Archive index

A11y Slackers Gitter Channel Archive 10th of May 2017

What fresh hell is THIS now? - Patrick Lauke
  1. zakim-robot
    @zakim-robot
    May 10 00:56
    [alice] (whoa those rainbow skin toned thumbs ups are awesome)
  2. [alice] (For those who can't see them - they are thumbs-up "reacjis" that cycle/fade through all the supported skin tone colours)
  3. zakim-robot
    @zakim-robot
    May 10 01:14
    [albert] i saw a tweet today from a “learn to code” company person that dismissed including <head> and <body> in html documents. what say ye, a11y?
  4. zakim-robot
    @zakim-robot
    May 10 01:26
    [karlgroves] Well, that’s one way of doing it
  5. zakim-robot
    @zakim-robot
    May 10 02:24
    [jamesn] I don't think it would cause any a11y issues but I would feel dirty doing it.
  6. [marcysutton] You wouldn't be able to specify a page title or HTML Lang though...? That seems wrong
  7. zakim-robot
    @zakim-robot
    May 10 02:33

    [jaredsmith] If you don't add <head> or <body>, the browser will auto-generate them. Run this through the HTML validator - it's a perfectly valid and WCAG 2.0 conformant web page (though not necessarily a good idea):
    `<!doctype html>

    <html lang=en>

    <meta charset=utf-8>

    <title>blah</title>

    <p>I'm the content`

  8. stevefaulkner
    @stevefaulkner
    May 10 05:40
    @mfairchild365 It is no longer an issue in Firefox https://bugzilla.mozilla.org/show_bug.cgi?id=693948
  9. zakim-robot
    @zakim-robot
    May 10 13:07
    [oliviernourry] Hi all, I'm currently testing a page with a form, where error messages are in a block container that has role=alert on top of the form. This container is in the DOM, displayed after submission only. I didn't expect that, but the alert is spoken out on page load (by VO on Safari/Mac OS at least, have to check on others). I was surprised because my understanding of role=alert was that it's triggered only when the content changes, not at its first display. I was about to invalidate this, but then I realized the error message is the first thing you hear when you get this page, which is rather relevant: you know why you are here, and that you have something to do about it. What do you think about that? Any adverse effect you could think of?
    Also, out of curiosity: is this a VoiceOverism, or does it work as it should?
  10. zakim-robot
    @zakim-robot
    May 10 14:54
    [sethkane] Can someone explain to me why maxlength isn't communicated to a screen reader. In JAWS and NVDA the user can keep typing but it is never notified that they have reached a limit. Thoughts? I can think of a few ways to solve for this but I just want to know if there is a rationale reason why this isn't a default behavoir with screen readers.
  11. zakim-robot
    @zakim-robot
    May 10 14:59
    [tink] Don't know about a rationale, but as a screen reader user I don't ever recall coming up against this as a prblem.
  12. [sethkane] @tink interesting, thank you for your response.
  13. [tink] Has this come up in usability testing for you?
  14. zakim-robot
    @zakim-robot
    May 10 15:05
    [alastc] Ah, that could be a problem in itself! Visually a maxlength stops you entering more text, but doesn’t stop you typing (obviously), I wonder if some of your form entries have been shorter than you thought? I might not be apparent until well after the form itself, kind of an invisible error. It seems unlikely that’s the case though, it would be a data-loss issue a screenreader maker would want to avoid.
  15. [tink] I've never noticed information being truncated after the fact - doesn't mean it hasn't happened of course, just not that I've noticed :)
  16. [car] @sethkane It seems that there is no platform accessibility api mapping for it? https://www.w3.org/TR/html-aam-1.0/#att-maxlength
  17. [sethkane] We have a client that is calling it a defect in NVDA and after testing it myself in NVDA, JAWS and VO I would jsut made an assumption that if a input field required only eight characterrs and the users wasn't notified of that ahead of time like in a label they would never know that they reached the limit. Now like I said I can think of a few ways to solve for this but I was wondering why the browser object and the screen readers are sync on this topic.
  18. [sethkane] @car That is a huge help! Thanks
  19. zakim-robot
    @zakim-robot
    May 10 15:10
    [tink] I would classify it as a screen reader defect, yes.
  20. Mark Holmes
    @markholmes
    May 10 16:40
    i did not know this channel existed
  21. but i'm very happy that it does
  22. Steven Reynolds
    @stvnrynlds
    May 10 16:43
    Seriously, very happy
  23. zakim-robot
    @zakim-robot
    May 10 17:15
    [mayabenari] anyone know of a good primer on the different assistive devices people use?
  24. zakim-robot
    @zakim-robot
    May 10 18:11
    [karlgroves] @mayabenari Not sure if one single resource exists. WebAIM has good resources https://www.google.com/search?q=site%3Awebaim.org+assistive+technologies
  25. zakim-robot
    @zakim-robot
    May 10 18:56
    [hellomelo] Hi everyone! I have a question regarding link text (https://www.w3.org/TR/UNDERSTANDING-WCAG20/navigation-mechanisms-link.html). Currently we have a page with two donate buttons. The first says "Make a Monthly donation" and the second says "Make a one time donation". Is this approach considered inaccessible because both buttons, while having different link text, actually take the user to the exact same page?
  26. zakim-robot
    @zakim-robot
    May 10 19:05
    [backwardok] is there a reason why you have two buttons that go to the exact same page instead of one button that’s more general?
  27. [hellomelo] I believe the client wanted to be explicit that there were two donation options... but yes, as I wrote out that question I think I answered myself.
  28. zakim-robot
    @zakim-robot
    May 10 20:01
    [cmegown] has it been suggested that one button that goes to one page with something like a set of radio buttons would be better?
  29. [cmegown] at least i feel like that would be more expected...
  30. [cmegown] if they won't let go of the two button thing at least make the button they select set the related radio button or something