like me can be less of a pain to people like you.

<jeanne> +1 to Jen

jenStrickland: as a person with a multi intersectional identity. When Makoto speaks I feel that I can relate a lot. There are people who have been here a long time from a more traditional cultural background. When I bring up equity issues I feel a bit funky. I am grateful for the private messages that support me. I wanted to response to Makoto's comment. Speaking for me, I have never found you offensive in your communication. I am immensely grateful.

<GN015> +1 to Jen

jenStrickland: happy to clarify on anything I said.

<Chuck> +1 to Alastair's comments

<Zakim> Ben_Tillyer, you wanted to say that the queue prioritises a certain sort of person

alastairc: I put myself on queue that it is very rare that we ever consider people as a pain. The only times it really comes up in any way is when people choose to be repetitive and dominating. It was clear that people are concerned about it. We are much more concerned about the people who are not able to express themselves.

Ben: The terms you used is important. I also worry about the number of people in the call we don't hear from. My suggestion to the chairs is to explore an alternative method to the current way of working but I'm not sure what that is.

<jeanne> +1 to respect, gratitude and admiration for the contributions of Makoto

<Zakim> chaals, you wanted to talk about onboarding...

alastairc: It is a perenial topic and suggestions are very welcome.

chaals: I joined this group in 199? so in some ways I've been here forever. I recently joined it again through the official process. I think its reasonable that many people ask questions. I think the chairs are very open to people asking questions.
… I think in the onboarding thing, saying much more explicitly "Yes there is a lot. We understand it takes a while to get your head around. Please ask questions."
… this should demonstrate that we understand it is hard. Set expectation that it is hard to onboard here.

<ShawnT> +1 to the buddy

chaals: the other idea is to think about some kind of buddy scheme. Find someone whose been here a while and who can be your buddy to help out.
… worth thinking about. The question about queing. The usual suspects are called that because they are the first in the queue. They happen to think or work in a way that gives them an advantage to getting to the queue. This is mentioned in every group, so its not unique here but to ask each individual to avoid being repetitive.

<jenStrickland> In a meeting recently I noticed one of the PWE chairs actively invited those who had not queued to speak — only one of the four didn't respond. That invitation helps to make newbies feel welcome to contribute.

chaals: ask people to +1. We can see that people are engaged that way. In this call we have 36 participants.
… putting up questions and providing time to think about it.
… it can also be difficult with writing. This is hard to do. I encourage the chairs and group to continue to discuss it. Chairs volunteer to do the work and its hard. Getting input is helpful.

Wilco: To add to that point, my comment is more negative than usual. I want to acknowledge the chairs have an incredibly tough job. I'm not sure what to suggest. Its tough to coordinate a group with this many strongly opinionated people on it. While I often disagree I also appreciate the work put in.

<Zakim> tzviya, you wanted to comment on what I am hearing (as co-chair of PWE)

tzviya: I am one of the co-chairs of the PWE which published the code of conduct.
… I just wanted to offer some observations. This group strongly wants to have a cohesive and comfortable work environment. There is a lot that people are struggling to accomplish. There are people who don't feel comfortable, welcome or able to express their opinions. Some people are hesitant to express their thoughts and are grateful when people reach out to make them feel welcome.

<jenStrickland> +1 to Tzviya — the PWE has some amazing training coming, too.

tzviya: there are a lot of conversations about how to make the queue work better. It really comes down to each individual. We all have to make an effort to pay attention. If you see someone on queue who hasn't spoken in a while, you can q- to let them go first.

<cwilso> strong +1 to Tzviya

tzviya: the chairs have a hard time with a large and wide reaching spec. The CEPC has a lot of recommendations.
… the slash me "that was a great comment" is also great.

<Zakim> bruce_bailey, you wanted to suggest we might ask more recent members to speak on their experience , maybe suggest ways to improve

<tzviya> cepc https://www.w3.org/Consortium/cepc/

<jeanne> +1 ti rzvuya and thanks for making the time to attend

bruce_bailey: I want to echo the many thanks to the chairs. I also want to suggest the more recent members who feel they have gotten over the onboarding.

<chaals> [I would be terrified of being called out by Zakim to speak on a random topic]

<alastairc> q/

alastairc: I joined 10 years ago and don't think I commented much for the first 2 years. The topics range and selecting people at random could backfire.
… next steps: The Chairs will go over this and get some feedback from Chaals. We will come back to the group with suggestions to change it in the future. Also want to note that things we do to make doing the work easier also help. We will be revisiting that as well.

