Meeting minutes
WCAG 2.x issues
Chuck: welcome, everyone
… some housekeeping first
… first, introductions
… any new people or new roles?
Chuck: second
… next week, we’ll start call 30 minutes early for onboarding
… feel free to join
… it’s a quick whirlwind about what we do, how we make decisions
Chuck: third
… regarding ACT rules
<Chuck> Summary Element has non-empty accessible name: w3c/
Chuck: there was a rule proposed
… please review the link above
<alastairc> It's also fine to go to the onboarding if you've been here a while and need a refresher, or just have questions. No dumb questions rule in place...
Chuck: there will be a lot of emails going around about this, please review
… if anyone has concerns, please raise them
Rachael: reminder that TPAC is coming up
Chuck: any new topics?
… for future agendas?
WCAG 2.x issues
<alastairc> https://
alastairc: if you have a W3C GitHub-connected account, the link above should work
<julierawe> Sorry, joined late. Can someone please reshare the link?
alastairc: issues sent to the AG group are in the “Sent for WG approval” column
https://
… we are looking for a thumbs-up emoji reaction; no need for a formal PR review
… that will count as support
… we’ve got quite a few in this status presently
… most of these are straightforward
… we’d be grateful for a quick look
… there will be more coming through tomorrow
… if anyone is interested in contributing, please join the Friday call
<bbailey> https://
<bbailey> email on last set
<bbailey> Re: WCAG 2 proposed changes (was review by Aug 18, now Aug 25)
mbgower: a mail should have gone out yesterday with proposed changes
… for review by Sep 8
<bbailey> https://
<bbailey> WCAG 2 proposed changes (review by Sep 8)
mbgower: I incorporated everything that had support
… please review within in the next 2 weeks
… pretty light, with the exception of some changes kenneth put through
… but are essentially housecleaning
… each PR has an explanation
Chuck: any questions?
Publication Preparation https://www.w3.org/wbs/35422/aug-pub-feedback/
Chuck: moving on
… publication preparation survey
… going to share my screen
… our August publication
… earlier in a prior meeting
… we’d discussed ways of reviewing the content
… and then put out this survey asking people to exercise those questions
… to determine whether anything needed revising or removing
… our results are
… very few
… no substantive changes
… so we can use this time to have a quick conversation to confirm and get approval to publish
… any final concerns?
<Chuck> Draft RESOLUTION: Move the August 2025 WCAG 3 draft to publication.
Chuck: here is a resolution for the group
<Ben_Tillyer> +1
<sarahhorton> +1
Chuck: please vote
<Rachael> +1
<Chuck> +1
<Adam_Page> +1
<julierawe> +1
<BrianE> +1
<mfairchild> +1
<Gez> +1
<Makoto> +1
<Jennie_Delisi> +1
<kirkwood> +1
<Charu> +1
<Azlan> +1
<alastairc> +1
<bbailey> +1
<GreggVan> +1
<filippo-zorzi> +1
<mike_beganyi> +1
<ShawnT> +1
<Laura_Carlson> +1
<stevef> +1
<kevin> +1
<todd> +1
<mbgower> 0 (been away, haven't reviewed_
Chuck: +1 means you support, 0 means you’re ambivalent, -1 means you have a concern
<jtoles> +1
RESOLUTION: Move the August 2025 WCAG 3 draft to publication.
<hdv> +1
Chuck: this is the last opportunity for significant concerns
bbailey: does a survey response count as a +1?
Chuck: no, the survey was just an opportunity to weigh in with feedback
… we strongly encourage participation
<bbailey> +1 to the current survey lite approach
<Rachael> +1
<scott> the survey didnt' make much sense to me as to what should have been done with the survey - or if content should have been made in the github thread?
alastairc: a note for the editor’s: we can also include some simple multiple choice options, such as “I’ve reviewed and have no concerns”
… for future surveys
<bbailey> i reviewed and it was great !
Chuck: we’ll move to CFC next
<scott> it didn't seem large enough for feedback about the spec. but didn't seem to imply that the github PR was where feedback should be?
