08:32:59 <RRSAgent> RRSAgent has joined #acd
08:33:04 <RRSAgent> logging to https://www.w3.org/2025/11/12-acd-irc
08:33:04 <Zakim> RRSAgent, make logs Public
08:33:05 <Zakim> please title this meeting ("meeting: ..."), wendyreid
08:33:27 <wendyreid> meeting: Accessibility Compat Data Update
08:33:35 <wendyreid> chair: lola
08:33:52 <wendyreid> scribe+
08:33:53 <patrick> patrick has joined #acd
08:36:57 <wendyreid> lola: Thank you for attending, this is an overview session about Accessibility Compat Data
08:37:10 <wendyreid> ... I'm leading this endeavor, with the help of many
08:37:24 <wendyreid> ... I'm also co-chair of the TAG, and the W3C Docs CG
08:37:34 <wendyreid> ... I'm also an artist, I hate cheesecake, and I think you should too
08:37:39 <wendyreid> [general shock]
08:38:00 <wendyreid> ... This is an issue that Adrian Roselli opened in BCD
08:38:22 <wendyreid> ... the <mark>, <del>, <ins> and other elements were not properly vocalized
08:38:40 <wendyreid> ... [example] VoiceOver would only communicate part of a sentence
08:38:48 <wendyreid> ... only n issue in Safari
08:38:56 <ananya> ananya has joined #acd
08:39:06 <ananya> present+
08:39:08 <wendyreid> ... If I was a software dev, I might use CanIUse, and that would tell me these elements were well support
08:39:18 <wendyreid> ... and if I went to BCD, it would tell me it's supported
08:39:21 <wendyreid> ... baseline too
08:39:28 <wendyreid> ... available, supported, nothing wrong
08:39:33 <wendyreid> ... where would we submit this issue?
08:39:52 <wendyreid> ... BCD, Adrian opened them there, and it was decided it was out of scope because it was screenreaders
08:40:14 <wendyreid> ... because the issue wasn't in other browsers, it was determined to be the screenreader
08:40:17 <wendyreid> ... it wasn't known right away though
08:40:42 <wendyreid> ... Adrian was able to file the bug with the WebKit team, who determined it was a VO issue
08:40:44 <wendyreid> ... if it was a different combination, how would it have gone?
08:41:14 <wendyreid> ... If teams can't communicate, and Adrian knows these areas well, but how would a random developer approach this
08:41:25 <wendyreid> ... this is where something like accessibility compat data comes into play
08:41:47 <wendyreid> ... goals, define machine readable datasets for this data, get it into MDN, we want to make it searchable
08:41:57 <wendyreid> ... support developers to build accessible experiences, and support web users of AT
08:42:04 <wendyreid> ... hopefully it's obvious why it matters
08:42:23 <wendyreid> ... currently developers have no way to check if a feature works in a browser, there's no way to know what working means
08:42:38 <Daniel> Daniel has joined #acd
08:42:41 <wendyreid> ... tools show browser support via BCD or Baseline, but accessibility information is not included
08:42:48 <patrick> present+
08:42:56 <wendyreid> ... Baseline gets info from BCD, BCD doesn't check the a11y tree
08:43:12 <wendyreid> ... other projects are out there, like a11ysupport, [other name]
08:43:30 <wendyreid> ... some of these are closed source, some aren't usable for devs, some have different focus
08:43:41 <wendyreid> ... ARIA-AT was focused on APG, not other sources
08:44:01 <cyns> q?
