<Chuck> meeting: AGWG-2025-09-09
<kenneth> scribe+
<kenneth> Chuck: Housekeeping first. Any new introductions?
<alastairc> That's the UK Gov digital service...
<kenneth> RichardMorton: I'm head of accessibility at UK Government Digital Services, just took on the role of accessibility and digital inclusion. Just recently joined the group.
<kenneth> Chuck: Do you have familiarity with IRC?
<kenneth> RichardMorton: I tried, but it timed out for some reason
<kenneth> Chuck: Alastair or Rachael can help via Zoom Chat.
Welcome Richard and the UK Gov DS !
<kenneth> Chuck: Moving on to announcements: I have none.
<kenneth> Chuck: Are there any topics to add to our list for future agendas?
<kenneth> todd: I was wondering if we were going to go over the comments from the survey, that I don't remember where we left off from, but I put a list of comments together from Safety and Deception subgroup, and we didn't get through the full list, so I wondered if/when we'd get back to that.
<kenneth> Rachael: The intent is for the subgroups to take the survey results and work them back in; if there are questions, you can reach out to the individuals, or we can arrange to meet with subgroups, but there wasn't an intent to bring it back up here
<kenneth> todd: OK, I'll invite the specific people then. Thanks
<kenneth> Chuck: First agenda item is reviewing WCAG 2 issues. Mike, can you give a short review?
https://lists.w3.org/Archives/Public/w3c-wai-gl/2025JulSep/0089.html
(as in agenda)
<kenneth> mbgower: I'm one of the facilitators of the Backlog TF. We do a 2-week review period, and usually remind the group, but we missed last week, so we're providing an additional week this time.
<kenneth> ... The email always goes in a particular order grouped by type of change
<graham> present_
<kenneth> ... Up until now, we've held bug fixes until the end of the cycle. From now on we're going to notify of bug fixes but implement (merge) them right away, so that there's less work required to review
https://github.com/orgs/w3c/projects/56
<kenneth> (Mike shares screen showing the WCAG 2.x Default Board, showing Sent for Approval column)
(url for project board that Mike is screen sharing from)
W3C sign-on required to see the board.
<kenneth> mbgower: Things to be reviewed will involve visual/content changes to TR space or informative docs. Things like scripting changes will be classified as bug fixes.
<kenneth> bailey: So to confirm, we'll give another week for this review, I'll send an email out to the list.
<kenneth> Chuck: Next, over the last year we've been managing the w3c/silver repo issues, to determine which raised issues can be closed vs. should remain open.
<Francis_Storr> there's ~30 old silver issues left to review
<kenneth> ... We're asking the group to review; if there's an objection to closing, note it in the issue, and we'll migrate it to the wcag3 repo without debate.
<kenneth> Chuck: Next, we're continuing conversations we've brought up regarding WCAG 3, and today we want to discuss the purpose of providing different conformance levels. There are a lot of different concepts, ideas, benefits, perhaps even concerns, RE what different levels look like or mean
<kenneth> ... we're trying to come up with a framework that encourages continuous improvement, that incentivizes content authors to invest in improving their content
<kenneth> ... we want to encourage people to see _all_ of the requirements and assertions and invest in addressing more of them, rather than narrowly focus on a subset.
<kenneth> ... One of the things we wanted to ask is, what other purposes are there / what other concerns?
<kenneth> Rachael: What is the benefits of levels, pros/cons; what are the reasons we would consider having them?
<kenneth> bailey: One of the ideas we had talked about much earlier, I guess, was instead of levels, being more specific with tags, e.g. level of difficulty, who it affects, can meeting the requirement be done invisibly?
<Zakim> Chuck, you wanted to ask about meeting invisibly
<kenneth> ... I want to re-raise that level of categorization, hoping the levels may fall out of those types of tagging
<kenneth> Chuck: I don't remember the detailed conversations about meeting a requirement "invisibly", could you describe more?
<kenneth> bailey: i.e. that you can meet a requirement without having much visible effect on the visual presentation of the page. (e.g. alt text)
<kenneth> alastairc: Picking up from Bruce, I think that's more, once you've got levels, how do you assign things to each level?