Chuck: where there will be a final opportunity for people to express support
Foundational/Supplemental Requirement Level Exercise https://www.w3.org/wbs/35422/requirement_levels/
Chuck: our other survey was to express support for what level certain requirements should be
<alastairc> scott - feedback in either is fine, some people struggle with github (and possibly vice-versa)
<bbailey> https://
Chuck: foundational, supplemental, or something else
… there are 7 results
… the first one was for “Captions prerecorded” — everyone agreed it should be Foundational
… if you have thoughts now, please queue up
<scott> alastairc as separate issues, or in the PR linked in the survey email?
Rachael: the results of today’s conversation will not affect the current draft, but the next one
<alastairc> We are trying to answer: "Criteria for placing a requirement in foundational or supplemental", and "Is foundational sufficient for base conformance or part of base conformance?"
Chuck: next req was “Captions live”
… 3 said Foundational, 2 said Supplemental, 3 said something else
… ripe for debate
… bbailey suggests WCAG 3 need a quality metric for captioning
… would you expand on that?
bbailey: I do think it’d be Foundational
… as with audio description, trying to treat this as binary is not working
<Zakim> alastairc, you wanted to comment on taking the "feasibility" as a thing. Also, whether it is best separated from pre-recorded?
alastairc: for this req in particualar, it will help this group to decide how feasible it is with what authors currently have available
… would other reqs for pre-recorded versus live be different
… we have a bit more nuance in WCAG 3
<bbailey> contradicting myself just now, maybe since captioning and AD is so variable, then maybe its Supplemental (per our definition of distinction between Foundation/Supplemental).
alastairc: do captions exist? and bbailey suggests the quality metric — would those be different
… for example, timing/synchronizations also needs to be sensible
<Ben_Tillyer> Apparently so
<alastairc> I haven't tried live captioning, beyond what I've seen at events, how easy is it? And what is the quality difference?
Rachael: I believe this should be merged with pre-recorded captions; they’re not that different as far as feasibility
… and then could start doing methods to distinguish
… these should be together as Foundational, and then there could be Supplemental to cover _quality_ of captioning
Chuck: I thought of this as Supplemental
<bbailey> +1 to Rachael point
Chuck: and should be distinct from pre-recorded
… but not a strong opinion
… I think keeping them separate affords us opportunities to treat them with some level of difference
<bbailey> FCC quality regulation: https://www.ecfr.gov/current/title-47/part-79/section-79.1#p-79.1(j)(2)
Chuck: but I am intrigued by Rachael’s proposal
<Ben_Tillyer> Trying to work it out
<bbailey> 47 CFR 79.1(j)(2)
<Ben_Tillyer> Ah, 10 secs!
<jon_avila> Live captions are part of AA today. Moving to supplemental seems to say we no longer believe the previous requirement is achievable.
bbailey: I posted a link to FCC quality metrics
<julierawe> +1 to jon_avila
bbailey: my concern is that those are more performance metrics
Ben_Tillyer: I think pre-recorded and live should be combined
<bbailey> https://
Ben_Tillyer: one caveat
… there are some corporate environments with digital conferencing
… that use hardware + software combinations where it’s not possible to live caption
… where the audio stream is completely separate from the video stream
… that could cause undue burden
<jon_avila> I think they should be separate. For better understanding user impact and potential scoring. More granularity has befits to knowing the impact.
Ben_Tillyer: but for either live or pre-recorded, the traditional tools — Teams, Zoom, etc. — are incredibly good for someone that has a good quality microphone
… good quality internet connection
… and an accent that isn’t too strong
… and that doesn’t change whether it’s live or pre-recorded
… the only benefit in some tools is that pre-recorded can produce slightly higher quality captions
… so I vote combined for presence, separate for quality
… Foundational for presence, Supplemental for quality
<alastairc> +1 to combining them for "presense of", and then separate for quality measures.
<bbailey> +1 to merged version as foundational
graham: there are subtle differences: music playing in the background, speaker names
… are we structured to merge
Rachael: I don’t think there’s anything in our structure that would prevent this
<scott> +1 to foundational for live captions. don't have a strong opinion on merging pre-recorded and live.
