Archive index

A11y Slackers Gitter Channel Archive 5th of January 2017

What fresh hell is THIS now? - Patrick Lauke
  1. zakim-robot
    @zakim-robot
    Jan 05 08:23
    [iandevlin] Hello everyone! I am currently looking for up to date information on the accessibility of Web Components and the Shadow DOM. I have come across some great articles by @marcysutton, but they were written in 2014, so I was wondering if they still hold true or if there is something else more up to date? Thanks in advance!
  2. zakim-robot
    @zakim-robot
    Jan 05 11:25
    [michiel] Mornin’ slackers
  3. Job van Achterberg
    @jkva
    Jan 05 11:29
    @michiel o/
  4. zakim-robot
    @zakim-robot
    Jan 05 11:31
    [michiel] @jkva \o
  5. stevefaulkner
    @stevefaulkner
    Jan 05 11:41
    @StommePoes , @marcysutton 'summary' is an interactive element, it is either exposed as a button or a disclosure triangle (window/Mac) neither of which allow child interactive elements. the HTML spec(s) does not reflect this. Have filed an issue so the discepency between spec and implementation reality (firefox/webkit/chrome) is fixed
  6. zakim-robot
    @zakim-robot
    Jan 05 11:46
    [michiel] Nice: <button tabindex="516" disabled="disabled" class="disabled" id="playButton" aria-disabled="true" aria-live="off" istabbable="false"><span class="accHidden" id="playButton\_altSpan">Play</span></button>
  7. zakim-robot
    @zakim-robot
    Jan 05 13:21
    [karlgroves] They really don’t want you to use that button. Or even know it exists.
  8. Mallory
    @StommePoes
    Jan 05 13:51
    @stevefaulkner you happen to have a link to that issue? I need to pass it around, a lot of devs here want to use details/summaries because it's presented as good solid HTML that's just waiting for more support (so they'd rather polyfill it than use separate button+div)
  9. I understand the sentiment, except I can't get around the nested-focusables thing. However, Firefox on Linux seems to know what to do-- it treats the begin of the focusable as "first" and the first nested as "next" etc, forwards and backwards. I'm guessing that's what all the browsers will have to do and AT will have to figure out what they consider names vs content in those.
  10. zakim-robot
    @zakim-robot
    Jan 05 14:03
    [melsumner] Reading some of the new roles and properties and excited about some of these, they seem like they will be incredibly useful!
  11. Mallory
    @StommePoes
    Jan 05 14:04
    role role role yer dirk, gently down the scream... melanie melanie melanie melanie roles aria dream
  12. wow that would be a great song if I were drunk in a bar
  13. zakim-robot
    @zakim-robot
    Jan 05 14:05
    [melsumner] very drunk haha
  14. Mallory
    @StommePoes
    Jan 05 14:06
    yeah I should have added some slur to that
  15. ok, plan: get drunk at CSUN, replace melanie with my own name, and sing it loudly at the google party
  16. if there's a piano I'll wear a Jessica Rabbit dress. not that I look good in those, but they fit pianos.
  17. Job van Achterberg
    @jkva
    Jan 05 14:10
    I'm interested.
  18. zakim-robot
    @zakim-robot
    Jan 05 14:10
    [melsumner] the new stuff is genuinely going to make development easier in a practical way so I think the excitement is okay :)
  19. Mallory
    @StommePoes
    Jan 05 16:18
    I actually never follow the new stuff until people are talking about using it...
  20. so much stuff, so little support :(
  21. and... I'm just no lover of polyfills. They promise that "one day, in the future, you'll remove it and stuff will now Just Work!" and that's kinda never how it ends up
  22. zakim-robot
    @zakim-robot
    Jan 05 16:40
    [car] Yeah. I was really looking forward to <dialog> and <menu type="toolbar"> in w3 html, and they didn't make the cut due to lack of support. <pout>
  23. Mallory
    @StommePoes
    Jan 05 19:36
    I def need a <pout> tag
  24. James Nurthen
    @jnurthen
    Jan 05 19:58
    @stevefaulkner you around?
  25. I have a question about ARIA in HTML document
  26. ARIA in HTML states that I can use aria-required anywhere I can use required in HTML5. The ARIA spec only allows aria-required on certain roles (for example it allows on textbox but not on checkbox whereas required is allowed on any INPUT element.
  27. am I allowed to have <input type="checkbox" aria-required="true">
  28. zakim-robot
    @zakim-robot
    Jan 05 21:45
    [melsumner] I'm not him, but I know you don't need to use aria-required if you are already using required
  29. James Nurthen
    @jnurthen
    Jan 05 21:46
    But we don't want to use required as it has other effects.
  30. powrsurg
    @powrsurg
    Jan 05 21:55
    Can someone explain to me what the updated Section 508 ICT standards means? Is that the 508 refresh (based on WCAG 2.0 AA?) is finally going to happen?
  31. zakim-robot
    @zakim-robot
    Jan 05 21:58
    [cameron] friends, anyone using this color contrast checker for Sketch? Thoughts? https://github.com/getflourish/Sketch-Color-Contrast-Analyser
  32. zakim-robot
    @zakim-robot
    Jan 05 23:33

    [tylersticka] Question!

    I have a menu toggle. The structure of the menu is kinda like this (super simplified): 

    <a href="#menu">Menu</a> <nav id="menu">   Awesome Menu Contents! </nav>

    When JS is disabled, the toggle is still functional by virtue of a :target pseudo-class. But when JS is enabled, the menu is "enhanced" with animation and some other bells and whistles that make it easier to use.

    But this is a responsive thing... at large sizes, the menu control is hidden, and the nav is always visible.

    So my questions are:

    • Is it helpful to add aria-expanded/aria-hidden/etc.? Or will the hash-target relationship likely suffice?
    • If so, should I then be listening to the browser's resize event and managing aria-hidden and aria-expanded accordingly when shifts from the small-screen, toggle experience to the large-screen, always-visible one?

  33. Mallory
    @StommePoes
    Jan 05 23:37
    yup, many devs would rather have aria-required so they don't get the half-assed, poorly localised and often uninformative error messages, but instead can make their own, or not show any, or not make an input look invalid when require and the user hasn't even been given the chance to screw it up-- their wrong before they even start
  34. James Nurthen
    @jnurthen
    Jan 05 23:38
    @StommePoes you are a fellow fan of the required attribute i see.
  35. zakim-robot
    @zakim-robot
    Jan 05 23:52
    [tylersticka] Here's a pen illustrating a simplified version of the scenario, including the JS-less fallback and the responsive visibility. (Thanks @melsumner for suggesting I include an example.) http://codepen.io/tylersticka/pen/2cddc9d886e7d1fa08dbca795be96c21/