<chaals> [+1 to Alastair, making work easier is helpful]

alastairc: thank you chaals for coming.

chaals: I will make one more comment before I go. Its very important that the chairs come back to the group. Things we will keep doing and will continue doing but make better. There is work for the chairs to do on this.
… As andrew said in they survey. Remember all the things we have acheived.

Assertions https://www.w3.org/2002/09/wbs/35422/assertions-jan-12/

<alastairc> https://www.w3.org/2002/09/wbs/35422/culture_check_in/results

alastairc: please see results list...

Additional references: WCAG 3 editor's draft: https://w3c.github.io/silver/guidelines/

<Chuck> https://www.w3.org/2002/09/wbs/35422/assertions-jan-12/results

alastairc: request to share screen, so we will do that

Assertions and procedures draft: https://docs.google.com/document/d/10Rl3YJhc4kTyaQwBoQYFp-YvFRK_TdBKkA3kLQKHYwk/edit#heading=h.k1n0svmeyc0

alastairc: survey about assertions, first question asked about documentation ...

Rachael: A little more context as we return from vacation, and our focus on WCAG 3...
… Strong recommendation last quarter about assertions .., define and look for consensus on how to handle...
… please see WCAG3 current editors draft and working document (both links just above)

alastairc: 1st survey question about what document should be required? three suggested tiers

alastairc: fairly even split in terms of voting response , so substantial responses.

GreggV: Yes, my survey response was long, so I will summarize.
… In brief, I argued for a minimal requirement.
… Companies are reluctant to name "people in the room" so detailed requirements for assertions might result in fewer assertions.

Reference to last week's survey results: https://www.w3.org/2002/09/wbs/35422/assertions-jan-2023/results

alastairc: You also suggest the "things that count"

<chaals> [Bruce fairly well expressed what I think for this first question. Documenting things, by itself, encourages people to do *something* - and although it isn't always what we want, there are plenty who decide that doing the right thing is as efficient as trying to work around it]

GreggV: For an organization say that they conform for some long document will not be productive in actually.

Alastair ask Gunddalla about her response, that significant documentation required -- but no expectation to share.

<alastairc> Bruce: Should be limited.

Wilco: From survey, we should leave question to regulator.

Bruce: No strong feelings, but agree with Wilco. Agencies subject to FOIA anyway.

Jaunita George: We have experience with VPATs and similar , for years , and they are not so useful. So this will not be an added value unless more goes into the requirement.

Mary Jo (from survey): Limit burden of doing this.

<Zakim> Rachael, you wanted to ask Wilco for his thoughts on how we would balance needing clarify from us but not stepping into regulation

alastairc (from survey): 3rd party auditors are likely to create own reporting form.

Rachael calls on Wilco to expound on leaving to regulators.

Wilco: For public organizations, they will have larger reporting and assurances and legal requirements. A WCAG assertion is not going to change their real-world a11y.

<Zakim> chaals, you wanted to respond to Juanita

Wilco: We might recommend making information available, but i do not see how we have any authoritative say in it.

<maryjom> +1 to Wilco

Chaals: Replying to Jaunita's concern with low-information VATS -- there is liability with making those assertions
… so it is important to provide the guidance on what organizations following wcag really need to do.

<Glenda> +1 to what Chaals is saying. I find more large companies being achingly honest on their VPATs. Adobe is a great role model for their transparency. Thanks Adobe!

<kirkwood> +1

Chaals: More extra document might not change much. Increasing the requirement to say what you did comes at cost of doing the work

<Zakim> GN, you wanted to discuss asking for more details on unreliable VPATs (they are legally binding)

+1 to both jaunita and chaals. Both points seem valid

<Zakim> Rachael, you wanted to suggest possible middle ground of WCAG requires minimum made public, suggests kind of documentation that might be used to support, recommend making public

<jon_avila> In my experience VPATs are often not legally binding

Gundala: I have seen VPAT which are non sense.

<chaals> [dishonesty is universal... not just limited to the US :) ]

<Glenda> https://github.com/GSA/openacr

Rachael: We could also do middle possition.