Adam_Page: I generally support foundational combined for live and prerecorded and supplemental for quality
<Rachael> +1
Adam_Page: Requiring the presence should characterise some minimally viable quality as well
graham: we need exceptions to be scoped
… for circumstances where authors don’t have control
<Zakim> mbgower, you wanted to say maybe categorize by features as opposed to accuracy?
mbgower: thinking about how this is dealt with in 508
… they specify features
… ability to control size
… font characteristics
… quality of the UX more than the quality of the captions themselves
… easier to measure those things in isolation
… the quality of the captions can be so variable
… difficult to quantify
… versus changing background color, etc., easier for authors to pull of consistenly
<jon_avila> Latency seems measurable.
Chuck: one thing I like about keeping them separate is that we could introduce a quality component to the pre-recorded at a Foundational level that we might not be able to do for live at a Supplemental level
<graham> +1 jon
Chuck: don’t know if they mix very well
GreggVan: pre-recorded can be done accurately by anyone, it just takes time
… and can also be checked for quality
… live, you don’t have that
… if they’re mumbling or using jargon
… asking for accuracy with pre-recorded should be uncontroversial
… but for live, I don‘t think there’s anything we could really require
… other than as an aspiration
… and it’s very expensive
… with AI, it becomes less expensive, but still defeatable
Rachael: we’re diving into writing new requirements for this particular area
… but our goal with this exercise is to think through *how* we will classify these
… it sounds like maybe we’ll have 2 Foundational requirements and then many Supplemental
<Zakim> mbgower, you wanted to say are captions so mainstream that we need to worry less about measuring quality for accessibility per se, and if not, are there specific considerations for accuracy we focus on?
mbgower: are captions fundamentally different in how they are broadly used by the public?
… in that they are widely used by a lot of people
… therefore, accuracy is so much more obvious to everyone
… do we have research
… on how quality affects someone who is Deaf or HoH versus others
… are there visual cues we could use to signal to users low confidence in a particular word or transcription
<CarrieH> my two cents, why someone may need captions not result of hearing loss? Auditory processing disorder and/or auditory processing lag
mbgower: we did work on this many years ago and the field has changed, but still could be applicable now
<Zakim> Chuck, you wanted to say the difference in required and supplimental
<Ben_Tillyer> I'm already planning out an app that does that on my notepad mbgower...
<Ben_Tillyer> What an ace idea
Chuck: bringing this back to Rachael’s point
… we want to think about how we choose Foundational versus Supplemental
… for this particular issue, I do see a problem with merging pre-recorded and live
<CarrieH> some people have difficulting digesting verbally or have delays with processing, and captions do help (recordings are better)
<Rachael> note: Key takeaway from this is that feasibility is important in assigning these.
GN015: I’ve also been informed by colleagues that the nature of the media matters a lot
… if it’s a movie, quality matters very much
… if it’s a live talk, they say they don’t need punctuation, they can guess what’s been said, etc.
<kirkwood> live and asking questions is a very good point
GN015: pre-recorded and live could have very different quality requirements from a user perspective
kevin: bringing this back to how to think about whether something is Foundational or Supplemental
… the fundamental user need here is the same
… someone who needs captions needs captions, whether live or pre-recorded
… that was my mindset in proposing merging the two and classifying them as Fundamental
<kirkwood> very differnt technology, very different level of effort, very different process.
kevin: distinguishing between the 2 is more of a policy choice
GreggVan: I think we need to keep them separate
<kirkwood> against merging. +1 to keeping separate
GreggVan: the user need is the same, yes, but the practicality of the solutions are totally different
Ben_Tillyer: I argue the opposite
… in WCAG 2.X, what I’ve seen is that people see the guideline called “Captions Pre-Recorded”, looked at what needed to be done there, and then didn’t read further
… and wrongly assumed that Live captions were AAA
… we have caused that problem in the past *by* separating them
… combining them solves that problem
… GreggVan is right about the practicality at this time, but the advancement of LLMs is rapidly changing that
<Zakim> alastairc, you wanted to say why not both?
<Zakim> Chuck, you wanted to move us to the next requirement
Chuck: important to keep exploring how we will generally choose Foundational versus Supplemental
Chuck: moving on to audio description (AD) pre-recorded
… 3 for Foundational, 3 for Supplemental, 1 for other
<bbailey> https://
Chuck: julierawe voted Supplemental because it is very difficult to do with current technology
<julierawe> I'm here but have nothing to add to my comment--unless folks on this call are aware of tools that can help with the problem I stated?