I agree, chicken and egg a bit.
<kenneth> ... Sort of goes to the encouraging continuous improvement - if people need to prioritize, how do you help them prioritize which things will have more impact than others?
<kenneth> ... Another separate thing: enabling regulators (in the general sense) to pick different levels for different circumstances, e.g. public sector services may be more critical for citizens than a holiday company, public sector vs. private sector, etc.
<kenneth> ... I get there could be pushback, i.e. they're all important to someone
We also talked about different levels for intranet versus public facing
<kenneth> Chris_Loiselle: Outside of this call, in us talking as a group, about whether there is enough data about what members of the W3C do go above and beyond to the level of AAA? Would be useful to look at a data set for driving this conversation.
<kenneth> Chuck: I don't know about any surveys of groups RE quantities of sites achieving AAA. I have observed there are sites that measure across public space for AA conformance, and a lot of people bring up that very few sites perfectly meet AA.
<kenneth> GreggVan: Another reason for levels is we wanted to be able to incorporate and provide incentive to go beyond A/AA. A/AA has kind of come to be viewed as a single level (as the "required" level).
<kenneth> ... A lot of really good advice we wanted to have was getting excluded. We wanted to make it possible to restructure in such a way that additional things such as assertions and recommendations and thins that couldn't necessarily be tested but were critically important, could be incorporated in a fashion that people would look at them
<kenneth> ... Wanted to establish places to look at things that weren't being looked at before. Want to incentivize and recognize those who go beyond bare minimum.
<kenneth> alastairc: On Chris's point RE how many do AAA: not a survey of W3C members, but from our own clients who have asked for testing, we've only been asked maybe 3 or 4 times out of hundreds to audit for AAA, so maybe 1%-ish
<kenneth> ... sometimes we do sort of a workshop to go through which AAA requirements are most relevant to a client
<kenneth> ... If we have levels, I'd be wary of combining A/AA into one level. There's a well-known behavioral economics thing, where the really expensive option makes the medium option look more desirable. i.e. having 3 levels helps sell the middle one
<johnkirkwood> +1 to Alistair
<kenneth> RichardM: AAA is rare, also I've never seen evidence of anyone claiming / aiming for only A
<kenneth> Glenda: I wanted to +1 what Alastair said about clients asking for AAA. Only seen a handful in the 14+ years I've been at Deque. We've had some clients try to do it, or made us do it, but then after a year, they pull back to AA, in terms of testing
<Zakim> Chuck, you wanted to advance the conversation
<alastairc> /me back
<johnkirkwood> +1 Glenda
<kenneth> ... In terms of training and designing, we can go to AAA in terms of best practices, etc. When the code is already there, it's a tough sell, doesn't seem pragmatically possible in the current state
+1 to all points raised
<kenneth> [Rachael speaking to slide 29]
<kenneth> Rachael: This was an experiment with levels early on. Sample set of supplemental requirements and assertions
<CarrieH> did anyone post a link to the doc in here
<Glenda> What if we have “tech requirements” as one part + “assertions” as the other part. And then the great best practices as the 3rd part < not required
<kenneth> ... there are some assumptions in what we're looking at here. Assumes we're using the foundational set, and assuming all foundational requirements are met, and that N/A counts as a pass. These seemed to be the directions the group was leaning towards.
Over the years, as something of a sidebar, I have gotten lots of utility from AAA SC as evidence that what someone is asserting about a similar AA SC is not correct.
<Chuck> Test mapping: https://docs.google.com/spreadsheets/d/13GNx7Ck70B84_Ri6W9yxMvYr8x-G8oBUr7oiF6jPZiY/edit?gid=1385430754#gid=1385430754
<kenneth> ... I've taken those assertions and mapped them against the different functional needs that are listed, as to whether each assertion or supplemental requirement applied to that functional need.