<alastairc> suggested overall response: Limit the documentation required to information about the assertion itself. Recommend that organizations maintain internal documentation on the assertions. Third-party testers would not have access to information to support the assertion unless it was volunteered.

<jeanne> +1 to Mike Gifford's work

<kirkwood> never seen a VPAT be legally binding

Glen Sims: I want to recommend working going into Open VPAT and making reports machine readable.
… Some things hard to measure. We do not have to tell goverments what to requrire.

<ShawnT> [Overview | OpenACR Editor | GSA](https://gsa.github.io/openacr-editor/)

<mbgower> +1 to Juanita

Jaunita_George: there is not lots of expertise in industry for completing vpats , vary widely in completeness and quality
… companies do not really know how to fill out in practice
… if WAI does not validate , assertions will not be meaningful.

<kirkwood> +1 to Juanita

<Zakim> GreggVan, you wanted to say that I think the main role of assertions is to put non-testable recommendations into WCAG when we wouldnt otherwise be able to -- so people see them and consider them -- and that they can get credit for doing them.

alastairc: Please focus on requirements for assertions themselves for this conversation.

GreggVan: Need for assertions come from difficulty with putting not-so-testable objectives into documents.

<Zakim> jeanne, you wanted to say I gave a proposal in the answer to the next question that may be helpful -- to have more paperwork for Bronze outcomes

GreggVan: VPATS are not required, v is for voluntary , sometime procurements require but maybe nobody reads

Jeanne Spellman: Please see response in next question, but requirement and detail might scale with tiers (B/S/G).
… We would not do enforcement, organization might be required to compile.

maryjom: V for voluntary not correct if agency asky for one. If part of bid response, can be binding legally.
… as far as checking on assertions, certain want a way to encourage, but maybe not include as requirement because then legally biding and default part of bid.
… please see AWK definition.

<mbgower> An assertion is an attributable and documented statement of fact regarding procedures practiced in the development and maintenance of the content or product to improve accessibility. Assertions are only testable in that one can test that the assertion has been made correctly - not that any desired result has occurred. The results are always true/false.

[from survey]

<AWK> That wasn't mine so much as an edit on what someone else had, fyi

<Zakim> bruce_bailey, you wanted to say something about vpats

[from original survey]

<alastairc> bruce_bailey: VPATs were a response from industry to section 508.

<alastairc> ... we didn't require those as part of 508

Bruce: Access Board did not include 508 with Refresh
… VPATs industry response to Original 508 Standards.

<jeanne> +1 to having a higher level person in the organization certify that records are kept.

jon_avila: I like FCC requirements under CVAA as a model....

<Zakim> alastairc, you wanted to talk about appicability across orgs

jon_avila: FCC includes performance standards , that covered entities compile documentation , and that they consult with PwD

alastairc: Our guidelines are meant to scale, work for individuals as well as multi-national corporations....

<Wilco> +1 Alastair

<GreggVan> +1 to that. and what do teachers, and small orgs or a cafe have to do>?

alastairc: at best we might just make recommends , that organization are encouraged to make assertions and then recommend level of documentation.

<alastairc> Limit the documentation required to information about the assertion itself. Recommend that organizations maintain internal documentation on the assertions. Third-party testers would not have access to information to support the assertion unless it was volunteered.

<GreggVan> +1 to alastar suggestion

[informal hybrid position from survey]

<Zakim> mbgower, you wanted to say IBM has 3539 published (publicly accessible) VPATS

<jon_avila> In my experience current VPAT reports do not show much about process is being followed and doesn't really show impact to users in using the software or service.

Jaunita_George: Many organizations already do something similar. Can we not build on that?

Mike_Gower: IBM as an example, we have thousands of VPATs / ACRs which are public facing...

<jeanne> +1 Jon Avila -- VPATS doon't apply to the types of information we are talking about for assertions

Mike_Gower: Vary in quality since decentralized and dozens of people completing / submitting results.

<Zakim> alastairc, you wanted to say what value it can addd

<jeanne> +1 mgower

<kirkwood> big +1 to mgower

Mike_Gower: The agency puts their name on them and takes them seriously and knows there is liability.

alastairc: Current VPATs are an attestation to 508/wcag2 , wcag3 assertions are more focussed on documenting processes.

<maryjom> VPATS are based on WCAG, 508 and EN 301 549. Updated whenever standards/regulations are updated and can be updated to make assertions as part of the content (just like we have test methodologies).

