Expand "Understanding Levels of Conformance" - #5303
Conversation
Based on the lengthy discussions in #3889 - which kept stressing the idea that no, Level A/AA/AAA does not try to imply any kind of hierarchy of importance/severity - this PR rewrites and expands this section, to hopefully provide some more clarification. * removes the use of the word "essential" - this currently very clearly *does* imply that some SCs are "more essential" (thus more important/severe) than others * moves the bullet about workarounds right after the first one, as it's related conceptually * adds a bullet about subjectivity (yes, it's a dirty word that WCAG doesn't want to acknowledge most of the time...but it's true) * gives an example after the bullet about what these bullet points actually mean - this is primarily what was missing in the "WCAG doesn't explain itself about why levels were chosen the way they were" * adds a call-back to WCAG 1.0 - where "priority" was used - and explains how this concept is NOT used in WCAG 2. further hammers home the idea that all SCs are important, and that deciding how to prioritise or even what level to aim for are decisions for developers/policy makers in the end Closes #3889
✅ Deploy Preview for wcag2 ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
To be clear: I think this is going in a good direction and improves the document. |
Co-authored-by: Patrick H. Lauke <redux@splintered.co.uk>
porting some of the gist of #5309 directly here
I disagree... there are many cases where subjectivity plays a role (heck, even just looking at 1.1.1 and deciding whether a text alternative is appropriate/correct/sufficient or not comes down to subjectivity - there are no absolutely objective measures)
i posit that even the categorisation of an SC into a particular level had ... shades of subjectivity that went beyond just the stated bullets ... so instead of pretending retrospectively that this was all done purely on those stated points, this acknowledges that the process was not fully mathematical/objective either (noting that the "Some of the common factors evaluated when setting the level included:" sentence has already been in this document since its inception, and is not a change introduced here) |
|
imagining what would happen if all that argumentative energy was instead applied to fixing and improving parts of wcag that actually mattered... |
|
Rather than ballooning out, I'd replace "criteria that address problems
that can't be addressed..." with a specific level-A criterion, and explain
how it got assigned the level it did. Then (_mutatis mutandi_ which to
spare a Latin grammar lesson means "changing the relevant details" but in
broad terms implies repeating the exercise) add an example for something
that got assigned to double- and triple-A.
Yes, there is always some nuance and clear concrete statements are
regularly less completely true than they might claim to be - but if they
are true enough to be able to work with them, I don't have a problem with
being clear about that - I think it helps shortcut a lot of really not very
worthwhile arguments hinging on "what-about...".
|
because it closes an open issue that was filed against the spec, and which caused a LOT of discussion. but hey ho. |
|
There might be details from WCAG 2.2 levels discussion worth harvesting for this update. |
Feel free to contribute further to this PR accordingly |
|
On backlog call 21 Aug, suggestion to add something like “the level assigned to a success criterion is not indicative of accessibility issue severity or priority” and use examples even of 1.1.1. (Sometimes missing alt is a hard stop, sometimes it’s trivial.) |
There was a problem hiding this comment.
I don't object to the mention of WCAG 1.0 and priorities. I don't object to roughly equating priorities with levels.
What I object to is the implication that there is significant correlation between Priority 1/2/3 Checkpoints and Level A/AA/AAA SC. There is some correlation but those tables show that the mapping is all over the place.
Co-authored-by: Patrick H. Lauke <redux@splintered.co.uk>
there's no implication. it's a statement of fact that in WCAG 1.0 priorities mapped to levels https://www.w3.org/TR/WCAG10/#Conformance ... or am i missing something? the table you point to is the mapping from WCAG 1.0 to WCAG 2.0, which is not what this sentence is talking about. i edited this sentence and the following one to make this crystal clear, hopefully |
…insic to the SC itself, but depends on what exactly is failing (as suggested by @giacomo-petri in last week's meeting)
|
What I object to is the implication that there is significant correlation between WCAG 1.0 Priority 1/2/3 Checkpoints and WCAG 2.x Level A/AA/AAA. Sorry not to be clear about that. I also don't think WCAG 1.0 Level A is all that similar to WCAG 2.x Level A. |
bruce-usab
left a comment
There was a problem hiding this comment.
The most recent edit resolves my concern, thank you!
Based on the lengthy discussions in #3889 - which kept stressing the idea that no, Level A/AA/AAA does not try to imply any kind of hierarchy of importance/severity (though this may perhaps be more wishful thinking) - this PR rewrites and expands this section, to hopefully provide some more clarification.
Closes #3889
Preview: https://deploy-preview-5303--wcag2.netlify.app/understanding/conformance#levels
Diff: https://services.w3.org/htmldiff?doc1=https%3A%2F%2Fwww.w3.org%2FWAI%2FWCAG22%2FUnderstanding%2Fconformance%23levels&doc2=https%3A%2F%2Fdeploy-preview-5303--wcag2.netlify.app%2Funderstanding%2Fconformance%23levels#levels