<kenneth> ... Feel free to go through these, there are 3 different scenarios (3 tabs/sheets)
<CarrieH> Thanks alastairc and ShawnT
<kenneth> ... Back to the slides: Slide 30 shows what percentages a few sets of requirements/assertions would score for particular types of challenges/disabilities
<kenneth> ... This breakdown becomes an interesting thing to at least report on, but also to consider setting levels around.
<Glenda> I love the Percentage of “Supplemental” table with “grading” based on functional needs. (I just dislike the word “Supplemental” because it sounds optional)
<kenneth> ... Observe that percentages can vary widely for different needs, with the same satisfied supplemental requirements
<kenneth> ... If we have levels, we could say e.g. Bronze is all foundational + 20% of the possible supplemental requirements / assertions that you've done
<kenneth> ... In first draft of Silver, we were doing it by counts and got pushback on counting and complexity. This has simplified that down, which seemed to have more positive reception
<kenneth> ... (Slide 31) Do we want to set levels, and is that useful? If so, by functional need or overall?
<kenneth> ... With conformance being tied to functional need, would it be beneficial to create sets that we recommend people to try? (e.g. bare minimum, vs. education, vs. very large complex product...)
<kenneth> ... Wanted to open up questions about these thoughts
<kenneth> Makoto: wanted to make a comment on providing different levels from a different point of view. There are many countries/regions where digital accessibility is still not legally required, e.g. Japan.
<kenneth> ... in Japan, at this moment, AA is baseline for public sectors and global companies in Japan. However, in such countries, at least in Japan, it can be a starting point or first step for more organizations if WCAG3 were to provide a lower level with minimum requirements
<Zakim> mbgower, you wanted to say that judging WCAG 2 level AAA as a predictor of adoption for WCAG 3 should be done cautiously
<kenneth> ... we will be able to make more things more accessible as a result. WCAG is recognized internationally.
<mbgower> https://www.w3.org/TR/UNDERSTANDING-WCAG20/conformance.html"It is not recommended that Level AAA conformance be required as a general policy for entire sites because it is not possible to satisfy all Level AAA Success Criteria for some content."
<kenneth> mbgower: Going back to Chris's question about adoption of AAA in WCAG 2, I think it's important for us not to use how WCAG 2 was structured to influence how WCAG 3 is carried out.
<kenneth> ... WCAG 2 informative docs remark that AAA is not always possible.
<Chuck> +1 to mike on the messaging may have influenced the results
<kenneth> ... going forward we can focus our messaging for adopters. Regardless of what tactic we go with, it's important to have really clear communication on it. WCAG 2 felt very all-or-none which seemed to stifle motivation towards anything besides AA.
https://www.w3.org/TR/WCAG22/#cc1 5.2.1 note 2
<Glenda> +1 to anyone just testing to “A”. I only saw it when Canada was rolling up to AA. But in my opinion…A/AA are just noise.
<kenneth> GreggVan: It is not just the nature of WCAG, it's the nature of all standards and regulation, that you set a threshold and you pass or fail. We can do whatever messaging we like, but it'll still come down to laws/regulations saying "this is what you need to do".
<kenneth> ... the key I think we've been trying to do with WCAG 3, is we structure it so that when they're reading what they "have" to do, they also see what they "should" and "could" do.
+1 to GreggVan points about could/should being in close proximity to must
<kenneth> ... it'll be hit and miss, but we can more incorporated if we keep the requirements and what you could do beyond the requirement, close together / mixed in.
<todd> question is, do they (the engineers) have the affordance to do so when they have a manager demanding they have something done yesterday?
<kenneth> ... You see organizations go through spurts where they aim high, then fall back to doing only what's required. Need to think about that reality.
<kenneth> ... people will often go above and beyond if it's right in front of them, but not if they have to dig for it.
<alastairc> todd - as Gregg said, it's hit and miss, but if it's there then some will.
<kenneth> ... RE Slide 30: I love it, but I want to point out some of the problems in how we do this.
<Rachael> +1 to the breakout of functional needs will affect this.
<todd> alastairc - thank you, i had missed that part
<kenneth> ... Right now we have 6 items for cognitive and 1 for vision. The fewer requirements there are in a group, the more widely the results will vary between groups. Might want to think about having fewer bigger categories, lots of little categories.