<Zakim> graham, you wanted to say foundational if new method is not "fail / pass" for WCAG as previously discussed, otherwise supplemental
graham: this is foundational if we’re not going with Pass/Fail
… if we go with Pass/Fail, then this needs to be Supplemental - because it’s too onerous
<Zakim> Ben_Tillyer, you wanted to talk about audio descriptions
<GN015> Julie, I see your point and agree it is a pain.
<Laura_Carlson> YouTube supports alternate AD soundtracks, if you have access to multi-language audio, a new feature that YouTube is rolling out.
<Laura_Carlson> https://
Ben_Tillyer: re: different versions of videos for AD
… I’ve come across this problem at non-profits with no budget
… sites like YouTube don‘t require you to upload an entirely new video
<alastairc> I think, in reply to Graham, that foundational is binary pass/fail, then supplemental is where people can pick which criteria they use to score more.
Ben_Tillyer: you can do it as an audio track, and it’s presented as a different language
… e.g., “English” versus “English (audio described)”
… however, we decided to produce the videos in such a way that the original audio track already provides the essential information
<graham> Thanks @alastair, in that case this needs to be suplemental as it is not feasible for many organisations.
Ben_Tillyer: and so AD was no longer necessary
<Zakim> bbailey, you wanted to mention that i know of no quality metrics
bbailey: quality of AD is even trickier than captioning
… much more of an art form
<Adam_Page> +1 to bbailey
<bbailey> https://
Rachael: what the group has decided is that all Foundational be required
<Charu> +1 to bbailey
Rachael: and some subset of Supplemental
… but haven’t decided how big that Foundational group will be
… the goal today was to talk through the rationale for how we place things
… for the next step of deciding
… we also have not decided how Supplemental will be selected
<bbailey> American Council of the Blind Audio Description Project https://
Rachael: chair hat off, I do support AD being in supplemental
… because the impact of descriptive transcripts is bigger; it covers the deaf-blind community as well
… so Supplemental, but only if descriptive transcripts are Foundational
<Zakim> mbgower, you wanted to say the big problem with audio descriptions is the lack of foundational support
<Adam_Page> +1 to Rachael
mbgower: the real assessment for providing information on video content is almost entirely reliant on AD right now in WCAG 2.X
<jon_avila> We need something at the foundational level. It is a huge user need to be synchronized.
mbgower: the more foundational requirement is that I can *read* the information in text of what is conveyed in the video
… if you consider feature-length Hollywood videos versus shorts on the Internet
… it’s not that difficult, even for AI perhaps, to describe some visuals usefully
<jon_avila> Ai can create audio descriptions
mbgower: we have no requirement for that right now; it’s AD or it’s the kitchen sink
… from a practical perspective, we’ve had no success saying AD is foundational
<jon_avila> Audio descriptions is a current AA requirement
<bbailey> +1 that quality text alternative can be preferable / better than AD
graham: the more I think on this, this one would be Supplemental *assuming* we require descriptive transcript
… thinking of a TED talk, with slides on the screen, where describing the slides are a bare minimum
… in that case, AD is supplemental
<alastairc> There is a current requirement (I think foundational) for "A transcript is available whenever audio or visual alternatives are used."
<Zakim> alastairc, you wanted to comment on AD being foundational or supplemental, and it seems to depend on other alternatives.
<Ben_Tillyer> graham - alternative *synchronous* alternative to the AD?
graham: and AD would possibly be a bad fit for describing the slide content anyhow
<Ben_Tillyer> (e.g. comparing captions to transcripts)
<jon_avila> Seems like there might be a misunderstanding of what audio description is.
alastairc: I think where we’re landing is that if there’s another requirement that’s more basic — a sufficient text alternative — then I would support AD being supplemental
<bbailey> +1 that text can also very nicely augment weak AD -- but 2.x SC don't well promote this sort of approach
<Zakim> Chuck, you wanted to ask for a scribe change
<graham> Ben_Tillyer I think that solve is one of the closest we have currently
thank you, sarahhorton!
