Meeting minutes
Introductions
Announcements
Adam: Quick announcement that next August 4th there is a group onboarding session 30 mins prior to the meeting
<hdv> +1 break
<Ashlee> +1 break thank you!!
Adam: Today we are introducing a process change in our meetings; taking a 5-minute break at the halfway point. Expanding that so folks can rest and recharge.
kevin: Editors have been putting together their PRs for the provisions that you've all been working on. We're meeting on Friday to pull all unmerged issues and pile them in. Hopefully get the latest version available for CFC for the group then out to another draft publication. This isn't going to be massive, we're trying to update selective bits and getting informative materials out there.
kevin: If anyone has questions about this, please give us a shout!
Do we want to meet at CSUN 2027?
Adam: CSUN is at Anaheim, thinking of arranging a face-to-face at the event, looking to survey options that are most attractive
… if you have thoughts if you'd like to meet at CSUN 2027!
<wendyreid> +1
<Jennie_Delisi> Hybrid option?
<nattarnoff> +1
<Bernardo> We usually have a Google Booth in there!
<Charles> 0 to in-person meeting. unable.
<CClaire> +1
hdv: If we are meeting I would probably go to a different event, going there would mean preparing something so curious if folks are going and interested in planning something; would love to see this happen in Europe but seems CSUN is a place many people go to already
<nattarnoff> Intend on going, not locked in.
<Jennie_Delisi> +1 if hybrid option is available.
<Ashlee> +1 to hybrid
Rachael: Typically we hold hybrid on the Monday before in California time, not Eastern Time
… that is an option if you are interested in joining hybrid
<Wilco> -1 personally. CSUN is a long week already.
GreggVan: I think about our members who can't meet face-to-face, and want to weigh the benefits versus keeping participation fully open
GreggVan: if there is some way for us to replicate face-to-face online that won't require members who can't get the travel money
<Patrick_H_Lauke> (noting that a lot of folks outside of the US will be loathe (?) to travel to the US ... currently)
Ben_Tillyer: Haven't been to CSUN because of distance, we should be mindful of what sort of work goes on there, and the decisions are not being made without the full cohort of AGWG; also happy to host one outside the conference
… there's TechShare Pro happening in London in November
hdv: I like face-to-face for these types of group, it will helpl us spend a longer amount of time on one subject; to have a whole day talking about something can be helpful from experience from other groups
hdv: I prefer it to be outside of the US and many may have difficulty joining/travelling there; maybe we can look at Asia that has W3C locations or Europe at the Netherlands with the Dutch government or Canada
<nattarnoff> As an American without a current passport I support hdv's recommendation of a non-USA location.
<Patrick_H_Lauke> conference often makes it easier for folks who need to justify their travel/work absence
<Wilco> +1 Hdv, dedicated meeting, not tagged onto an already busy week
<HaTheo> +1 to not at a conference
<Ben_Tillyer> question to chairs/kevin: what do the other groups do re: f2f cadence?
hdv: would love it to not be a conference; advantage of face-to-face
<Zakim> Rachael, you wanted to say understood about the US can we get suggestions
Rachael: we will schedule all-day sessions that are completely virtual over the next year, but CSUN is out; we need a location in spring — the advantage of conferences is people can get funding to go, if we want outside of conferences, maybe we can look at separate funding opportunities for people who can't get funding to do a WCAG-specific travel option; we're looking for 6 months out from TPAC in spring, reach out to chairs if you have
… locations to offer
<kevin> I can check into that Ben... I suspect it 'varies'
<Charles> note: funding is not the only restriction to travel.
GreggVan: in WCAG 2 we had face-to-face, it was useful but when they're not done as part of the conference it can be a small number of people who can get funding to do it; if we can't figure out how to do it in a 2- or 3-day virtual meeting, so we don't restrict others with less resources
<Charles> GreggVan one major factor for in-person is sensory memory. if senses are not overstimulated, the sensory input improves the working memory and memory mapping of new information for many people.
wendyreid: There are no accessibility conferences in the first half of the year that aren't based in the US — CSUN, Access U… — one thing to consider is teaming up with IAAP since they have a chapter in Canada and in Europe, maybe suggest to them to co-organise an event as an excuse to get everyone together at the same place at the same time
<HaTheo> in 2018
<HaTheo> yes it was on site, but it had a dedicated conference room with the other ones
<Charles> the design studio we did 2018 was very useful
Matt_King: Several tradeoffs: one day only and distracted by the rest of the conference — the Silver Task Force did two days consecutively off-site from CSUN — being offsite in a dedicated space and spread across two days did a good job; CSUN is a juggernaut that it ends up competing with it
<HaTheo> in 2018 I presented but that was the only reason I was able to attend.