<kenneth> ... RE e.g. flashing, sometimes having no flashing is impossible, but if there is flashing, it's sometimes impossible to get rid of. e.g. stripes in a crosswalk effectively causing flashing as the view scrolls. There needs to be possibility of scoring high even when your page isn't extremely simple e.g. no pictures/multimedia etc.
<Zakim> alastairc, you wanted to comment on differences across functional needs by FN (some have more requirements than others), and scenario (education might skew to learning
<kenneth> ... We want to be a measure of accessibility, not a measure of effort. This example is fascinating, but it brings up some warning flags.
<kenneth> alastairc: RE percentages: I'm a little wary of building it in as a core conformance thing, for 2 reasons.
<kenneth> ... Differences in cross-functional needs. e.g. some are kind of all or nothing like sensory disabilities, most of which will likely be foundational. But others e.g. COGA things could have lots of different requirements. So it may seem imbalanced.
<kenneth> ... Differences by scenario: in some scenarios you may want to skew more towards certain requirements than others, as they're more important to that scenario. I'm not against embedding metadata towards that kind of reporting, but I'm a little wary about the core conformance aspect.
<kenneth> RichardM: I think it's important to have percentages both for functional and overall. Depends how you measure. A large proportion of sites don't meet AA; I think there's incentive for people to improve, e.g. if they're at 50% and need to get to 70%.
<johnkirkwood> “percentage of what” <— can we answer that question? it seems undefinable … how will that fly with public, regulators in real world scenertios. I could see if we are talking about incorporating into a “maturity model” paradigm
<kenneth> ... As Gregg said, we shouldn't be rewarding based on effort, but you need to be pragmatic about the circumstances
<GreggVan> Thanks great info
<kenneth> maryjom: Everyone makes the assumption that AA is what all jurisdictions are picking up. In India, draft regulations are thinking of level A. Some jurisdictions e.g. Ontario leverage A, AA to ramp up accessibility, i.e. provide prioritization
<alastairc> I ad assumed that originally, people (in AG) thought that some would go for level A, and then move to AA. It was a surprise (and a success) when most places went straight to AA.
<kenneth> ... RE percentages: I agree with Alastair, worry a bit about leaning too much towards particular disabilities. I know it's hard to find a good balance. We'll have to keep that in mind.
<Zakim> Rachael, you wanted to balancing
<kenneth> ... In current regulations, they want COGA into one thing. They have a certain number of functional criteria to categorize things by, that we might be able to leverage.
<Charu> +1 to be a measure for accessibility and not a measure of the effort
<kenneth> Rachael: RE Alastair/Gregg's point, chair hat off, the reason for using percentages here is to even things out. Score points (e.g. A+) might make it even less even across types of disabilities; percentages helps even it out.
<GreggVan> +1 to impossibility of weighting
<kenneth> ... Gregg mentioned functional need categories; there are a lot of factors that affect balancing this. Number of requirements, size of assertions / how they're broken down, percentages vs. points; when we first tried points we tried weighting different requirements differently
<kenneth> ... not getting into the details right now of weights / applicability of each requirement to each need
<Zakim> Chuck, you wanted to say I'm concerned about it being used for conformance, but I could see value in scoring
<kenneth> ... chair hat still off, I think percentages still help even it out, because you can focus on one area, and this helps make the effect of that more visible
<kenneth> Chuck: Chair hat off, Oracle has never been fond of scoring, particularly under the current conformance model and set of standards. This does look interesting and promising; of course the jury's out
<kenneth> ... wiould have to see how it's implemented
<johnkirkwood> +1 to Chuck
<kenneth> ... echoing some earlier comments, I do have some concerns, but this percentage approach does have a whole lot of appeal and I am excited by the possibility. I recognize this sample only represents a subset.
<kenneth> ... Question, when you came up with e.g. 58%, was that average of percentages over the functional needs, or percentage of the count of standards met vs. not yet? (I think each would produce different results)
<Zakim> alastairc, you wanted to ask what we do with requirements/assertions that do apply to several FNs?