<jon_avila> https://
GreggVan: Using AI for pictures, powerful, not for audio
GreggVan: Audio description is art, not everything on screen is important
<Ben_Tillyer> GreggVan I used Gemini 2.5 Pro to diagnose my car's engine from the sound alone after querying it at the weekend...
<Ben_Tillyer> It's getting better!
GreggVan: have to work within gaps, timing, AI can't yet, maybe someday
GreggVan: have to work within gaps, timing, AI can't yet, maybe someday
<kirkwood> author should approve it though
<mbgower> +1 to Jon
jon_avila: people are using AI for audio description
<kirkwood> it does hallucinate, though
<Laura_Carlson> +1 to Jon
jon_avila: Transcript might work in some situations, missing synchronization
jon_avila: Not filling every pause, not describing every details, essential equivalent is what's needed
<Adam_Page> +1 to jon_avila
jon_avila: Need some level of equivalence at the foundational level
<Ben_Tillyer> +1 to jon_avila
<bbailey> +1 to jon_avila
GreggVan: +1 to comments, need to separate movies and be my eyes when talking about audio description
GreggVan: Talking about audio description of movies
GreggVan: Valuable for everyone, learn more by listening to AD
<Jennie_Delisi> +1 to Gregg - cognitive supports sometimes via audio description
<Chuck> +1 to putting a pin in this one
<bbailey> +1 to everyone here using AD (and captions) whenever you can
GreggVan: Recommend putting on hold, might have tools upcoming at end of our process
GreggVan: All agree on need, discussing practicality
<bbailey> +1 to focusing on need versus how easy
<Zakim> Chuck, you wanted to say that it is ok for us to have different opinions, and that we are exploring how to think about foundational and supplimental
<GN015> IN fact, I benefit from sign language, though I actually do not understand most signs. IN movies, I have seen sign language interpreters acting, this way the tone of voice is visualized, and it was easier to grasp who is talking.
Chuck: Strong agreement, expressed in different ways, tasks is to explore what's foundational and supplemental
<GreggVan> +1 to chucks comment
<Zakim> alastairc, you wanted to focus on the scenarios aspect of what Jon said, if we can't say "for each video", then it isn't foundational.
<kirkwood> What if it’s a safety video?
alastairc: Discuss what's appropriate for different scenarios, need to work out caveats or needs to be supplemental
alastairc: Pin at level at now or guess about the future
Adam_Page: +1 audio description foundational, was involved, got a vendor to help, was staggeringly affordable
<Laura_Carlson> +1 to Adam
<kirkwood> Good point Adam
Ben_Tillyer: Caveats should be type of sync'd audio/video, blanket term included all would include Zoom and everyone's videos
Ben_Tillyer: Scoping applicability is key, needs to be foundational, people not able to participate in work and life
<kirkwood> of course, Mike. That does not solve ror incorrect information in description causing safety issue (incl death).
Ben_Tillyer: Having alternatives to AD, has tried scripts but not always possible, some things are unscripted
Ben_Tillyer: Foundational, defined scope, defined alternatives, amount of information intended to get through AD
<Zakim> mbgower, you wanted to say that we actually have a Get Out of Jail Free Card with https://
mbgower: Already talking head video doesn't need AD, no additional information
<alastairc> We have clients with social-media output and I honestly can't see how they could incorporate audio-desc without it being purely automated.
mbgower: Working group decided it was okay, explore those edge cases more, insist that's the info that must be there
<Adam_Page> +1 to mbgower to that odd loophole in 2.X
<Ben_Tillyer> +1 to mbgower - "air quotes" in a talking head video comes to mind
GreggVan: What if there is no audio gap, should that pass? If all gaps, pass; if no gaps, pass
GreggVan: Nothing useful to put in gaps, hard to figure out how no useful information
<kevin> -1 to no way to pass if no gaps, pausing can be used to add gaps
GreggVan: Need clarity to keep from endless discussions
<alastairc> I'm concluding from this one: Feasibility is important, but there are a lot of scenarios and in some cases an alternative to the video might be a better option than AD.