WCAG 3 document breakdown (continued) — https://www.w3.org/wbs/35422/wcag3-document-breakdown/results/
<CClaire> I am not sure, but is it necessary to be the accessibility event? Or we could try something like MWC, Computex, etc.?
Adam: Chairs will review the minutes and weigh the options spoken of today
Adam: Resuming our WCAG 3 document breakdown discussion from last week; picking up from conformance concept of severity
<Rachael> slide: https://
Slideset: https://
Rachael: This is the analysis of our breakdown; slides are not decisions only stimulate thought and looked at majority answers that fell into majority and which fell into least objections. I want to acknowledge that neither of those are the same as a consensus.
Rachael: It provides us a direction of next conversations, we'll go through the details today; majority of people were leaning for scoring to go in informative documents, the fewest objections were leaning towards putting something in about levels and conformance
Rachael: From a progress towards conformance and partial conformance; people seem to want it in conformance and had the fewest objectios
Rachael: This is not the same as consensus but shows some ways to think about building a proposal to come back to the group
GreggVan: Progress toward conformance is all about scoring so it's weird that we have scoring and informative and progress in conformance. Partial conformance is a contradiction, no such thing.
Rachael: Partial conformance was changed to Progress towards conformance — you could also have levels and progress towards conformance-based on levels and not scoring
GreggVan: They're progress towards higher levels of conformance, once we make that normative, it stops to create a better measure for it
Rachael: Start with Severity — 1 person said it must be in conformance, 8 it should, 9 it should not but should be informative, 3 it must not and 2 felt it should not be there
<bbailey> https://
Rachael: there's a general spread around severity can be simplified to almost automatic inter-rater reliability be below 80%
… severity conformance is needed for someone's work, and their understanding of what blocks someone might be wrong, I'd rather people focus on making content accessible to as many as possible rather than focusing on a few accumulations and applying it across the site
… many commented on severity and conformance, and some saying not to have it normative but containing informative documentation
… conformance alone does not always provide enough information to differentiate products with different levels of impacts on users
<Zakim> Charles, you wanted to clarify severity definition
Rachael: we have a group of people who feel strongly that there is a lot of value in prioritising content and fixes based on severity and they are leaning towards putting that in conformance, the o ther side of the equation is so subjective
Charles: I agree some of them need definitions and are difficult to measure; did severity intend to mean severity of harm?
<kirkwood__> it is impossible to define
Rachael: We've defined severity in a number of different ways; can we find one that is objective enough to use it?
Charles: I don't have an opinion and can't have an answer to this question unless we have that definition
<kirkwood__> agree with Charles
Wilco: What it comes down to is what problem are we trying to solve with these? I wish there was more clarity behind it; hard to see if this is a good idea to address with this potential solution
<kirkwood__> +1 to Wilco
GreggVan: I think severity is important but like other important factors, it's subjective and individual. If we define it as amount of harm, define harm? Is it harm or inconvenience or barrier or hindrance?
<Charles> if severity means severity of harm, the best i can imagine is: death; completely blocked; and is difficult. and even that is still subjective and may favor some functional needs over others.
<kirkwood__> severity of effort? seems like we are talking about. if it’s inaccessible, it’s inaccessible. period
GreggVan: If you can't keep up, get a job, those are harms as well
<kirkwood__> +1 to Gregg
GreggVan: we'll have to face this a lot. If we want to figure out how to do and figure out how to do with WCAG 3; we have to figure out how to get them into being practiced in other ways. Let's see what we can do with what's important.
