Challenges: We Aren't Striking at the Root

[mgifford]

One of the biggest challenges to scaling conformance testing is that we aren't thinking about and investing in the systems that produce the modern web.

My experience comes from a decade or so of working with Drupal, trying to fix the problems up-stream to see that people are at least starting with reasonable defaults.

All of the resources on accessibility are focused on site specific accessibility problems, but this ignores the ecosystems of software libraries that define the modern internet.

In my A11yAxioms I advocate for community based approaches to accessibility rather than institutional.

With Drupal, accessibility problems we fix in Core get applied to over a million websites when the site owners decide to upgrade their site. This is huge.

If you know that your site is starting with something with good accessibility defaults, it is much easier to test it for compliance as you can know where to focus your testing efforts.

[patrickhlauke]

who is the "we" in this? AGWG? nothing is stopping developers who are involved in systems like Drupal (or, for my own part, Bootstrap) from applying WCAG requirements to their work. WCAG defines what the end result should look like/behave like/contain. developers then need to transpose that to the specific way in which their own site is built.

is this something you're suggesting for the Education and Outreach part of WAI?

[edit: and now, checking up on my unread emails/RSS feeds, I get that this is likely related to https://www.w3.org/TR/accessibility-conformance-challenges/ ... which now makes a bit more sense]

[mgifford]

Sorry for the lack of context @patrickhlauke - and ya, I've been reading this as what are some of the challenges for scaling WCAG to large, dynamic sites.

And yes, there is nothing in WCAG to stop a developer from fixing a system like Drupal or Bootstrap, but there is nothing to encourage it either. If we want to talk about making accessibility scalable we have to get people fixing bugs where they originate rather than creating work-arounds that may temporarily fix the site in question.

[patrickhlauke]

the problem is though who is meant to "encourage" these developers? w3c? WAI? and who foots the bill (even if it's "just" developer time)? i.e. who is doing the asking, and who are they actually addressing?

[JaninaSajka]

There was once an [https://www.w3.org/WAI/standards-guidelines/atag/](Authoring Tools Working Group) in W3C/WAI with the mandate to develop guidance on tooling to support web accessibility, together with a [https://www.w3.org/WAI/fundamentals/components/](view of essential components) that all needed to work together. All that work of at least 3 Working Groups was folded into the new [https://www.w3.org/2017/01/ag-charter.html](Accessibility Guidelines Working Group) early in 2017. I suspect, though, AGWG has been too consumed by WCAG 2.1, and now 2.2 to have much energy left for the old ATAG and UAAG work products, though Silver (WCAG 3.0) is trying to comprehend some of that. I would also note similar issues were raised by Oracle representatives during the AGWG discussions preceding FPWD publication of the Challenges document, and those comments are captured as subitems under the first question in the [https://www.w3.org/blog/2020/06/accessibility-conformance-challenges-draft/](W3C blog post announcing this document), though they're not in the Status Section of this published version itself. We continue open to text submissions to address these and additional challenges. Tbhere's no reason we need stop at 5 specific Challenge sections--there's room (and arguably need) for more.

[WayneEDick]

I think the Guidelines are a system of axioms. Because they need to be exact, they appear to be abstruse. That cannot be avoided. When you try to express truths, you must define context, assumptions, scope and limitations carefully. That limits what can be said, because what we assert must be true. I really like systems of principles, and that is what I think you are attempting to express. It helps us find the places where we can begin to define axions.

[mgifford]

@patrickhlauke - I do think that W3C documentation should acknowledge that digital tools are built with common libraries. We use those libraries as it saves us time & money. Part of the responsibility of using that library is to contribute back to it (bugs or features). The WAI can acknowledge this.

As far as who pays for the up-stream fixes, this is a good question. I'm convinced collectively we'd be paying less and getting more, if we took this approach.

I would like to see governments and big business taking a leadership role in footing the bill. It is nice to be involved in projects like the We4Authors Cluster which is the first initiative I know of where governments are investing in supporting initiatives that contribute to ATAG 2.0 - Part B. In the Drupal community, we've managed to do quite a lot, but there is always more to do.

I'm just asking for the WAI to include fixing the problems up-stream as being an important part of seeing that future Webaim Million studies get better.

@JaninaSajka - We need leadership to make it clear that accessibility can't continue to be overlooked. Going back to the source, procurement is such an important part of this problem. If IT is choosing their frameworks with accessibility being a key requirement, ATAG needs to be in RFPs, common patterns from popular CMS's will help with browser and assistive technology support. Decentralized teams are normal, and we need to see that accessibility champions are included in those distributed teams. Culture change is way more powerful, than the bottleneck accessibility authorities that lots of organizations have.

@WayneEDick - Thanks. People can consume tweets. People will read the GDS posters. WCAG guidelines have been good for putting most people to sleep.