<kenneth> Rachael: Within each category, the number that were met out of the possible ones that would apply
<johnkirkwood> the number of criteria of those that apply to memory. a good example. but some may be more critical than others, no?
<kenneth> alastairc: What about requirements or assertions that would apply across multiple functional needs? Do they score for both in the percentage?
<julierawe> Pasting in a comment from Stefan Chitu in the Zoom chat for this meeting: Follow-up question: I suffer from epilepsy. Nice approach on the %. Flashes/flickers + red color + >3:1 contrast => seizures => cognitive decline or death. How would we ensure that flashing is 100% to all the categories?
<Zakim> Rachael, you wanted to answer alastair
<kenneth> Rachael: RE Stefan's question. The lower-level flashing requirement (only essential flashing) is a foundational requirement, so that would need to already be met. This is the enhanced "no flashing" requirement.
<kenneth> ... motion, pseudo-motion, e.g. each have foundational and supplemental (enhanced) versions
<kenneth> ... RE Alastair: Each requirement is mapped against functional needs, in the spreadsheet linked in the previous slide
<Zakim> bailey, you wanted to say that if AG doesn't come up with % scoring -- the tool vendors will -- and they won't match
<Zakim> Rachael, you wanted to say where it gets embedded would likely depend on whether we tie this to conf. or reporting
<kenneth> Chris_Loiselle: My question is around dashboards/metrics. Companies reporting in-house vs. externally. Is this looking to benchmark companies against other companies for procurement? Conformance documentation needs to be clear about levels, full conformance, etc. Various vendors need to apply consistent measurements
<kenneth> Rachael: This should ostensibly be simpler, the percentage is based purely on passes/fails. It's not counts or severity
<kenneth> ... I would hope this would be built into tools. We would provide the mappings needed
<kenneth> ... RE reporting internally vs. externally, depends on what we decide for conformance, e.g. Bronze = all foundational + some percentage of supplemental requirements and assertions. Then provide mappings for functional needs
<kenneth> ... not sure we can make it a consistent set of reporting, because we've left out the ambiguous parts
<Chris_Loiselle> regrets for second half of call.
<kenneth> scribe+ bailey
<scribe> scribe: bailey
Chuck: I will close this out,
perhaps on a hot not, Chris asked about comparing across
products...
... but if we give out meta data -- that will open possibility
to gaming system -- but that might be okay...
<RichardM> +1 to gaming being potentially useful
Chuck: it gives all companies way
to improve score to next level.
... Moving on, would it be beneficial to create sets of
recommended supplement requirements and assertions...
<Zakim> Chuck, you wanted to answer if Rachael does not and to say and to change scribe
Chuck: or can we count on scoring to do that work.
<johnkirkwood> to the percentage by functional need. is it a pure percentage of the NUMBER of functional needs or do some functional needs have a larger (weighted) percentage?
alastairc: Coming back to any scoring (points or percentage) remind folks that treatment of not applicable skews results.
<Chuck> +1 to including meta-data
<Zakim> alastairc, you wanted to comment on comparability
alastairc: Functional needs as meta-data is different question and should allow mapping. I'm not sure that we should include it in core conformance..
Chuck: Moving on to representative sampling, slide 33, is it already covered?
Rachael: We *think* current approach of defining scope and with getting started information, representative sampling does not need to be explicitly incorporated into the scoring.
<johnkirkwood> some sites one could only do “representative sampling”, no?
Chuck: Some sites have unbounded permutations. Everything is so dynamic, not just users but topics.
Ben_Tillyer: An example of the difficulty is that there is no canonical page view -- every user every place every time every instance is different content all from the same "page" or even same URL.
<Zakim> alastairc, you wanted to comment on the complexities of sampling
Ben_Tillyer: Good to see web developers select components and evaluate those.
alastairc: With WCAG2 its up to
site making conformance claim to scope what is being covered by
the claim. So this is really just the status quo.