08:44:13 <dom> dom has joined #acd
08:44:21 <wendyreid> ... ACD addresses these issues by identifying web features as they are developed, get into review at the right time
08:44:33 <wendyreid> ... immediately start planning what tests are needed in WPT or ARIA-AT
08:44:42 <wendyreid> ... I'm using my position on TAG to keep on top of what is happening
08:45:04 <wendyreid> ... Matt Atkinson, member of TAG and co-chair of APA, but that is not sustainable if I'm not on TAG
08:45:15 <wendyreid> ... we need a strucutured process with the horizontal review group
08:45:29 <wendyreid> ... want to make our tests part of WPT, the infrastructure is there
08:45:39 <wendyreid> ... it makes sense to contribute to what exists and take what we need
08:45:51 <wendyreid> ... we want to present the support data in places developers go
08:45:58 <wendyreid> ... hopefully this will help teams
08:46:13 <wendyreid> ... why now, I don't know how many have seen the MDN HTML 2025 survey
08:46:32 <wendyreid> ... under the education and learning banner, a11y was listed as a main pain point for develoeprs
08:46:38 <wendyreid> ... they need to understand a11y support
08:46:58 <wendyreid> ... when I was at Bocoup, doing a11y work and trying to figure out how to bring things to developers
08:47:08 <wendyreid> ... it's been a discussion for a long time
08:47:21 <wendyreid> ... a11y is a pain point for develoeprs and they are communicating that
08:47:38 <wendyreid> ... there is compliance pressure, it's not the main reason but it communicates priority
08:47:58 <wendyreid> ... we have EAA, other governments expecting accessibility
08:48:12 <wendyreid> ... WPT have a large test suite with tests that already cover soem things
08:48:16 <wendyreid> ... we have data out there
08:48:21 <wendyreid> ... we can bring it all together
08:48:52 <wendyreid> ... the ARIA-AT project, I worked on that at Bocoup, there was little interest in SR tests for WPT, but now there is interest where there wasn't
08:49:01 <wendyreid> ... how do we keep that conversation going
08:49:10 <wendyreid> ... the infrastructure needing to be build is minimal
08:49:25 <wendyreid> ... just to be clear, ACD is not an auditing tool, develoeprs can't use it to audit their websites
08:49:46 <wendyreid> ... there is a WebDX meeting later this week into considering ACD or other a11y data set into baseline
08:49:49 <cyns> q?
08:50:09 <wendyreid> ... we don't want ACD to be a baseline, we don't want people to assume they meet the criteria and its accessible
08:50:18 <wendyreid> ... its so you know what to use to build accessible experiences
08:50:42 <wendyreid> ... ACD [diagram], we pull from WPT and ARIA-AT to build compat data and send it to developers
08:50:52 <wendyreid> ... ACD feeds into MDN, IDEs, CanIUse
08:50:54 <wendyreid> ... who benefits?
08:51:10 <wendyreid> ... web users using AT will have more accessible experiences, room for improvement still
08:51:26 <wendyreid> ... web devs will have access to information on support features and problem areas in their codebases
08:51:35 <wendyreid> ... browswrs and AT vendors will see how their products perform
08:51:53 <wendyreid> ... browsers will see how their platform interacts with SRs and any gaps
08:52:12 <wendyreid> ... browsers are here, AT vendors are not, but there is no standard for them, standards bodies
08:52:34 <wendyreid> ... most ATs are OS-based, which is part of the complexity, we're web focused, but SRs consider the entire operating system
08:52:38 <wendyreid> ... what we need
08:53:03 <wendyreid> ... I've made a rough draft of a scraper that scrapes WPTs, insight into what is missing
08:53:03 <wendyreid> ... especially for older web features
08:53:27 <wendyreid> ... WPT is missing support for features and we need to make sure they are tests
08:53:34 <wendyreid> ... we're focusing on HTML AAM mappings
08:53:41 <wendyreid> ... many missing tests
08:53:55 <wendyreid> ... when I did the rough scrape, it looks like only 8 HTML AAM features have tests
08:54:02 <wendyreid> ... we need to write tests to fill in the gaps
08:54:07 <wendyreid> ... we also need better filtering options
08:54:32 <wendyreid> ... the way I am filtering is shaky, it cannot produce consistent results
08:54:40 <wendyreid> ... I'm pulling down WPTs, a test file can have multiple sub-tests, and how they categorize tests is not how we need
08:54:45 <wendyreid> ... we care about the subtests
08:55:21 <wendyreid> ... in one example, there is one computed name test, with subtests for elements, we would care about each element, then build out the further tests
08:55:29 <wendyreid> ... WPT doesn't have a way to filter on a subtest level
08:55:49 <wendyreid> ... I'm doing string pattern matching, it's flaky, they have a lot of contributors and no set naming conventions
08:55:56 <dom> q+ to compare WPT scraping to bcd-collector
08:56:12 <wendyreid> ... some tests have el-element-, and some don't, it's convoluted and unnecessary
08:56:25 <wendyreid> ... good news, Google and Bocoup are working on annotations to work on this
08:56:34 <wendyreid> ... better categorizing tests and cross testing
08:56:42 <wendyreid> ... thats a work in progress
08:57:06 <wendyreid> ... we also need SR tests, ARIA-AT is now looking to do SR tests for ARIA features
08:57:14 <wendyreid> ... we can reap the benefits of their work
08:57:17 <wendyreid> ... we also need funding
08:57:23 <wendyreid> ... we have had some funders
08:57:29 <mehm8128> mehm8128 has joined #acd
08:57:38 <wendyreid> ... a11yproject, open web docs, microsoft, mozilla, tetralogical
08:57:51 <wendyreid> ... Cynthia Shelly, we have support
08:57:54 <wendyreid> ack dom
08:57:54 <Zakim> dom, you wanted to compare WPT scraping to bcd-collector
08:58:18 <wendyreid> dom: I'm curious about the approach to rely on the WPT data for browser support, I know it was explored for BCD, the conclusion was it was too granular
08:58:35 <wendyreid> ... for what BCD needs, which led to the collector which does more targeted testing
08:58:46 <wendyreid> ... has that been explored as an alternative
08:59:14 <wendyreid> ... if you look at all the ways tests are failing for supported feature, the definition of interoperable vs how people actually view it
08:59:19 <wendyreid> ... will you hit the same?