<Zakim> Chuck, you wanted to introduce comparable keyboard effort
Chuck: Comparable keyboard effort, story, used airline website to demo comparable effort, loads of keystrokes
Chuck: Audience counted, around 150 keystrokes to get to common item
Chuck: Experience indicates it can be pretty important
Chuck: Reads the survey responses (very quickly)
GreggVan: Anyone who is a keyboard user, want to have session with input group to learn about keyboard users
GreggVan: Refer to GreggVan and bbailey
GreggVan: What's comparable? Is 12 keystrokes comparable to 2 mouse clicks? Needs to be reasonable, not comparable
<Ben_Tillyer> PgDwn? :D
<Chuck> +1 to gregg, I see the problems to requiring this be foundational.
GreggVan: If required, has to be true of all sites, or all sites except for conditional requirement
<Chuck> This may not be "feasible" to make it foundational.
GreggVan: If memorize commands for 1 page, only works on one page
GreggVan: Not foundational or supplemental, would be best practice because can't figure out comparable
GreggVan: Modes are different: one is serial, one is random access
Rachael: Would be good to have small foundational set, tie all quality items into modules or grouping in advisory for policy makers
Rachael: Can we put things in supplemental that policy makers can use
<Zakim> alastairc, you wanted to comment on it being an assertion, and we should expand it out a bit to involve an optimization process.
alastairc: Assertion, not requirement — process
alastairc: Certain tasks, drawing, moving things, will take longer with keyboard than direct input method
alastairc: Want to encourage people to think about what would make an interaction more usable using a keyboard
alastairc: Get people to try, think about it, as an assertion
<Zakim> mbgower, you wanted to say time on task is a very common measure for UX; it's an obvious way to assess equivalent effort
mbgower: If it's an assertion, not foundational, automatically makes it supplemental
<alastairc> We haven't said assertions can't be foundational, but we haven't found one that we thought could/should be foundational yet...
mbgower: time on task common measure
mbgower: can list time on task for critical tasks
mbgower: Provide research-based number for time on task
<Zakim> Chuck, you wanted to say this conversation has swayed me to believe this is best left as an assertion
Chuck: Makes sense for it to be supplemental
Chuck: Moving on to point pressure alternatives
Chuck: Reads survey responses
Chuck: No disagreement in comments, moving on
graham: Speech alternatives, how are we defining essential, fluffy area, what's essential to functionality
Chuck: Case-by-case essential, generic existing definition, drilling down in scope, otherwise not best discussion for today if not related to foundational and supplemental
<Chuck> +1 we have a definition in WCAG 2.
GreggVan: We use it all over, need to discuss it, pressure for stylus interaction is essential
GreggVan: Is there an alternate way to accomplish the task, define essential in context
<alastairc> Current definition in WCAG3: https://
GreggVan: Alternative with exceptions when essential lower level, higher level without exceptions
GreggVan: Having extra points when something is essential seems unfair
<Zakim> Chuck, you wanted to react to GreggVan to say I moved us to speech alternative
LoriO: Not sure whether foundational or supplemental, title wording is confusing
GreggVan: Meant speech isn't only way to achieve any task unless it's required
GreggVan: For example, sampling voice task
LoriO: Title as written isn't clear
Chuck: Focus on foundationalness
graham: Essential, platform specific, Amazon thing that you talk to, is that included in scope
<Zakim> alastairc, you wanted to comment on the foundational-ness of alternative inputs
<graham> and for clarity I meant Alexa, but did not want to set anyone's device off!
alastairc: Not relying on one input method is foundational, expressed in several requirements
alastairc: Whether voice-only devices are in scope, not at the moment, but would want to tell policy makers to look at service as whole
<Zakim> Chuck, you wanted to verbalize Rachael's comment
alastairc: Comments on the language of requirements, add comments to survey and on github
GreggVan: All voice products have rule to control with keyboard
GreggVan: When not an alternative, then definition of essential, if can't do this way, can't do task
GreggVan: Shouldn't be just one way unless keyboard interface; note not keyboard, keystroke data interface because can use different inputs
<alastairc> Regarding keyboard / keyboard interface, agreed, but it's a method of not relying on one input type.