+1 to hooks and +1 to assertions we are discussing here being an add to VPAT like documents after WCAG 3 vs a replacement

shadi: Appreciated the discussion and intersection between accessibility and conformance not actually being the same thing...

<GreggVan> credit without documenting -- I hate documenting - especially every meeting I have. so forget about it. (and walks away from even looking at them).

+1 Shadi

shadi: they are overlapping Venn diagrams , not subsets , and enforcement is another huge factor.

<shadi> https://w3c.github.io/silver/use-cases/

shadi: We are going to have these intersecting issues keep resurfacing , our work is more on the technical side.

<Zakim> GreggVan, you wanted to say I like accountability - but I worry that documentaion will drive people away. Even those who would DO don't like to document The logic runs backwards - and not to good conclusion. 1) I have to document this? that is too much. I have too much paperwork already - I am designer. 2) I'm not going to document so - I don't get any credit 3) so I guess it is easier to just skp these. They arent required, I can;'t get

alastairc: Summarize what I am hearing: Very limited documentation about the assertion itself. We can make suggestions in addition.

<alastairc> draft RESOLUTION: The agreed approach is to: Limit the documentation required to information about the assertion itself. Recommend that organizations maintain internal documentation on the assertions.

GreggVan: Even folks who like accessibility hate documentation of the work. So reaction can be to do less, or claim less at the very least.

<mbgower> I would remove the word "internal"

<Wilco> +1 Mike

<jenStrickland> +1 to mbgower's suggestion

<mbgower> It can be, but let's not prescribe that

<Zakim> GreggVan, you wanted to say that you keep the end of it. that

GreggVan: liked 3rd sentence from above that documenting is voluntary...

<alastairc> Limit the documentation required to information about the assertion itself. Recommend that organizations maintain documentation on the assertions. Third-parties would not have access to information to support the assertion unless it was volunteered or legall required.

<GreggVan> +1

GreggVan: do not want to imply that there is something which auditor might ask for.

<Zakim> mbgower, you wanted to say that the proof is in the pudding

mbgower: I know some people are concerned with only requiring the statement, and not the work perhaps. It is not a "get out of jail free" card...
… if organization says "we use plain language" for example -- that will testable if site is in plain language.

<mbgower> Look at AWK's example in the next survey question

+1 to jeanne

jon_avila: What are we considering beyond the assertion?

<Zakim> jeanne, you wanted to say that we agreed not to wordsmith -- let the editor's take the ideas and bring it back.

alastairc: Assertion requirements in survey, basically level and date.

jeanne: I would remind folks this is not a word-smithing session. Looking to get direction from WG, not the final language.

<alastairc> draft RESOLUTION: Limit the documentation required to information about the assertion itself. Recommend that organizations maintain documentation on the assertions.

alastairc: Agreed, looking for consensus on direction.

<Zakim> mbgower, you wanted to say I can provide at least one example for Jon

<Chuck> +1

<joweismantel> +1

+1

<ShawnT> +1

<jeanne> +1

<jenStrickland> +1

<Azlan> +1

<Makoto> +1

<Jaunita_George> -1

mbgower: AWK example in survey is pretty strong. Is that what we are saying?

<jeanne> +1 that AWK's example is good. I think we can apply that as we get into more detail

<mbgower> +1

<maryjom> +1

<laura> +1

<mbgower> Statement of assertion (example: An evaluation against plainlanguage.gov was conducted)

<jon_avila> +0 I believe we should have some basic information that is required to be maintained such as date

<Chuck> There's a -1 and a 0

alastairc: Chairs and editors will go through responses, we are trying for direction.

<chaals> +1

<Chuck> 13 +1s

Jon, we will revisit exactly what goes into an assertion and supporting docmentation in a future meeting

Jaunita_George: Asking for greater description of documentation.

RESOLUTION: Limit the documentation required to information about the assertion itself. Recommend that organizations maintain documentation on the assertions.

<Wilco> +1 Jaunita, although I'm not sure that contradicts the draft resolution?

<ToddL> +1

Bruce: i think "limit the documentation" needs more definition

mbgower: VPATs have very little requirement on its form. They are not required to be dated for example. Who filled it out?
… I am concerned with level of tracking we are suggesting.

