Meeting minutes
https://
HR process for reviewers
Any thoughts on what you as reviewers need documented/done in the review process? (Existing docs on the wiki)
https://
Neha: Can we have some kind of use cases/examples of issues raised, and how to track them? Which repo will it be tracked under? Could help newcomers understand.
Mat: Thought I had it documented, will follow up and get it done
Mike: A mock issue with steps of how it looks i github ..
Neha: We newcomers could make an FAQ
… with respective links to the answers to the questions such as which issues are assigned to us.
matatk: Wonderful!
Fazio: Helix Opportunity created a GitHub classroom account, which is a new service. We could create an assignment that steps people through the process.
fazio: We integrated github plugin with Helixopp.institute
Fazio: It can be useful for learning purpose
Possibility of more asynchronous working?
matatk: we should focus on comments added as we assume that we have answered all FAQ
matatk: We use time offline to review everything and discuss blockers during the calls
matatk: Thought on how useful agendas are, and possibilities of async work.
Neha: We can do more async work. Can use the comments section in issues before the call. People can assign themselves. Can discuss blockers during the call to save time.
chiace: To save time yes but we can work offline
chiace: This is a good point. I am in a few other WGs and sometimes we work together during the session, but I think it takes a long time. Can take an hour to modify a few sentences. It's meaningful that you understand deeply what's going on but it's also good to be able to work offline, and discuss blockers on the call.
… when discussing together you can get to a meaningful solution, as everyone has a different perspective. but takes longer
<mike_beganyi> +1
matatk: the concern that I have is people who are reviewing specs can come up with a nuse case andthese things should be discussed during calls. There is a risk working asynchormously but worth trying.
mike_beganyi: opportunity for people who really want to do meaningful contribution
mike_beganyi: It does not have to asynchrous completely but a hybrid model
mike_beganyi: being in pairs is great
<matatk> +1,000 to Janina - *everyone* has great things to contribute
janina1: we are trying to save time, everyone can contribute, key in asyn work model is to talk about concerns and to make familarize with issue and how to work together
janina1: .. come prepared
<mike_beganyi> +1 to Janina
Lisa: I can not communicate via github. I can go through an issue but communication is a problem
Lisa: Names for issue need to be accessible not just the number but some meaningful names
<Lisa> and cant comunicate via github
matatk: we can have someone to work on congnitive accessibility of github platform
Lisa: We can ask this on COGA task, we can use google docs but is can not be accessible for everyone. Email is the best to get hold of me. There can be a redundancy
Lisa: Notification email can be useful as well
matatk: spreadsheet is a good way to track and follow up through the issue
<Lisa> reviewed once a month (first monday)
<Lisa> github training for coga we made https://
matatk: Outside APA communication works well with github
In the next session (later on) they are talking about about Agentic AI and Accessibiity: https://
Overview
Next steps on Maturity Model; Support on responding to contributions
<Lisa> stepping away for a min
matatk Maturity model, plans future versions of the documents
open discussion
matatk Background we published, people are using it. Few feedbacks to be addressed. Challenge is that occured were in the communication with the people who posted the issues.
janina1 bring forwad some documents they are going to run through a CFC process in APA
matatk 2 parts: 1st what do we actually want to do with maturity model
matatk 2nd a bit more general, gather any particular accessibility barriers
Fazio create, based on the comparison between the groups, create a protocol
Fazio suggested a protocol timeline for responses
matatk we tend to leave issues until we get the person who raised the issue till they seemed satisfied. We can try to get a timeframe expectations as Fazio suggested. Just at the end of the process, not for all
matatk when somebody files an issue, there is a few different possibilities. Or more than this, like in which case we need to come back to them.
another option is that we decide we accept the feedback
matatk we can decide to make a change to the pull request. It would involve making a full request
We can give the person who raised the issue of the link to the discussion.
Fazio it is self explanatory. the group decide what we want and the reason, then we agree
janina1, Fazio got an important point
matatk we need to be ready for this eventuality, and if we get to that stage and the task force and then move. Person given feedback because they feel they have been heard, we need to be prepared for that
Fazio we should write everything in a protocol, we can have in a page a process documente, we should be aware of...
janina1 all of is absolutely critical but not sufficient. There is some assumptions. I f we really thought that we could do all of this asynchronously, it would be.
janina1 People have very strong opinions that come up, she doesn't believe that we need a process that mediates between them. we need to develop them. It needs to be a W3C process solution . The decision policy we have, an PA, id an outgrowth of HTML decision policy
janina1 The fancy process happended because there were disagreements
janina1: Normative document preparation and the communication that goes through. Got consensus around W3C.