<Patrick_H_Lauke> it will be impossible (?) to make this objective, as by definition "harm" and how it affects people is extremely subjective
<Charles> +1 to GreggVan on different types of harm
Rachael: Conformance Subgroup had come to the point where the group thinks that stepping back out of the conformance conversation and approaching from the perspective of surrounding materials to better think of the purpose and coming back to conformance as a way to talk about this — the purpose of severity is to be able to prioritise fixes and address the issue that we've brought up in partial conformance and stages toward complete conformance
Rachael: We're trying to deeply understand each of these areas, the concerns, stances, and how people think about them and bring it back to the Conformance subgroup.
Matt_King: The problems with this by itself, on its own severity and separate from dependencies, even defining the purpose of using it — what Rachael said about figuring out the problems and looking at what is the most useful purpose of severity, until we have that defined, it's hard to have an opinion about whether or not it should be part of conformance or measuring conformance because you're further from conformance if you have 3 failures
of the exact same provision and they all inflict different levels of harm however we define that. If you fix the one that removes the least harm, you didn't come closer to conformance in a meaningful way in the user's point of view.
<HaTheo> +1 to what Charles, Wilco, Gregg etc. have said. but, I also feel like Severity is contextual and important for organizations, regulators etc. to define. I would be in favor of a common definition if it's helpful, but I worry about applying it as it feels hard to do in a general way.
<Patrick_H_Lauke> the original WCAG 1.0 had a concept of severity/impact when it devised its Priority 1 / Priority 2 / Priority 3 and the matching A / AA / AAA conformance level https://
<Patrick_H_Lauke> e A, AA, or AAA bucket ... but (supposedly) without defining what the bucket is/represent, which frankly I find hard to believe...)
Ben_Tillyer: I think it's an impossible task to define the words we're discussing to apply to all users on all digital interactions for every type of digital application, but maybe it's less impossible for owners who want to claim conformance to understand the potential levels that accessibility bugs might cause to end users with disabilities after having those conversations with these users; is there something we can do to discover, define, and
measure the potential harm in their smaller digital world.
<HaTheo> +1 to Ben
giacomo-petri: Severity is critical however because since it cannot be objectively defined, I think it shouldn't be part of normative conformance. I see audits covering the same scope and producing different results; severity might help in linearising the results — instead of adding all those parameters (informative or normative), what if we require a performance audit and define a conformance audit
<kirkwood__> severity of the issue that creates non conformance or WCAG guidelines. Harm is a legal term. (in Disability rights) think we should stay away from it
<Charles> prioritizing as an approach to remediate conformance failures seems to be about harm to the organization and is very different than severity of harm to users.
<Rachael> +1 to parameters to ACR / audit
<Zakim> Jennie_Delisi, you wanted to severity as part of specific criteria vs overall
giacomo-petri: … not necessarily prescribe a specific formula or testing methodology — we could potentially define what constitutes a conformance audit and define the minimum parameters such as scoring severity not necessarily defined by WCAG, but are parameters that if you evaluate against WCAG, you have to consider them otherwise you are not meeting the conformance audit definition.
<Patrick_H_Lauke> (related to my previous point about the circular discussion ... this is what i was thinking of https://
<Jennie_Delisi> Will this decision impact the ability for us to apply specifics related to severity within individual criteria? For example things which impact large financial actions - or emergency communications impacting life and death?
Jennie_Delisi: Can we talk at the criteria level — will this decision impact how we apply specifics related to severity within individual criteria? If elements on a page impact a large financial decision or emergency communications impacting life and death, are those going to be impacted if the group says severity only belongs in the informative documentation.
<kirkwood__> severity based on what Hidde?
<kirkwood__> LOE? Harm? IT infrastructure? personnel?
<Patrick_H_Lauke> +1 most auditors / companies I worked with/for do include some concept of severity when reporting to their customers, to help them prioritise fixes
<Charles> hdv is that use of severity in reporting about prioritizing remediation? or used some other way?
<wendyreid> +1 Patrick_H_Lauke
hdv: I've always added severity levels to my accessibility reports, those are only internally consistent and I'd use a different skill in different projects because a lot of severity depends on the project — it's hard to come up with one system that works for every single digital product so maybe our severity guidance should be how to build your own severity system for your product or project. Have guardrails or guidance that impact severity
… or could take into account when deciding severity and let people specify their own severity levels. It is more useful and possible to make a list of things that people can take into account when they make their own.
