WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: To what degree does failure to convey structure violate 1.3.1 or other success criteria?

for

From: Birkir R. Gunnarsson
Date: Feb 6, 2018 7:56AM

ON your first example, my understanding is that there are plans to
change the <address> definition so that it can be used more widely.
Maybe that's already true in HTML 5.2, unfortunately my browser/screen
reader combos all crash when I try to load that page.
Here is the MdN article:
https://developer.mozilla.org/en-US/docs/Web/HTML/Element/address

I think your reading of WCAG 1.3.1 is correct.
If no information is communicated visually it doesn't have to be
communicated programmatically.
Accessibility is about equal access, if something is not marked up as
a heading for anybody it's a usability issue, not an accessibility
issue.

You can still recommend to use the appropriate HtML for the job for
other benefits, and you could try to argue that 4.1.1 applies to some
degree, though I think that is an overly liberal interpretation.
And 4.1.2 is the catch all accessibility SC that we can use when all else fails.
I also agree with you that 2.4.6 is one of the weakest success
criteria. I"m not even sure if an empty heading violates 2.4.6.
I've not used it often and I've ended up in long discussions with
other experts in situations where I have.
I like the intension and with a stronger set of technique examples,
e.g. creaing a failure techniques under 24.6 for an empty heading
element, this could be <addressed>.
heading as a fila b

On 2/6/18, Robert Fentress < <EMAIL REMOVED> > wrote:
> To what degree is conveying information and relationships a required thing
> under WCAG Success Criteria (SC) 1.3.1 Info and Relationships
> <https://www.w3.org/WAI/WCAG20/quickref/#qr-content-structure-separation-programmatic>
> ?
>
> As a reminder, that SC reads, "Information, structure, and relationships
> conveyed through presentation can be programmatically determined or are
> available in text."
>
> A strict reading of the text of 1.3.1 seems to indicate that the issue is
> not *whether* information and relationships are conveyed, but rather that
> any information and relationships that are *conveyed through presentation*
> are also conveyed semantically or with text. An example of something that
> you might think violates 1.3.1 but that, if you just look at the plain
> words in the SC, is *not* wrong is when the <address> tag is used to mark
> up any old address, when, semantically, it should only used to define the
> contact information for the author/owner of a document. So, the tag is
> used incorrectly there, but there is nothing in the presentation that
> suggests this. Would this violate 1.3.1?
>
> Similarly, I have always thought that using asterisk characters to denote
> bulleted lists violated 1.3.1, but, since the info and relationships are
> "available in text" through the use of the asterisk character, one could
> argue SC 1.3.1 is not violated.
>
> Another possible candidate that might catch this sort of thing is SC 2.4.6
> Headings and Labels
> <https://www.w3.org/WAI/WCAG20/quickref/#qr-navigation-mechanisms-descriptive>,
> which reads, "Headings and labels describe topic or purpose." Would
> information about the type of element that is being exposed count as a
> "label" under this success criteria? Reading the Understanding 2.4.6 page
> <http://www.w3.org/TR/UNDERSTANDING-WCAG20/navigation-mechanisms-descriptive.html>;
> makes
> clear they are not only referring to the <label> HTML element, but I find
> it a little squishy, based on the wording there and the rather anemic
> techniques listing, what *exactly* that term entails.
>
> Finally, it is possible that SC 4.1.2 Name, Role, Value
> <http://www.w3.org/TR/UNDERSTANDING-WCAG20/ensure-compat-rsv.html>; might be
> violated by this sort of thing, since it requires that the name and role be
> programmatically determined. However, this only applies to user interface
> components, which lists and addresses are not.
>
> Another note: If we *do* decide to interpret SC 1.3.1 or SC 2.4.6 more
> liberally to include incorrect or inadequate semantic encoding, where do we
> draw the line in terms of marking something as technically violating a
> criterion? Does a failure to use definition lists, where appropriate, or
> using unordered lists instead, count as a violation? How about <dfn> or
> <cite>? Where do we draw the line?
>
> If you've gotten this far, you obviously have too much time on your hands
> and should get back to work (winky face)! But, seriously, I'd appreciate
> the community's thoughts on the matter. I feel like I might be missing
> something obvious.
>
> Best,
> Rob
>
> --
> Rob Fentress
> Senior Accessibility Solutions Designer
> Assistive Technologies at Virginia Tech
> Electronic Business Card (vCard)
> <http://search.vt.edu/search/person.vcf?person54847>
> LinkedIn Profile
> <https://www.linkedin.com/in/rob-fentress-aa0b609?trk=profile-badge>
> > > > >


--
Work hard. Have fun. Make history.