08:59:20 <kadirtopal> kadirtopal has joined #acd
08:59:24 <kadirtopal> q+
08:59:28 <wendyreid> lola: ACD is significantly more complex, we may hit this
08:59:43 <wendyreid> ... BCD is only looking at the DOM tree, don't need the granularity
09:00:00 <wendyreid> ... with ACD, we need to test the a11y tree, do certain things happen, do properties change.
09:00:12 <wendyreid> ... if we add aria-label to an element, does the computed name chage?
09:00:27 <wendyreid> ... we can't do the checks like BCD does, with WPT we can pick and choose tests
09:00:35 <wendyreid> ... we can use the tests that make sense for us
09:00:51 <wendyreid> ... we'd end up replicating WPT if we didn't use it
09:01:00 <wendyreid> dom: ???
09:01:15 <wendyreid> lola: Maybe we'll have something ready
09:01:21 <cyns> q?
09:01:37 <wendyreid> ???: Filtering, with ACD we're looking for info that is in WPT already
09:01:39 <dom> s/???/do you think WPT + annotations will make it possible to make that filtering robust?
09:02:08 <wendyreid> kadirtopal: WPT-related, with WPT are in the browser, what else is needed to complete this work
09:02:17 <tzviya> s/???:/ florian
09:02:17 <dom> s/???/florian
09:02:30 <wendyreid> ... are WPT tests mostly testing things in the a11y tree?
09:02:46 <wendyreid> lola: So both, there is an interop focus area and just deals with a11y
09:03:05 <wendyreid> ... they are focused on WPTs focused on accessibility, we'd combine those tests with the general tests
09:03:18 <wendyreid> ... the existing tests, aside or address, they just test that it works
09:03:31 <wendyreid> ... the a11y tests look at computed name and so on
09:03:38 <wendyreid> ... what else is needed? ARIA-AT
09:03:49 <wendyreid> ... some kind of screen reader testing section
09:03:58 <wendyreid> ... WPT's automated testing, SR testing is not there yet
09:04:00 <wendyreid> q+
09:04:04 <wendyreid> ack kadirtopal
09:04:16 <dom> q+ to ask about assistive technology support beyond screen reader
09:04:26 <wendyreid> lola: ATDriver is a new spec for automating AT testing
09:04:53 <cyns> q+
09:05:05 <wendyreid> ... I'm hoping screen reader testing can be automated because its expensive otherwise
09:05:07 <Daniel> q+
09:05:11 <dom> -> https://w3c.github.io/at-driver/ AT Driver
09:05:22 <kadirtopal> q+
09:05:26 <dom> q- later
09:05:26 <wendyreid> ... in this case specifically SRs, if we can put energy behind this work
09:05:45 <wendyreid> ... on Dom's point on the granularity, ARIA-AT is also very granular and we don't need it all
09:05:59 <wendyreid> ... part of AT Driver is looking at intents and more than what is vocalized
09:05:59 <keithamus> keithamus has joined #acd
09:06:14 <keithamus> q?