Lisa GitHub is not accessible for COGA. Just a few people can menage comments and discussions being coded
Lisa only 25% of people in COGA are comfortable in using COGA
Lisa *only 25% of people in COGA are comfortable in using GitHub
Lisa People inside of W3C have barriers inside W3C, It will be limiting in the partecipation and consensus. It is a practical problem. They also made a e-mail list for commenting. Significant feedback from that.
Lisa what they also do is using Google Docs, redundancy of docs, process is just ridiculous
JenStrickland people feeling hurt is big deal as you were talking. the dynamics of W3C's shutdown' people. We need a protocol to let people learn how to listen, ask question, be curious, kind to find a process, a solution, more truly inclusive. As a global institution we have a variety of cultures, genders, a variety of... we need to figure out
how to truly democratize that. We need to documenting social clues, trying to be as cristal as we can, so people can feel heard
<JenStrickland> "people feeling hurt" s/b "people feeling heard"
<JenStrickland> We've had similar issues in the Sustainable Web Interest Group with new invited experts, they are struggling to participate in the W3C GitHub.
matatk close the topic before moving on to the next agenda. The AB is looking for advice on the process and this isn't what they expected, but they asked.
They need to hear it, so he would try and do whatever he can to encourage
matatk Tag design reviews, issues with comments in GitHub, joining in a call they fix the discussion and misunderstanding. Reopen some issues, there will be a 2.0 at some point, we need a first 1.1 first draft.
Fazio there are some misunderstanding, he doesn't think we need to change anything in the process. the journey in is mind is everything we make a 1.1. first draft. this assessment tool that we're talking about putting together until we get that. To statement status very soon, this is the progress. 1.1 for sure, maybe 1.2. W3C statement status,
we are going to explore more applications. Other groups like sustainability are using it. That should be a game plan. It helps us embed our work
matatk, Fazio we are going to support your vision
<Lisa> +1 need to get to issue papers , lol
janina1 she loves about the goal of W3C Statement
JenStrickland there are other maturity models and it is going to do without reinventing the wheel
COGA Issue Papers
<Lisa> https://
<Lisa> https://
<Lisa> https://
<Lisa> not in: full citations, credits, debugging Always can do with more edit, more reaserch
matatk janina1 there are few new people in the task force, COGA wants to published as notes, APA is happy to support them, to make the content usable and successful
For background: Making Content Usable for People with Cognitive and Learning Disabilities: https://
Lisa making content useful, it means to design it for the audience. People are looking for new solutions, or adapt. If you are a designer and you want to do more, maybe a new solution, policy making, funding allocations, or something like that for strategy as a document, it could be useful for researchers who are looking for ideas for their next.
We want to integrate with more plain language, we can work and integrate better together. They had an old template.
Could you have a different section or next issue papers, Sometimes is not testabe. https://
Lisa use the direct links to go to the new versions (see before in the chat).
The voice systems: Alexa, but also it needs to be voicemail, as W3C specification. For like voice menu, which people are doing. Has been included also mental health disabilities, there was a huge overlap. As well anxiety or illness, not be lower with dasabilities, even if it is temporary
Lisa ( she is eplaining all the contents of the page) User needs, user persona, looking for solutions, best practices. They are also reviewing specifications that this is relevant for, news need. A lot of these issues that they talk about is talking decision making. it is not guardianship, user are in control. for example, someone is manically
depressing. They might buy a very expensive drum kit because they've decided that they want to became a famous drummer. but then afterwards the bank account is repeated. They have debt for the next three years and they realize the do not have a good sense of beat.
Lisa sometimes people need a supportive decision-making. So the alternative is ask for support, like a parenting control. What about a financially abusive relationship? Be gving a mechanism for financial control. Look for priniple and some proposed solutions. On supported decision making as a feature and why. Sometimes they are limiting people and
not support them. So, what about decision making in W3C? they tried to extend it. In public private partnerships. Organisations that kind of thing where you want participation in a public forum.