... We have WCAG EM which starts to address this. It's a
process approach. With an infinite variety of websites, can it
be any more specific? Are we missing anything.
julierawe: My company is moving from "is it possible" to "what is feasible" and we focus on high traffic sites...
<Zakim> alastairc, you wanted to comment on gamable vs feasible
julierawe: worry on "gamble" seems less problematic to me, because low use videos not having audio description seems less of a risk.
<alastairc> If people check that, true. But with millions of pages, how would people notice?
alastairc: My larger concern for "gamble" is sites narrowing scope of their conformance claim. The pick only the simpler pages. Or only the pages targeted to accessibility issues.
<CarrieH3> I would assume that some companies may prioritize accessibility work may weigh how much of an impact x probability of it happening. So like what Julie was saying if it's a site with low traffic, they might prioritize work on a site that has larger impact..
Ben_Tillyer: Sites scoping out videos will be pretty transparent, so I am not too worried about that.
<Zakim> Chuck, you wanted to read some comments for the meeting
RichardM: I had sites acknowledge risk of not covering everything, so I agree that concern for gaming the score is less of a concern.
alastairc: From conversation, I think we can close this conversation for now. This is not end of conversation, but nothing to add to conformance for now.
<RichardM> my point was more that uk public orgs are sometimes accepting the risk of not complying with regulations, so they may be tempted to game the sampling side.
GreggVan: I would like to see us consider that we work on guidance to encourage people to go on beyond bare minimum...
<alastairc> s/but nothing to add to conformance for now/
GreggVan: Paper would include possible incentivization going forward.
<Ben_Tillyer> Don't we start with the foundational level?
<alastairc> Ben - see slides 17/18
GreggVan: issues beyond conformance, but how to be most productive with the effort.
Chuck: Seeing no one on queue, moving to next topic, third party content.
alastairc: Clarifying question from GreggVan "non-normative content" in this context means not covered by conformance model.
[page 35 in slide deck]
<laura> +1 to Gregg
GreggVan: My perspective on
"third party" is that the site owner shows it, its is covered
content...
... site could have used different vendor, not offered the
service, etc.
... If third party is letting end-user post content, that is
different...
... but third party posting stuff for sale, then the
responsibility of the site is more...
<Chuck> ack
<Zakim> Chuck, you wanted to offer some other types of 3rd party content
GreggVan: at a minimum, the site needs to facilitate accessible content from public.
Chuck: That is where I want to
steer conversation, that third-party has different
meanings...
... One example is news aggregation site. Another is allowing
public to post -- and the public has low awareness of a11y.
<alastairc> Some further reading, where the group has gathered info about this before: https://www.w3.org/TR/accessibility-conformance-challenges/#Challenge-3
kenneth: By third party I thought of that as payment processors and the like; is it worth having a distinction between third-party content and user-generated content?
RichardM: I think of third-party as any user-generated content.
<Poornima> regrets for last half of the meeting. thanks!
Chuck: In Oracle we are responsible for anything we incorporated. Customer does not care that something came from elsewhere.
<Zakim> alastairc, you wanted to comment on difficult situations for things like advertising
<GreggVan> should be POSSIBLE for content to be created or added as accessible or the site is not accessible. Even though not responsible for the actual content -- it would be good to provide tools or checkers that encourage accessible content - though this is not required by WCAG (unless we get into authoring tools requirements.
giacomo-petri: One variant on this incorporating and API, and as developer I might not know exactly what are dependancies. Especially since API behavior might change without any change on my site.
alastairc: maryjom raises good example of advertising. Site owner doesn't know what is in the banner!
<johnkirkwood> the user needs to be made aware that it is third party content
alastairc: Might not be many choices, and without adds, site can't exist.
<johnkirkwood> +1 need to be responsible
<Zakim> GreggVan, you wanted to sau Types 1) components you buy to build your site. (you are responsible) 2) content you gather and then resell (you are not responsible as
todd: There are a lot of dark patterns in advertising, those need to be called out.
GreggVan: I would remind us that
WCAG is the ruler not the rule...
... If something is not accessible -- even if not feasible to
do so -- WCAG should score that as not accessible.