09:06:16 <wendyreid> ... JAWS, VO, NVDA, ACD doesn't care about how, only does it happen
09:06:16 <keithamus> q+
09:06:45 <wendyreid> florian: another thing that is needed is filtering the AT driver work, ACD format making choices for what do we test, create results
09:07:19 <wendyreid> lola: Exactly, Florian has opened a PR on the repo, what the JSON could look like, allows for facilitation and integration into the tests
09:07:29 <wendyreid> ... we want to use the same shape and testing formats
09:07:36 <wendyreid> florian: make it easier to adopt
09:07:49 <dom> q?
09:08:12 <wendyreid> cyns: There's also the acacia work Valerie is doing, there are layers, helps pinpoint issues, the other thing we need is writing bunch of tests
09:08:12 <dom> ack wendyreid
09:08:14 <dom> scribe+
09:08:15 <cyns> ack me
09:08:55 <dom> wendyreid: my team at ebay was looking at taking over allysupport
09:09:22 <dom> ... I'm wondering if these tests that are valuable and good could be ported into ACD
09:09:35 <dom> florian: I've been looking into porting some of that data
09:09:44 <dom> ... likewise for the data from Tetralogical
09:09:55 <dom> ack Daniel
09:10:16 <wendyreid> Daniel: Several things, according to what MAtt shared in ARIA WG, we're closer to better automation
09:10:18 <kadirtopal> q-
09:10:24 <wendyreid> ... there's work in NVDA, in JAWS, Applescript
09:10:24 <dom> Daniel: there is a public in nvidia able to run the automated tests, work to include it in jaws
09:10:33 <dom> ... and work with applescript for voiceover
09:10:39 <dom> ... all of that is already happening
09:10:53 <dom> ... re high-level approach, we should be careful in not missing key details
09:11:13 <dom> ... e.G. moving from one heading to the next without moving scrolling will have an impact on what gets rendered
09:11:36 <dom> florian: when designing the ACD data format we were exploring what to express exactly
09:11:50 <dom> ... not just a binary supported, but a list of potential issues (without expressing "it works")
09:12:22 <dom> lola: a lot of the data that exists on screen reader has "supported", "unsupported", "partially supported" because there are nuances
09:12:26 <dom> ... some depend on user settings
09:12:39 <shawn> present+
09:12:41 <dom> ... some are screen reader dependent based on product decision
09:13:05 <dom> florian: it's also that exposing differences might help convergence among screen readers
09:13:07 <kadirtopal> present+
09:13:37 <dom> daniel: my worry is throwing away useful data towards that "high level" approach
09:14:00 <dom> cyns: it's also about writing good test cases - e.g. on headings, make sure it reflects the scrolling situation
09:14:22 <cyns> q?
09:14:30 <dom> lola: part of the challenge is keeping this interpretable by developers who already struggle with the level of complexity in the much simpler BCD
09:14:36 <wendyreid> ack dom
09:14:36 <Zakim> dom, you wanted to ask about assistive technology support beyond screen reader
09:14:47 <wendyreid> dom: You partially addressed this, you talked a lot about SRs
09:14:49 <wendyreid> ... not the only AT
09:14:58 <wendyreid> ... I wonder how you manage extending to other ATs
09:15:11 <wendyreid> ... or manage the more political aspects of focusing on one
09:16:00 <wendyreid> lola: Follow on ARIA-AT's approach, it feels like a place to start, SRs communicate to other ATs like braille readers, we may get some benefits along the way, with the goal of expansion
09:16:11 <wendyreid> florian: Start small, then we can grow the dataset
09:16:18 <wendyreid> lola: See the shape of the data
09:16:27 <Daniel> q+ to ask if links to the relevant materials have been posted to IRC
09:16:28 <wendyreid> cyns: With the AT Driver spec, it helps there
09:17:11 <wendyreid> vadim: in the beginning, you mentioned the bug, there are also unsupported features and ones deliberately not supported
09:17:16 <wendyreid> ... browsers are opinionated, have reasons
09:17:25 <wendyreid> ... SRs are much more opinionated, and much different
09:17:49 <wendyreid> ... there are already three levlels of not supported, and maybe a fourth, the user intervention of disabling or modifying features
09:17:55 <wendyreid> ... how you you plan to suport
09:18:00 <wendyreid> lola: Great questions
09:18:08 <wendyreid> cyns: SR testing are done with defaults
09:18:19 <wendyreid> lola: Even ARIA-AT keeps to defaults, and modes
09:18:35 <wendyreid> ... how we manage the three levels, that is where communication comes into play
09:18:59 <wendyreid> ... to understand the decisions behind the data, we need to build relationships with the affected parties
09:19:04 <wendyreid> ... we have relationships with the browser
09:19:12 <wendyreid> ... if we can find out that something is intentional
09:19:16 <wendyreid> ... we can document it
09:19:22 <cyns> q?