Lisa online safety and well being, thet tried to include algorithms, data and social media, we have them in a swift in mental health
Fazio In Web ML, they are looking for this kind of info, because they are trying to understand at what point is ok for an AI agent to make a determination of what it is appropriate to make a decision safe for a user
Lisa in safety and well being, talking about hackers, there is a reason if there is a lack, like cybercrime. Privacy. Social Media, mental health. The algorithm are going to show more of the kind of things you click on. So if you are a bit paranoid, you became more anxious, isolated, it can be a bit addictive, also from a sense of safety. We are
touching on the representation of data set it is a basic thing in smart cities and this is decision making. An app can complicate or slow down. this is why thet did some user stories
Lisa wayfinding and orientation, for people who are problem orientating themselves inside the room or getting from one room to the bath. redirect or automatically change orientation, etc. also for the screen , new solutions and scenarios
matatk What are the next steps? What do need APA to do?
Lisa next week they will share this formally to APA. some bits are not in yet, like the course locations at the bottom. and they send it to get help for publishing.
janina1 asking for clarification about clear goals, title, abstract, etc
janina1 feedback: you can not talk to all the world for wayfinding, we need to be really specific
Lisa it is written in the introduction
janina1 there is a part on the why you wrote this paper?
janina1 the content usable, the use cases are extremely important, but we need to clarify this other way. Janina wants Lisa to succeed, reason why of all the feedbacks :)
Lisa the model paper covers safety issues, focs on users, also from crime, mental health, privacy, content usable, there are some safety issues, looking for clear advice
<Zakim> matatk, you wanted to mention user needs and reviews
matatk we want to provide user needs, to be sure that everything we provide is allign with them, so in this way we do not need to redo anything. make content usable,split it in different modules, how is that working?
Lisa the relationship between them, is to make the content usable, they did a lot of background readable and usable . adaptable colours, algo, increased anxiety. make it easy is a design thing. we can do a nice table to compare different options and other important info. there are som things that should be stress in making the content usable
Lisa content usable is going for web developers, we can have really good project to make the content more usable
<Fazio> I've created a document of 190 some odd testable statements from How To Make Content Usable
<Fazio> We're integrating it into our auditing tool
Lisa content usable is for people making the content usable. it should be tested , in particular for people that have disabilities. if you are looking for new ways to make your content better or safer whatever the topic is on. we've done background research, understand roadmaps, in particular for smart cities dev. What is happing today in our
community? maybe it is a science committee. People making strategies, or road map of where they go with their roadmaps
matatk I understand, though it is not because we are saying this is authoritative guidance for policymakers should be aware of. it is actually the opposite.
<JenStrickland> Something I really struggled with in the last few discussions here is the interrupting others and cross-talk.
Fazio document with around 190 testable statements from how to make content usable and is integratng it uditing tool. useful for COGA
matatk meeting about orizontal review + discoverable destinations about agentic AI. we have some issues about orizonral review
matatk discussion about scheduling meetings around the Thanksgiving period.
Traditionally, the week after the “tea bag” event is taken off, but this year APA will still meet on 26 November, while many task forces will decide independently whether to meet that week.
They also noted that non-American members may still be available to work or join activities.
Finally, they finalized the December schedule: APA will meet during the first three weeks of December, specifically on 3, 7, and 10 December
JenStrickland appreciate the emails sent out because even if she doesn't read them on Mondays, she tent to search for the before APA meeting
Roy_Ruoxi doesn not think that getting the email on friday will change much
<JenStrickland> matatk: there are two charters up for a review
@janina matatk
The group discussed how to better manage and communicate meeting agendas. They confirmed that:
Agendas will be sent out early Monday morning (US Eastern Time) so people have time to review them before the Wednesday APA call. Participants said these emails are useful and they often search for them before meetings.
Calls are considered on by default, and the agenda will explicitly mark the call as “confirmed.”
New TR items or topics can be added later without problems, since there are usually no strict deadlines; if something new appears, they can review it the following week or with less prep time.
They want more review time for new TR items and charters, and will keep the agenda updated accordingly.
The person posting the agenda will avoid including empty items, to keep things cleaner and easier to manage.
matatk W3C Interest Group is useful and how it differs from a Community Group, even if groups can turn in different kind of groups. What has been done is kept the group around the table and make everything visible, chose the right thing for "you" as a need/ group.