... Types 1) components you buy to build your site. (you are
responsible) 2) content you gather and then resell (you are not
responsible as
... 2) content you gather and then resell (you are not
responsible as website but are as seller of the product. the
sites is not inaccessbile but the product is. So may be liable
for inaccessible product. 3) content is by others and you do
not selle content (e.g. email program) Author is not
responsible since they have no control.
in type 3 it
GreggVan: Site should not be
responsible for something like email coming in. But it should
be posslble for something like email to be accessible.
... Other issues is to ensure third party content CAN be
accessible. So for example if users can post an image, the site
must provide a way to provide ALT.
Rachael: All the items we have covered here would not effect our conformance model.
Chuck: For third party content, how does site scope a conformance claim?
GreggVan: Site asserts site is
accessible before contributions, site facilitates end-users to
post accessible content....
... then conformance statement describes those
conditions.
... WCAG should treat third party content as an exception.
[slide 35 being updated to reflect live conversation]
<Zakim> alastairc, you wanted to comment on authoring tools methods
<johnkirkwood> +1 to Gregg
GreggVan: This stuff belongs in a policy document, not exceptions under WCAG.
alastairc: What if we had requirements specific to user generated content?
<Chuck> +1
alastairc: For example, users are
allowed and encouraged to provide alternative text....
... That could be one of our methods, and allows the owner of
the site to make conformance claims just on the mechanism.
Chuck: So site conforms by providing potential.
alastairc: Yes, just need to double check that we are not creating huge loop holes.
johnkirkwood: I have had to deal
with this 100s of times, we have oversight of dozens of units
using dozens of platforms...
... We are try to take this huge umbrella. For AG, its not a
useful conversation.
giacomo-petri: Building on
alastairc point, consider something like
accessible-name...
... that, technically, is not something a person in the public
can provide.
<giacomo-petri> perfect
alastairc: The User Agent working out how to overcome a barrier, but I think that's a different thing.
<johnkirkwood> who is the 1st party, who is the 2nd party, and who is the 3rd party?
alastairc to johnkirkwood i don't think we can take much responsibility for the broad organization problems. We are developing ruler, not rule. First party, the one making claim, has to account.
[meeting end]
This is scribe.perl Revision VERSION of 2020-12-31 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Succeeded: s/Mike:/mbgower:/ Succeeded: s/being is close proximity to/being in close proximity to/ Succeeded: s/ejbdccuuklgukdldvictlrbrvgikbtvldllecrdueduk// Succeeded: s/that AG doesn't not come up/that if AG doesn't come up/ Succeeded: s/I'm not sure about how that impacts scoring/I'm not sure that we should include it in core conformance./ FAILED: s/but nothing to add to conformance for now// Succeeded: s/but nothing game breaking for now/but nothing to add to conformance for now/ Succeeded: s/I think that means payment processing and the like/I thought of that as payment processors and the like; is it worth having a distinction between third-party content and user-generated content?/ Succeeded: s/rssagent, draft minutes// Default Present: Rachael, filippo-zorzi, Francis_Storr, mbgower, Ben_Tillyer, todd, hdv, Makoto, kevin, bailey, BrianE, alastairc, Chuck, ShawnT, kenneth, CarrieH, Azlan, Kimberly, julierawe, Laura_Carlson, AlinaV, johnkirkwood, jtoles, Charu, graham, JeanneEC, Glenda, maryjom, Roland, giacomo-petri, Jen_G, LenB Present: Rachael, filippo-zorzi, Francis_Storr, mbgower, Ben_Tillyer, todd, hdv, Makoto, kevin, bailey, BrianE, alastairc, Chuck, ShawnT, kenneth, CarrieH, Azlan, Kimberly, julierawe, Laura_Carlson, AlinaV, johnkirkwood, jtoles, Charu, graham, JeanneEC, Glenda, maryjom, Roland, giacomo-petri, Jen_G, LenB, CarrieH3 Regrets: Jennifer Strickland, Rain Michaels, RainM, JenS Found Scribe: bailey Inferring ScribeNick: bailey WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 People with action items: WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]