<Patrick_H_Lauke> (but make it clear that "you still need to fix everything, but here's the big showstoppers that if you only have limited time/budget you should tackle first)
<MartijnHoutepen> +1 to hdv to provide guidance on severity instead of prescribing
<Patrick_H_Lauke> (need to drop,sorry)
<HaTheo> +1 to the idea of Giacomo early of defining a conformance audit and establishing a clear/standard methodology for conducting it.
<kirkwood__> Think we should ditch severity. It is either accessible or not. That is hard enough given all types of disabilities.
GreggVan: It was mentioned that we would require a conformance audit as part of conformance — if they had a published one, we've always not required reporting as part of conformance; asking people to not only conform but to tell them whether they did or didn't is something that goes beyond what we want to try to do. Stay away from requiring people to claim conformance.
<Wilco> +1
kirkwood__: I think we should completely ditch severity — we cannot judge by level of effort , disability group, given all our knowledge and disabilities out there.
Bernardo: I think we shouldn't require conformance reporting and they can go and their own audit. For the baseline conformance level that everyone should aim for at a minimum, I don't think severity is a productive criteria or part of the discussion for anyone involved. Main suggestion is to remove severity from the baseline conformance level.
<Ben_Tillyer> +1 to Bernardo
<shadi> +100 to Matt_King
<Jennie_Delisi> +1 to Matt
<kirkwood__> +1 to Matt
<Zakim> giacomo-petri, you wanted to respond to Gregg: I'm not saying that people need to publish or report the results, but to validate WCAG conformance you have to perform a conformance audit in some way; otherwise, you have no way of knowing your level of conformance, and when conducting that audit, you have to consider the relevant conformance parameters
Matt_King: I'd like to encourage broader thinking in this space; want to make sure we remain open-minded. In my research I have found that there are output standards in security and medicine and safety, like software security, where having an evaluation conformance part of the standard and many of these have objective definitions of things like severity that I don't think are completely out of reach for us. It's hard to talk about severity out
… of context of reporting; we need the definitions and then judge them.
<Wilco> What problem are we trying to solve?
<Ben_Tillyer> +1 to Matt, was why I was using "more impossible" and "less impossible"
<wendyreid> +1 to Matt
<Bernardo> We should have a solid definition for "severity" nevertheless
<Ashlee> +1 to Matt, particularly I think it'd be great to see and consider formal definitions out there
<shadi> +1 to giacomo-petri
<Zakim> Rachael, you wanted to suggest we move forward to the next topic
giacomo-petri: Wasn't saying to require generating output/reports, just that if you want to validate your product against WCAG, you have to consider severity, scoring, whatever we decide should be included — when you perform audits, you take care of those parameters.
<Charles> if severity intends to mean severity of harm to a user, it is theoretically possible to create a somewhat objective matrix with type of harm; degree of harm; duration of harm; and functional need harmed. but then if used in conformance would dramatically complicate the model. if severity intends to mean a prioritization scheme for remediation, then
<Charles> that is clearly policy suggestions.
<Ben_Tillyer> could we have a slide or other visual cue in zoom that we're on a break? (in case people join now and cant sync the irc history)
Adam: Handing it to Rachael for Reporting discussion
<Adam> https://
Rachael: Referencing (5) Reporting Requirements survey results via link above.
<Ashlee> Is Rachael reading from the survey results just linked, or something else? I'm having a hard time following what's being said.
<Zakim> Charles, you wanted to ask about regional reporting requirements
Rachael: Asks for any additional comments to capture to understand where everyone is standing.
Charles: My understanding of the law is that there are various jurisdictions of the world that have reporting requirements.
<Zakim> Ashlee, you wanted to ask for clarity on what's being read aloud
Charles: If we required reporting or a specific format of reporting would it dramatically inflate the amount of work that occurs in those jurisdictions?
<Rachael> Document of notes: https://
<kirkwood__> It is required in some legal jurisdictions in the US
Ashlee: Struggling with following the read out of the responses. Please provide the link (above).
Rachael: Not sure what the access restrictions are. If you can't get to it, there is the link.
kirkwood: In response to Charles, there are some states that require it in the U.S.
hdv: Also in Europe, there are reporting requirements in the Netherlands. It's even required to follow a specific format.