GreggVan: Do one (keyboard data interface), opens up to others
<Ben_Tillyer> Never heard "keystroke data interface" before and google shows 1 result for that string..
<kevin> s|s/mbgower Already/mbgower: Already/||
Chuck: Errors persist
Chuck: Reads survey responses
<scott> +1 to supplemental
<Zakim> Chuck, you wanted to move us to the next criteria
GreggVan: Have as foundational because shouldn't be timing constraints
GreggVan: Exception on timing, real-time events, if error is announced as audio shouldn't repeat,
<Zakim> alastairc, you wanted to ask if an aspect of foundational is "it is mostly covered by something else".
<jon_avila> Seems like it would fall under Audio control in general.
alastairc: Deriving criteria for foundational, if general requirement for timing covers another (errors persist), then subtopic is supplemental?
GreggVan: No, because then it's an exception by putting as supplementary
<alastairc> Ok, so it's the more specific things that should be foundational... or we need to merge things.
GreggVan: Should only be at another level if exception at first level, add at next level to include it
Chuck: Shares Levels Exercise slide
Chuck: Today's discussion introduces possible criteria, feasibility, should feasibility be a factor in deciding
Chuck: Impact, how many beneficiaries, significance of impact
<kevin> Level Exercise
Chuck: Testability
<Zakim> Chuck, you wanted to move us along to concluding the conversation
<Zakim> alastairc, you wanted to add "universality"
alastairc: Feasibility, not mentioned in WCAG2, embedded in leveling
<Zakim> graham, you wanted to say feasability is a critical factor, otherwise we could say "all videos should be transcribed into every language and braille alternative available" as a foundational requirement.
alastairc: Foundation, need to be universal, good idea to do, if looking across scenarios where not universally beneficial, then not foundational
graham: Feasibility important, want to get more people to follow WCAG, putting things at foundational if they're not feasible
graham: Future of AI, work on what's feasible with current tech, can see where things are going, but costs
<Zakim> GreggVan, you wanted to say feasibility -- is not foundational or not but whether it is a requirement or not. Ditto for testability. EFFORT might be however
graham: Fewer blocker, more people will implement
GreggVan: Should be between any requirement, if not feasible, shouldn't be requirement, did have in WCAG 2
GreggVan: Had to be technically possible
<alastairc> Hmm, there are lots of things which are possible but not feasible (in general)
GreggVan: Testability, same thing, can't require if not testable
<alastairc> I'd thought large effort/cost is part of feasibility?
GreggVan: Effort and cost, if really hard to do could put at higher level, get more credit for higher effort
Ben_Tillyer: Feasibility, what take into account in determining
<bbailey> This might be last version of the WCAG2 requirements for SC that GreggVan was talking about: https://
Ben_Tillyer: E.g., framework doesn't support accessibility, can't add to list of exceptions because other platforms exist
Ben_Tillyer: One hour of consultancy can be unreachable for some
<Zakim> Chuck, you wanted to say that aren't these then valid criteria, just from the other side of the coin?
<graham> to address Ben_Tillyer point, as it is a good one, choice of platform to reach accessibility requirements can be a thing in WCAG 3 possibly. So if you chose a tech that didn't allow alt text, BUT there are other techs that do support it on a given platform, it is up to you to pick the right platform?
Ben_Tillyer: Focus on user needs in defining supplemental, fundamental
<Zakim> alastairc, you wanted to comment on terminology gap between effort and feasible?
alastairc: Large effort/cost is part of feasibility
alastairc: Take into account in generic and universal sense, let policy makers decide
<GreggVan> got it ALSO feasible to WHO and for WHAT technology but authors should not get out of jail because they choose a tech that can't be accessible
<alastairc> RRSAgent make minutes
<Ben_Tillyer> graham - exactly, but what if that framework is only available on a mac, or requires new staff to be hired?
<GN015> the choice of platform influences the user experience. We should stick to the user experience, not forbidding or promoting specific technologies. We might point out the influence the technology choice might have on conformance and on costs.
<graham> Yeah it would need a lot more thought than i gave it to put some edges on it, but I like the idea of putting the excuse of "but my platform doesn't support it" lower down the conversations!