Archive index

A11y Slackers Gitter Channel Archive 16th of July 2016

What fresh hell is THIS now? - Patrick Lauke
  1. Thierry Koblentz
    @thierryk
    Jul 16 02:01
    I'm afraid that if devs do not use <dl> because of the sketchy support they get from AT then we'll never have a chance to see these issues get fixed...
  2. James Nurthen
    @jnurthen
    Jul 16 02:04
    It's not like DL is new. Why do you think more people using it now will change the support?
  3. Thierry Koblentz
    @thierryk
    Jul 16 03:42
    @jnurthen Because of "prioritization" (why would AT spend resources on fixing <dl> when nobody use them? ).
  4. Mallory
    @StommePoes
    Jul 16 07:34
    I stubbornly continue to use DLs.
  5. zakim-robot
    @zakim-robot
    Jul 16 09:10
    [michiel] ++ DLs
  6. zakim-robot
    @zakim-robot
    Jul 16 13:59
    [krisbulman] If we don't use them, support will continue to suffer.
  7. stevefaulkner
    @stevefaulkner
    Jul 16 15:55

    @eagsalazar

    @stevefaulkner That is interesting because http://www.html5accessibility.com/, if you look at header specifically, versus what http://validator.w3.org/nu/?doc=http%3A%2F%2Ftink.uk%2F marks as "unnecessary" don't match up. I get that everything is supported and mapped to the accessibility tree for html4, it looks really spotty for html5, contrary to what your article is suggesting. What's the right answer?

    Funnily enough I am responsible for both HTML5acc and the conformance requirements for use of ARIA in HTML. I updated header info in http://www.html5accessibility.com/ to reflect Chrome's support more clearly and filed a bug about issues. The right answer is if you want to support legacy then you can add the roles for landmarks to native HTML5 elements that map to landmarks. In the case of header/footer ONLY when the elements are scoped to the body element. The validator throws a warning as its a SHOULD NOT requirement in HTML, it doesn't mean you can't but it is recommended not to.

  8. zakim-robot
    @zakim-robot
    Jul 16 22:17
    [cameron] Any work being done to surface accessibility properties of DOM elements via javascript? That would be <3
  9. [cameron] side note: Steve, just saw the html5accessibility.com redesign. Looks hot!
  10. [cameron] that's where I was inspired to ask the question above in fact
  11. [cameron] > If a feature is supported, the next step is to test if it is correctly mapped to the platform accessibility layer. This must be tested manually.
  12. James Nurthen
    @jnurthen
    Jul 16 22:29
    @cameron there is some discussion in wigc. https://discourse.wicg.io/t/dom-apis-to-expose-accessible-role-states-properties/693/4
  13. zakim-robot
    @zakim-robot
    Jul 16 22:35
    [cameron] +1
  14. [cameron] thanks @jnurthen!