An Interest Group helps bring specific stakeholders “around the table” who might not otherwise participate, which is important for topics like user needs, challenges, and best practices.
A Community Group is open to anyone, while an Interest Group is more like a Working Group: you generally need to be a W3C member or an invited expert to join.
Because of this, an Interest Group’s structure and governance are more suitable for some member organisations, and its work can be given clearer direction
APA sessions at TPAC (and some related breakouts): https://
matatk what the APA group should be aware of? based on the TPAC experience
<JenStrickland> https://
JenStrickland Sarvin made a pres on quality assurance that can be intresting
matatk Servin is also on the tag, really intrested in keeping the web indipendent, also web of trust, being able to verify what people said. and he also something that APA would be intrested in
@janina there is a group in web space, call WaySpace
https://
@janina the main thing that we should do, goinge to fix accessibility behind the source layer. the ,amdtra has always been fixed. the source in the view of this group. Published a report just over the weekend, Her with Lionel_Wolberger and some co-authors are presenting tomorrow at 10 am. They want wider conversation way as to when and how it is
appropriate to consider beyond source remediation. they call it post source
@lionel fortunate in the collaboration of the Information Architecture Community Group on the preparation of that report
@janina make a document that is more of a note docuent, rewording some additional content
@lionel "do not bring me spare problems. If you bring me a problem, bring me a proposed solution" mantra proposed
JenStrickland some of us do not presume to offer a solution, we have humilty to think that we do not fully understand whatever caused the problem enough to do so
JenStrickland she was having a bit of hard time with the interrupting in the cross talk
Accessibility Maturity Model - next steps
matatk plan of AMM, documents and issues at high level + issue management. COGA has provided some tutorials about that to their members. being transparent and suppporting them in feeling heard + @janina w3c process policy, orientation. matatk promoting the notion that some of this need to be written
kevin consensus, main aspect, what does it mean and why , we should cover that
Fazio we can work through the issues on GitHub, with the aim of 1.1 as W3C statement, start from that and build something around it. people in WCAG3 were concerned about the improvement. yesterday has been done standard diagnostic. 1.1 fix GitHub issues
kevin he thinks that Fazio is a long way away from a statement
kevin considerations about WCAG are particularly around that. We need to afreed to kind of corpus of language and cordinate of your view
Fazio we are not COGA. we 've actually implemented an AI powered auditing tool that we believe can run these tests and come back. see the comparison and is not saying that as a document itself.
<Fazio> The plans I just laids out are in our last task force meeting minutes
@Janina doing things layer after layer,
JenStrickland set a frame for how we are having discussion. we are here to hear people first, it is important to set expectations
matatk we should define the process that we need to adopt, after we define the process to follow
<Fazio> I don't like the ay I am talked to about this work
<Fazio> Don't feel supported or respected!
matatk the process is really not accessible to tons of people in APA and COGA groups, we need to find a way to make things accessible to the groups first, also some people has some accessibility problems first.
Proposed process includes:
Reviewing feedback to determine if it is actionable.
If the task force decides to act on it, the issue will be labelled accordingly, and work will proceed through a pull request, allowing the person who provided the feedback to review the changes.
After review, several outcomes are possible:
The group and the submitter are satisfied.
The group is satisfied but the submitter feels the work doesn’t go far enough — in this case, the issue can remain open for further discussion.
The submitter disagrees with the direction taken — then the topic may be revisited.
The discussion emphasised transparency, collaboration, and allowing adequate time for responses within the feedback and review cycle.
kevin there is lots of reasons why there might be a misunderstanding. The goal is the other part of it. GitHub does what we needed to do to address the transparency and so forth
Fazio implemented GitHub classroom into curriculum and program, it can be created assignments and tutorials
@janina GitHub is more than a platform
JenStrickland GitHub classroom is more than the GitHub video/ training
<Fazio> Github Classroom assignment https://
JenStrickland Lisa can provide more details about the problems of COGA, code help to find the solution. the designer, dev, etc would like to improve it. collecting the feedbacks has been done before
<Fazio> Helix Opportunity's GitHub Classroom page https://
<kevin> WCAG 2 Backlog TF process
@janina process really intresting . CFC to understand how they pop. circumstances so we can nip that in the bud. having wider converstion about how we clean up our expectations of one another, how we define. to sum up we have processes that we implemented in various ways. Llisa explained coga process matatk we need to create a process for
listening as JenStrickland said
JenStrickland learn how to listen and hear one and another is really important, to understand when we are in hearing mode, and all the other. we take for granted how to do things. "this is the people that want to speak, this is..." we need to make super that we adopt what we are talking about. create the work easier for everyone to let them do a
great work
matatk plans for maturity models , how we think about feedbacks and how to handle them. topic of consensus
<JenStrickland> Neha might you jot that down here?