09:19:42 <wendyreid> vadim: WebEngines Hackfest, session on automated SR testing, they have minutes and slides
09:20:04 <wendyreid> lola: There's acacia, funded to work on it, it'll have a big impact
09:20:35 <wendyreid> ... the a11y tree lives in the browser, it communicates with 4 different platform APIs, unique to each OS, which goes to the AT, to the user
09:20:47 <wendyreid> ... acacia is looking at the relationship between the a11y API and platform API
09:20:53 <wendyreid> ... funding right now is only linux
09:21:06 <dom> ack keithamus
09:21:33 <wendyreid> keithamus: Had related thoughts, this diagram is helpful, it's true, browsers consist of different adapted libraries to different messages to different services
09:21:40 <wendyreid> ... Acacia's aim is to test those things
09:21:56 <wendyreid> ... on windows you do foo and macos does bar, and the SR does what it does
09:22:01 <wendyreid> ... what the SR picks up
09:22:17 <wendyreid> ... I'm curious wirth combos like browser + OS + At
09:22:54 <wendyreid> ... depending on the granularity of data exposure, we can go as detailed as specific combinations, it can be false on multiple factors
09:23:08 <wendyreid> ... it's only green when all the factors are true
09:23:20 <wendyreid> ... currently ARIA doesn't specify mappings for Android
09:23:24 <wendyreid> ... plus plus plus
09:23:34 <wendyreid> lola: we haven't even started talking about mobile
09:23:55 <wendyreid> florian: we've had a hard time figuring out the gaps in data, and where in the traffic light things fit in
09:24:22 <wendyreid> lola: The knowledge of the ecosystem isn't even a guarantee of understanding
09:25:01 <wendyreid> cyns: Some of the value of the work, same WPT test, if you have all these areas testing the different points, if all pass awesome, if one fails, finding where it fails, great
09:25:10 <wendyreid> ... right now it happens in slack or over beers at TPAC
09:25:15 <cyns> q?
09:25:35 <wendyreid> keithamus: To conclude, if we have Acacia testing platform, and AT Driver testing the inputs to the AT, maybe we can connect the dots, we need that
09:26:14 <wendyreid> ... there's two tests if we know the two cases, if one fails or is wrong, you can find the differences, more tests in one layer
09:26:36 <wendyreid> Daniel: Just to say, if there are relevant links to share, put them in IRC please
09:26:39 <wendyreid> ack Daniel
09:26:39 <Zakim> Daniel, you wanted to ask if links to the relevant materials have been posted to IRC
09:26:53 <jgraham> RRSAgent: make minutes
09:26:54 <RRSAgent> I have made the request to generate https://www.w3.org/2025/11/12-acd-minutes.html jgraham
09:27:01 <wendyreid> lola: Thanks everyone, this has been an amazing discussion
09:27:01 <wendyreid> ... please keep up with what we're doing!
09:28:07 <lola> lola has joined #acd
09:28:51 <lola> aria-at: https://aria-at.w3.org/reports
09:29:52 <lola> lola has joined #acd
09:30:05 <lola> acd: https://github.com/lolaslab/accessibility-compat-data
09:30:16 <wendyreid> rrsagent, please generate minutes
09:30:17 <RRSAgent> I have made the request to generate https://www.w3.org/2025/11/12-acd-minutes.html wendyreid
09:30:32 <wendyreid> zakim, end meeting
09:30:32 <Zakim> As of this point the attendees have been ananya, patrick, keithamus, shawn, kadirtopal
09:30:34 <Zakim> RRSAgent, please draft minutes
09:30:36 <RRSAgent> I have made the request to generate https://www.w3.org/2025/11/12-acd-minutes.html Zakim
09:30:42 <Zakim> I am happy to have been of service, wendyreid; please remember to excuse RRSAgent.  Goodbye
09:30:43 <Zakim> Zakim has left #acd
09:30:52 <wendyreid> rrsagent, bye
09:30:52 <RRSAgent> I see no action items