<kirkwood__> +1
<Zakim> Sam-estoesta, you wanted to talk about reporting, requirements
<Ben_Tillyer> UK public sector bodies must follow this format: https://
<Charles> uniformity would take time
hdv: a WCAG evaluation methodology may be one way to try to make that reporting format uniform. It could make it easier for regulators who want the same thing
<kirkwood__> Regulators would be happy if you made it easier ;)
Sam-estoesta: Correct. Adding to John's comments. There are state and federal reporting requirements.
<hdv> for the record / those curious: our document requiring specific reporting formats and methodology is called 'Model toegankelijkheidsverklaring'
Sam-estoesta: Also, if you think about Canada and parts of Europe, public sector or government-identified organizations, should we be thinking about what they are currently positioning.
<kirkwood__> Laws change due to new bills put forward by disability groups too
Sam-estoesta: We're more likely to get adoption and widespread use across those public entities.
Adam: reminder to use q+ to and q+ ask then add a note.
Bernardo: I agree we need to provide everyone guidelines on showing evidence of conformance and associate reporting to conformance level.
… maybe associate the requirements for a few upper levels of conformance and keep lower level of conformance optional.
<Bernardo> Showing different ways to report on conformance
Bernardo: this is quite demanded for public services or private companies that provide services to the public service and public organizations.
… or guide them on how to show a demonstration of conformance for the guidelines.
<Ashlee> Having the google doc is very helpful! Thank you Rachael
<Sam-estoesta> There is quite a bit that is publicly available from these regulatory bodies; I wonder how much we could source from these bodies and then look for overlap
Adam: Seems like the discussion is certainly reinforcing the analysis.
Rachael: Queue is complete, moving forward.
Rachael: Reading from notes above, under heading "Sampling methodology"
Rachael: Asks for further comments to be added, or commentary on survey results.
MartijnHoutepen: I think about conformance within a web page, is this discussion around including whole sites or all applications?
Rachael: Purpose of this is to gather people's thoughts on it. We have current discussions around pages vs. paths vs. products.
… There are a lot of complexities. This is a discussion of how we help complicated large sites better deal with conformance and accessibility.
Ben_Tillyer: For people who don't think the template or method of reporting should be in conformance because of legislative bodies or industry bodies...
<shadi> +1 to Ben_Tillyer
Ben_Tillyer: we could put a caveat alongside anything that we made normative to say that if your company or org has their own rules, they supersede whatever we propose.
… second, I think that there are so many methods of reporting and they all differ so greatly. This is an issues for multinational corporations that span different industries.
<Matt_King> +1 to consistency in reporting that Ben is talking about!
Ben_Tillyer: likely because there isn't guidance from the W3C in the previous version. If we got ahead of it, we wouldn't have this problem right now.
<Jennie_Delisi> +1 to Ben
Charles: Representative sampling is a pretty well established methodology when applied to conformance generally because it was based on pages.
… If we move the scope of conformance from page to paths then it seems to me representative sampling would be opposed to that. Would it be either or, or not representative sampling of paths.
<bbailey> +1 to not sampling paths
Rachael: I don't think the path suggestion as it sits right now is to replace pages and views, but rather to supplement or provide an alternative.
<kevin> Also worth noting that representative sampling is not a method for 'conformance' to WCAG 2 but might be accepted for 'compliance' with certain regulations
<kirkwood__> could you not sample all the pages in a path?
Charles: Understanding then would be if you have scoped and claimed a path, you couldn't sample that path.
Rachael: agree
<shadi> https://
wendyreid: As for sampling, when needing to test a million pages, that is just not possible. For sampling methodology, we just do a handful.
… I think we have to make this informative. We need the space to allow for a lot of the variants that exists between different industries, regions, business sizes, and org sizes.
… I'm not against the concept of us recommending reporting formats, but would still want them to be informative...
<Charles> +1 to Wendy
wendyreid: because I don't think any one country or regions has been able to create one reporting format. The format needs to provide flexibility.
<kirkwood__> Different types of reports depending on audience: Business report, Process report, legal report? Reporting to whom?
Matt_King: We might want to look at what's being done in other complex industries and spaces to figure out what is practical in a normative sense.
<Charles> the world would be grateful for consistency if that were possible
Matt_King: You can specify aspects of what you're required to calculate or what you're required to include in a report regardless of its format.
<shadi> +1 to Matt_King
Matt_King: Let's take a look at what problems we're trying to solve and boil them down to their basics.