<jenStrickland> Would it be appropriate to let the editors wordsmith it, with this input?

alastairc: We need to create some assertions , work out what we are recommending for assertions.

<Zakim> Chuck, you wanted to say this gives us a direction on the approach

alastairc: We need to make some progress in the meantime.

Level Assertions Apply

Chuck: The current resolution and voting gives us a direction to pursue, so that is sufficient for now.

alastairc: The second survey question was about what level, bronze, silver+, gold only?

alastairc: I will summarize this time, owing to time, please put yourself in queue if want to add something.

[Alastair reads from survey]

alastairc: GreggV made case for Silver+

<mbgower> Agree with Jeanne and general comments that it is too soon to restrict to certain levels

<chaals> [Agree with Wilco / Jeanne. Assuming that people go for the middle option doesn't mean we should pile everything there]

alastairc: Gundaula noted in survey that terms are a little hard to track and maybe not being used consistantly.

alastairc: My conclusion is that we shaking out the issues, but no consensus yet....

<mbgower> +1 not ruling out bronze assertions

<alastairc> Suggested RESOLUTION: Bronze level would be where most 'outcome' based requirements will go, but we are not ruling out some assertions going at bronze level

alastairc: One outcome might be to make asserts only expected/required for Outcomes.

<mbgower> +1

<jeanne> +1

+1

<GN015> +1

<AWK> +1

alastairc: This does NOT rule out some assertions at bronze level. TBD.

<Makoto> +1

<ShawnT> +1

<jenStrickland> +1

<Azlan> +1

<Jaunita_george_> +1

<laura> +1

<kirkwood> +1

<jon_avila> +1

RESOLUTION: Bronze level would be where most 'outcome' based requirements will go, but we are not ruling out some assertions going at bronze level

<GreggVan> 0 they would be there -- but would not be required there

Approval to add content

alastairc: Third survey question asks if Editors Draft can remove procedure test and replace with current material.

Rachael: Some survey results do not match answer, so I will rephrase and re-survey

alastairc: Also see some editorial work clarifying processes versus procedure
… Was this a Note or replacement?

<Wilco> +1, right direction.

<alastairc> Question, For Assertions, are we moving in the right direction?

Rachael: Really, just asking if current work is a better direction?

<chaals> [My apologies, I have to leave..]

<Chuck> +1

<ShawnT> +1

<Jay_Mullen> +1

<jeanne> +1

<jenStrickland> +1

<Detlev> can't tell

<Makoto> +1

<shadi> +1

<chaals> 0 - it's too early to tell, let's see the assertions themselves take shape and then we can check.

<Jaunita_george_> I would say we would be if we made some changes

<alastairc> https://docs.google.com/document/d/10Rl3YJhc4kTyaQwBoQYFp-YvFRK_TdBKkA3kLQKHYwk/edit#

+1 to those edits. I will work on integrating that in over the next day or two

<jon_avila> +0

<laura> 0 not sure

mbgower: I note that second bullet in each group , from conversation , are in flux

<mbgower> I suggest "assertions can include" instead of "assertions include"

Jaunita_george_: Would like to know better how this material fits with other requirements or required documentation

WCAG 2.2 CR publication discussion (if needed)

alastairc: Hearing no further concerns on survey, moving to next agenda topic

alastairc: Any question on CR restart?

Wilco: We had some plans for research on current phrasing for focus appearance ?
… Are we aiming for 80% concurrence between seasoned testers?

alastairc: We are in new territory, but cannot have "at risk" item for CR...
… We do have feedback from some seasoned testers.
… We will need to make decision before going to CR.
… We are hearing that dashed lines are ambiguous for some audits.

Wilco: Can we go to CR again if testing results go south?

alastairc: We can potentially update an editorial clarification on one of the exceptions.

We will not be making a significant change to main body of SC.

alastairc: Please see list responses , we are still looking on GitHub for more examples
… it would be good to have more representative examples in understanding
… not just edge cases either, simple example are of value too.

<Zakim> mbgower, you wanted to say that I've been using the Focus Appearance concept on a lot of feedback. very useful

alastairc: If CFC continues as is, we will be going to CR very soon. We are still updating documentation for passing implementations at AAA.

mbgower: I would like to report that IBM is using present wording with working designers, and present phrasing has been useful...
… especially with barely passes examples.

alastairc: We have been using UK.gov for testing. Large scale and meeting AA.