Introduce "complete conformance" to conformance level section #193
Conversation
Here's a suggestion of how I'm thinking we can resolve the issue I raised on AG on Tuesday. I think it is important to keep a clear distinction between what the requirements are, and what the margin of error is that we're allowing on it. Put in other words, it needs to be clear that every image missing an alt attribute is an accessibility issue; That it isn't OK for content authors to intentionally pick images that are difficult to describe and leave the alt attribute off. This definitely needs more work, but it shows the direction I'm thinking for a solution. I do think it is important that we address this in FPWD, even if it is only a partial solution, some acknowledgement of the problem is important in my opinion.
|
Here's another approach, building more on the LEED rating system. LEED uses “certification levels”—Certified, Silver, Gold or Platinum—to certify the degree to which a building conforms with LEED requirements. Similarly, we could have ”certification levels“ to certify the degree to which a digital resource conforms with WCAG 3 requirements. By calling the levels “certification levels” rather than “conformance levels”, we are able to acknowledge progress and encourage additional efforts without allowing/assigning a conformance claim on a product that has areas of non-conformance with WCAG 3 requirements—areas that might present barriers that result in exclusion. So we would have “WCAG 3 conformance,” which is a measure of how much a product conforms with WCAG 3 requirements, and “WCAG 3 certification,” which recognizes substantial conformance with WCAG 3 requirements along with other activities. Separating certification and conformance with WCAG 3 requirements will give us more flexibility. It opens the door to ways to earn credits and “bonus points” to achieve higher certification levels, such as points for organizational maturity and holistic testing that may not be WCAG 3 requirements. Hope this is helpful, and apologies if it has already been covered and considered. |
|
+1 Thanks for adding Sarah, I do like the idea very much! |
|
I really like the potential utility of shifting from "conformance levels” to “certification levels”. [EDIT: fixed my typo.] |
|
Argh, I don't know what I did but I think I closed this? Sorry. Bruce, I meant to ask whether you meant "shifting from "conformance levels” to “certification levels”." |
|
Great idea Sarah. I think you're right, a different name for conformance levels is appropriate I think, given the shifting direction. |
|
@rachaelbradley Would you like me to try to rework this PR according to @gradualclearing's suggestion? |
|
@WilcoFiers Not at this time. This is a valuable proposal but it is a large change that needs broader conversation. We have added an editor's note to address this in the FPWD so AGWG and the Silver taskforce can discuss it further going forward and also get public comment on it. |
|
@rachaelbradley can you link me to that editors note you are referring to? I went through the document and could not find it. |
|
… On Tue, Oct 13, 2020 at 5:31 AM Wilco Fiers ***@***.***> wrote:
@rachaelbradley <https://github.com/rachaelbradley> can you link me to
that editors note you are referring to? I went through the document and
could not find it.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#193 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AFBUPBRDFE3ODIBKOGV2H4LSKQM7VANCNFSM4SIP43LA>
.
|
|
Hi @WilcoFiers @gradualclearing et al I mentioned this in the Visual Contrast subgroup, but on stumbling upon this issue 193, I think it might fit here, as opposed to its own issue. Revised: Scoring PsychologyAs I was exploring ideas for 0-4 levels of scoring for Visual Contrast, something occurred to me. Numbers carry no emotional weight(with the possible exceptions of 0, 7, and 13, LOL) The Abstract:Levels 0-4 or 1-4 are an abstraction. In our paradigm, 4 is best… but in some other paradigms, 1 is best. Of course adding in a 0 level makes it more clear that 1 is less desirable than 4 right?…. Or does it? Oh wait, no, there are plenty of examples where a zero score is a best score. Why this matters:The abstract nature of numerical values is useful for ordering things, but not immediately communicative of the value. Presently made worse by ordering the score list with the zero on top and the four at the bottom of all the Silver score lists i’ve seen, which is reverse of the conventional wisdom of best on top. In addition, communication can be substantially improved by associating each numbered level with a descriptive adjective:
This immediately makes it obvious what is important, and what is sub-standard. Adding such simple descriptors not only clarifies meaning, it adds an important element of psychological weighting to the understanding and value of each level. For example, it's easy for a manager to say “oh level 1 or 2 passes, so that’s fine” but it’s not so easy to set a goal like “oh poor or marginal is fine.” Who strives for marginal? No one of merit. We want people to strive for ideal, with a safe fallback to acceptable.We do not want to encourage people going for level 1 as if it was an acceptable minimum. Using a word like “poor” helps to end that problem — a problem I might add exists with the current WCAG 2.x, where people go for "AA" as good enough but ignore the value of "AAA", a problem that plagues contrast choices among other things. AA and AAA are also somewhat abstract. How much better would it have worked if the terms were "Minimum Accomodation" for AA and "Preferred Accommodation" for AAA? Or alternately "Optimum" for AAA vs "Ordinary" for AA. IMO, plain language and clear words are more than just eliminating jargon, but also a "modestly opinionated" approach, promoting the value of best practices and encouraging adoption of higher levels of performance. Saying outright "This is best" as an authority means the reader does not need expend their own effort in deeper-dives to come to that same conclusion. Humans are emotional animals and words can engage and motivate, even if subtly. In fact it is what we are used to, inundated daily with messages tailored by behavioural psychologists, there is some expectation of descriptive terms. This is how we break through the noise and solidify the message. The importance of being more clear in the meaning of each level on first glance, by using simple terms with a modest emo-weight, cannot be underestimated in its value. Standards documents are already highly arcane and often abstracted, and everything we can do to simplify and make “understanding immediate without digging” is a step toward general acceptance and adoption of these guidelines by designers and the public. Thank you, Andy Andrew Somers |
jspellman
added
action: discuss@meeting
jspellman
added
status: assigned to subgroup
Here's a suggestion of how I'm thinking we can resolve the issue I raised on AG on Tuesday. I think it is important to keep a clear distinction between what the requirements are, and what the margin of error is that we're allowing on it. Put in other words, it needs to be clear that every image missing an alt attribute is an accessibility issue; That it isn't OK for content authors to intentionally pick images that are difficult to describe and leave the alt attribute off.
This definitely needs more work, but it shows the direction I'm thinking for a solution. I do think it is important that we address this in FPWD, even if it is only a partial solution, some acknowledgement of the problem is important in my opinion.