<kirkwood__> +1 to Matt
shadi: +1 to Matt. I think conformance claims are quite similar. They are a normative part of WCAG.
… They are not required, but they do set certain requirements if you make a conformance claims.
… If it's supposed to be a WCAG conformance claim, it has to contain certain things so there could be a similar approach to the reports.
Ashlee: Generally is no comment/neutral because I don't feel like I had enough context to have a strong opinion.
… Note that for sign language, there are videos that could be considered for sampling? It might depend on the information that's being presented.
… How would the content creator respond to a sampling exercise? We don't know if the videos are accessible.
<Sam-estoesta> With regards to complex industries and spaces to look at for examples, government-funded organizations and federally regulated organizations would be an ideal set to investigate first.
Bernardo: There are so many different websites, some informative pages and others with fully functional applications like e-commerce.
… when coming back to something, it varies a lot to the goals and the function. It should not be functional to anyone who's trying to use it.
<Zakim> Rachael, you wanted to move on to the next topic
Rachael: Referencing Google Document WCAG 3 Document Breakdown results Pt1, 3rd party vendor content (procured) heading.
<kirkwood__> 3rd party is a legal term.
<kirkwood__> +1 to IP address comment. (last comment)
Rachael: Key question is whether or not there should be a carve out for third party that allows for partial conformance when third party is not conforming on a site.
Matt_King: Reading through the comments, it seems like people who say it should and shouldn't are saying the same thing.
<kirkwood__> We should define what 1st party and 2nd party is in our case. should we call it something else?
Matt_King: There's a bit of a difficulty of interpretation. I've experienced this. There's consistency in the remarks for the must/must not/shoulds. Rachael, do you see the same thing?
Rachael: I do. The core question is should partial conformance be allows in the conformance based on the third party not being able to meet it?
<kirkwood__> who is the 1st party and who is the 2nd party here?
GreggVan: Similar to Matt's comment, the must-nots - I look at those, and see a problem. (Reviewing the list and calls out they may be mis-categoriezed)
<kirkwood__> legal issue
<kevin> +1 John
GreggVan: If we don't say that third party content, then the person who chooses the third party provider doesn't take accountability of which decisions they make for contracts.
<Zakim> Rachael, you wanted to react to GreggVan to say we can look at the survey results
GreggVan: seems that the person making that decision should be responsible. It just creates a loophole.
<kirkwood__> +1 to Gregg
Rachael: Callout that the document shared are the notes, and we can always go back to the survey comments if that's helpful or needs clarification.
Ben_Tillyer: I'm in the camp of the third party should be treated equally.
<Adam> https://
Ben_Tillyer: Wanted to mention something that's not being talked about. The lines of what is First party and what is Third part are even more complicated .
… I do coding in my spare time, and use packages written by other people when creating my own web apps.
… In the most recent project I'm working on, I have 36kb of code I've written, and there's 212 mb of other people's code that's imported.
… I have agreed to the licensing upon installing packages, but have not entered into a financial contract with the developers.
<Charles> +1 to Ben_Tillyer third party does not always mean purchased
Ben_Tillyer: a lot of these libraries have benefit, but it's the idea of the of software, there's something we need to do if this does go ahead and we need to distinguish between First and Third Party.
shadi: Want to explain my thought process and how I responded to the survey.
… The question about third party vendor content; this example was used in contracting out my website to a subcontractor, but continues to refer to section 5.4 of WCAG.
<Charles> ads a great example
shadi: which talks about the statement of partial conformance and says web pages that will later have additional content added.
… think about news or articles coming from different sources.
… To me, it fell in this question and it is currently addressed in WCAG 2 in Section 5.4 and it says you need to do everything you can within what you control.
… I think Ben mentioned other examples and nuances or different kinds of situations between outsourcing or contracting out development between user-generated content and other situations.
<bbailey> +1 to Shadi that reference to WCAG2 5.4 was spurious for this particular survey question
<Charles> even a font is third-party content
kirkwood__: I think it's confusing as far as using this legal terminology of third party content, whether that's IP addressing, or if you're talking about a CMS that points to user generated content like a blog.
