w3c / silver

Accessibility Guidelines "Silver"
https://w3c.github.io/silver/
Other
199 stars 44 forks source link

Conformance Levels & VPATs #269

Open cheitman12 opened 3 years ago

cheitman12 commented 3 years ago

I've been reading over the new conformance levels section, and I really like the path that this is going down. I've noticed when I talk to people at conferences I am speaking at that a lot of companies don't even know where to start their accessibility program, and I think that this will help solve that. For example: I could say something like start with Bronze compliance and work your way higher once you achieve that. Versus the current just do all the things for level AA which is hard and confusing to explain.

What do we see the future of VPATs looking like with 3.0? I write close to 25 a year for my company's platforms, and we give them out to clients on a weekly basis for their accessibility reviews. I'm trying to figure out how the rating system, plus listing where the issues are at, your testing methodologies, etc. is going to fit into a document of some sort that is readable to people who aren't accessibility professionals. We have a lot of clients who know they should care about accessibility, but don't necessarily have an expert on staff to decipher a complex VPAT.

Myndex commented 3 years ago

Hi Cortney, @courtney-heitman thank you for the comment.

I can't speak for the other sub groups, but for Visual Contrast, we have a multi level scoring using font size and weight relative to a specific contrast. The dynamic tool is https://www.myndex.com/APCA/

These score levels apply to CONTENT text, i.e. text intended to be read: headlines, body text, menu items.

Copyright, by line, etc. have a lower standard of acceptibility, as noted on the lookup table. So does non-text like buttons.

The top, "Preferred" level,

is a "suggested best target", but does not in itself count toward a score. It is a preference for what is best to see sites, but the requirements are a bit challenging to make it a reasonable normative level (it far exceeds the current AAA).

Score 4 is the ideal minimum.

This is the minimum sites should attempt to conform to.

Score 3 is the acceptable minimum,

intended as a safety fall through for projects that attempted, but missed slightly on score 4

Score 2 is the deficient minimum.

It should never be a goal, but is intended as a way to pass sites and provide feedback on corrective measures.

Score 1 is POOR, the minimum passing level

This is a catch-all for sites that are passing some current standards, but that need revision to pass this emerging standard. As in Score 2, it is a way to pass a site while providing guidance on corrective actions.

Score 0 is a fail

This means the site was unable to even muster a score level 1, and corrective action is required.

There is more on this and discussion on the Visual Contrast Wiki: https://www.w3.org/WAI/GL/task-forces/silver/wiki/Visual_Contrast_of_Text_Subgroup

And is obviously a work in progress. Please let me knof of any questions.

Thank you!

Andy

Andrew Somers W2 Invited Expert Myndex. Color Science

cheitman12 commented 3 years ago

Thanks! That kind of helps.

I guess my question was more towards how we're doing the ratings on a website level - since the ratings are on a per page basis and often times a VPAT is for 100s of pages at once. Currently in a VPAT, we assign Supports, Partially Supports, Does Not Support or Not Applicable to each of the WCAG criterion, and then we note what the issues are and where they are at. With WCAG 3.0 are we going to assign an average rating to each criterion instead?

Example: I have a website that has 150 pages, 100 of the pages are a 3, 5 pages have major issues and are a 1, 40 are a 4, and the last 5 are a 5. Would I just give that a 3 since that's where the majority of the pages ratings are at? Or are we going to have to list the rating for each page?

I'm just concerned with how to make the ratings scales user friendly and easily consumable in a summarized form for a large scale websites.

Myndex commented 3 years ago

At the moment, visual contrast is per the lowest-rated content element. Thus, a 1. Again, the idea behind 1 is to allow a pass for highly deficient content to allow a site to correct it.

Based on your example, I could see possibly setting a threshold for number of "bad" pages and how they might impact an aggregate score. Something we'll discuss in the nexgt visual contrast meeting, thank you.

cheitman12 commented 3 years ago

Thank you!

jspellman commented 3 years ago

@courtney-heitman, the group working on WCAG3 have not yet addressed this issue, but we know it is on our long-term list of actions. We want to have a stable version of Bronze before we approach various groups about updating VPAT template. I had a short introductory meeting with members of the Access Board and Trusted Tester in February. From the discussion, I would conclude they are interested but are waiting to see how the draft develops.

We will be working on an official response from the group and will close this issue within the next few weeks. We have a lot of comments that we are working on. :)

cheitman12 commented 3 years ago

Thanks Jeanne! I assumed you have quite a few issues to go through. I'm very curious to see the thoughts from the group.

bruce-usab commented 3 years ago

@courtney-heitman, as someone who was working early in 508 when industry came up with the VPAT approach, I just want to give a big +1 to Jeanne's reply. VPATs and ACRs are quite important of course, and on the mind of many people working on WCAG3, but not specifically in scope for the AG WG or Silver Task Scope.

ChrisLoiselle commented 3 years ago

@jspellman This was labeled ready for survey then removed. For Visual Contrast, we have a multi level scoring using font size and weight relative to a specific contrast. The dynamic tool is https://www.myndex.com/APCA/ . The scoring concerns are in development in Silver so this is ready for survey, but the topic will be open until scoring and conformance groups are finished with their work.