Neha map the documentations, to support people in being more effective. i gave a high level outline of the feedback process, but i used github terms to describe the bits. In the google docs proess, it looks like this. in the github process. it looks like this. they are the same matatk approve
matatk: So, Kevin, re consensus, what should we discuss?
Kevin: I think the gap from my perspective, within this particular example…
… consensus was being argued based on industry usage rather than working group members, and that's a problem.
… when what you're looking for is for the group to agree or agree to disagree on how a document or resource is developing. That's the step where I can see a problematic outcome where it doesn't respect the working group's members input.
matatk: Janina has mentioned consensus. We've discussed it. It is where everybody could live with it, then to find that not everyone could live with it.
In one line, map every step from google docs being followed for issue management to be mapped with github process for issue management, so it will be easy for people to learn how to use github
matatk: if it was entirely based on "this is how industry is doing it" then that would be bad. That was part of the feedback we received. I looked at how the task force reacted to the feedback and how they responded internally in the document.
… In the vast majority of cases, a reasonable discussion had occurred, and had they been communicated back according to the process we said we should have followed…
… that was part of the balance that I made.
… we had to balance the overwhelming majority that were very plus one, and the few that were minus one, and it was extremely difficult.
… We do care about the feedback, and I'm sorry that we didn't communicate that adequately at the time.
Fazio: Consensus, at what point do you determine you don't have consensus when people who are not part of the work and they object to things that have nothing to do with them.
… for example, we had one objection based on other people's issues.
… and if the minutes had been read, they would see that those issues were tabled to be addressed later on.
… so, what right do they have to object based on someone else's interpretation of someone else's issues?
… then, "this can't go into WCAG 3," and we as the AMMTF didn't believe they were applicable to the publication.
… maybe this was not well communicated, my catharsis…
… we looked at the objections because they didn't have a lot of merit based on the situation.
… the comment from three years ago, which had been addressed, but wasn't communicated, but was.
Neha: there is a gap in communication that things were addressed, but were not communicated.
matatk: Context: there was a lot of support for the AMM, but there was some concern about it being published.
… there were two objections, but it was published anyway.
@janina: I think we at the W3C has been spoiled in our consensus process. We have no mechanism… we started to in HTML5… but if we waited we would not have HTML5. It was a painful time and I hope not to revisit it here.
… consensus is not unanimity. There are times when we will not agree and still need to move forward.
… Thankfully we have a decision policy and followed to the best of our ability. I think it's to our credit that we forget that consensus is not unanimity.
… We must have good policies in place. It's not just AMMTF, COGA, …
… It came over time because of experience.
… I do think we need to say a bit more than the process doc says currently.
<Lisa> AG usealy do not have 100% consensus. I actually made a formal objected to wcag 2.0 and got 40 people/organisations signing with me
Kevin: Responding to Fazio, forgive me if I misquote. You spoke about the right of the people not in the room to respond minus one, they do have a right. They're part of the working group.
… and that's problematic, because it is clear that these people have perspectives and they have a valid right to reflect those perspectives and their contribution to be respected.
… I did say it can't be normative, I think you're very far from being a statement.
<kevin> Consensus Building
Kevin: In terms of if there's specific guidance on consensus building, what it is, how to develop, how to deal with dissent. there's a lot of info there, a lot of history.
… and the last one you pointed to was an issue that had been addressed, was there for three years, and there wasn't anything linked to the solution / commit / change, so the burden is put on the original commenter to figure out. That's why we aim for transparency and connect things.
… Without it, it's very difficult for original commenters to understand.
… otherwise we have these docs that aren't reflective of what the entire community sees.
{Sorry, I failed to document the last bit of discourse.}