<shadi> +1 to Charles
kirkwood__: Understanding of what it is from a technical perspective rather than touching on the legal perspective - we're going to have a lot of trouble with that.
wendyreid: We definitely need a more solid definition of what we mean by this type of content.
<kirkwood__> the iframe points to a different ip address?
wendyreid: There's probably infinite examples of these types of situations, and they are standard in the industry today especially wen they're technical problems.
… some of this is controllable if you have a direct contractual relationship with an agency because you control the contract terms.
… but for an example like a payment processor, as that integration is being updated, it could work one day and break the next; this makes it really hard.
… We do have to carve out this content, what the scope means, and how to report on it.
<kirkwood__> once you report on it do you become responsible?
wendyreid: Should a website be punished because of the actions of one of its vendors? Probably not. Where there's a power imbalance, you're not going to have control in that situation.
<kirkwood__> +2 to Wendy
<Ashlee> +1 to wendyreid
GreggVan: Want to +1 Wendy's comments and examples.
… one of the things you have control over is who you pick.
… for the ATMs, they are never connected to a bank. ATMs are always connected to a host. The banks pointed at each other and said they couldn't support it.
… when the Access Board said they had to, suddenly, all of them were accessible.
… If a company is working with big or small vendors to offer accessible third party solutions, then it can get done.
… When people say they don't have control over a part of their site, like a side bar, then people need to complain to exert control.
<Zakim> Rachael, you wanted to react to GreggVan
GreggVan: Control is a funny animal. It's going to be complicated. In some cases it's unfair or unachievable. It gets ironed out in enforcement.
<laura> +1 to Gregg
GreggVan: We should do things because that's the way it should be done, and the exceptions will get ironed out in enforcement.
<Zakim> Ben_Tillyer, you wanted to respond to shadi
Rachael: To sum up, we're going to take all the comments and all these great conversations, and come back with suggestions to move forward.
… Rachael needs to leave the meeting a little early.
Ben_Tillyer: To respond to Shadi - if you have an advertising space or news coming from some news source, you design and decide the requirements for input.
… there are many ways to get and consume content from third parties.
… I think what Wendy said touches on the same topic. If you had to diligently review content from an ad provider, and you were sure it was accessible and put it on your site...
<kirkwood__> confusing of user has been an issue in some cases. Reasonably thinking it’s your content. [frames,ads etc]
Ben_Tillyer: You could be notified of the fact that it wasn't accessible and the truth would come out.
… Why should people be allowed to claim conformance when it's not accessible anymore?
Matt_King: Great comments and examples from Wendy, Gregg, and Ben there.
… The linkage between some of these topics like reporting third-party content, sampling, etc. What is the problem we're trying to solve?
… Do we want companies/organizations to acknowledge that they have inaccesslb content?
… Is it that we want an evaluation or reporting on that?
… None of these concepts can have a decision be made about them when they are standing on their own. We need to decide what to do based on the problem that we're trying to solve.
<kirkwood__> you can clearly attribute content on a site / 3rd party (legal)
giacomo-petri: Want to add a point about feasibility and the cost. There's an evaluation cost to evaluate the third-party content.
… What if there was a way for organizations to claim conformance for their own product, and clearly state that the claim applies only to what they control...
… and list the third-party components, leaving the responsibility for their own conformance claim. This is also a matter of company maturity, maybe evaluate available options, understanding, and identifying differences.
Adam: Leveraging assertions is an interesting idea.
<Bernardo> Most websites are using building blocks from others (SaaS, payment, content management, Support channels... etc)
GreggVan: My concern with Giacommo's comment. Just because I'm using everyone eles's tools to buidl something, that doesn't mean I shouldn't have any
<kirkwood__> reality: there is a split legal ‘responsibility’.. we just definewhat is not accessible. other is a matter of ownership & burden etc
GreggVan: It's about what you deliver and being accountable for it.
<laura> +1 to Gregg.
<kirkwood__> +1 to Gregg
shadi: Section 5.4 in WCAG2 does not say that you can call out parts of your website that may be conformant.
Adam: That brings us to the end. Hearing a lot of general alignment in terms of so-called third-party vendor content.
… Accessibility barriers should not escape accountability. It's a matter of how we define it, and how far WCAG could or should go in terms of trying to enforce it.
… Reminder to join 30 minutes early for onboarding session next week.
… Have a good day.