Archive index

A11y Slackers Gitter Channel Archive 5th of December 2016

What fresh hell is THIS now? - Patrick Lauke
  1. James Nurthen
    @jnurthen
    Dec 05 02:50
    Not accessibility related but facebook is really annoying me. Every time I go to the iOS app I have to log back in.
  2. I'm guessing it is due to the fact I just got a 2nd isp at home and have load balancing set up and there is some security related setting. It maintains the same IP address for a session but each time I open the iOS app it seems like it switches IP address and facebook barfs.
  3. stevefaulkner
    @stevefaulkner
    Dec 05 09:32
    @Muhnad_a_twitter what are the tests based on?
  4. zakim-robot
    @zakim-robot
    Dec 05 10:45
    [dean] Random hashtag question for you :) - anything I post on Twitter that is related to accessibility in the broad sense of the word I tag with #a11y - I’ve just read a great article about a ‘magic table' that is used for people with dementia, and is being reworked to use with people with autism. Is #a11y as broad as that? or do you guys and gals just use the tag for web / digital accessibility?
  5. Callum Macrae
    @callumacrae
    Dec 05 11:20
    Aargh I can't get aria-live to work in my app. It's just not being read out :(
  6. how do I debug this?
  7. I'm using VoiceOver in Firefox, but the same thing was happening in Safari too
  8. Callum Macrae
    @callumacrae
    Dec 05 11:27
    this isn't working, either, so it looks like it's not me… http://schne324.github.io/live-region-playground/
  9. Job van Achterberg
    @jkva
    Dec 05 12:18
    @callumacrae: that playground example works for me (VO/Safari) - what have you tried?
  10. Mallory
    @StommePoes
    Dec 05 13:17
    Dean: yeah, I totally include dementia and autism and dyslexia and all that with a11y. Because to me it's never going to be limited to someone's narrow-ass definition of "a physical disability"
  11. I'm running Firefox 50.0.2 and my SVG is not being cool.
  12. Someone took my code and made a button type="button" and inside is an SVG with role="img" and aria-labelledby="the id of the title element inside" with indeed a properly marked-up title element
  13. when I inspect using Sarkov's inspector, FF sees the SVG has a role of image, but no name.
  14. And all I hear is push button in Orca. Now on Windows, on both FF and IE with NVDA, things are great. But I'm wondering what's up with Firefox on Linux.
  15. Mallory
    @StommePoes
    Dec 05 13:57
    Hm, and while NVDA on Firefox is reading out the SVG's name, JAWS on Firefox is not.
  16. le sigh
  17. Mallory
    @StommePoes
    Dec 05 14:14
    Even Chrome and IE are giving out names of svg's titles in buttons, but not Firefox. Can't tell if it's JAWS or not since NVDA works in Firefox but how can the browser be offering the information differently to one screen reader than the other...
  18. stevefaulkner
    @stevefaulkner
    Dec 05 14:15
    Even Chrome and IE are giving out names of svg's titles in buttons, but not Firefox. Can't tell if it's JAWS or not since NVDA works in Firefox but how can the browser be offering the information differently to one screen reader than the other...
  19. Its that the AT are using diff methods to access, NVDA uses acc tree, JAWS does its own weirdness
  20. Mallory
    @StommePoes
    Dec 05 14:16
    I would file a bug (or look to see if one were filed) if I knew where the issue was.
  21. Apparently so does Orca, I can't get Orca + FF50 to announce these either.
  22. And when I inspect, I see lots of stuff but no accessible name...
  23. stevefaulkner
    @stevefaulkner
    Dec 05 14:17
    whats the code?
  24. Mallory
    @StommePoes
    Dec 05 14:17
    svg_screenshot_FF_50.png
  25. Muhannad
    @Muhnad_a_twitter
    Dec 05 14:18
    @stevefaulkner based on WCAG 1.0 & WCAG 2.0
  26. Mallory
    @StommePoes
    Dec 05 14:18
    so it's a <button...> inside is <svg ... role="img" aria-labelledby="the title id"><title id="the title id">accessible name here</title><use ... linking to a URL.../></svg></button>
  27. Muhunnad why WCAG1? for the US?
  28. And in that screenshot are two showing the svg and the button in Sarkov's inspector. Everything looks kosher except I do not see name listed anywhere which I did expect.
  29. stevefaulkner
    @stevefaulkner
    Dec 05 14:21
    @Muhnad_a_twitter suggest killing any WCAG 1.0 requirements
  30. Mallory
    @StommePoes
    Dec 05 14:21
    And yeah I did check a non-Angular page too because you never know what kind of crap Angular might give you.
  31. Muhannad I agree with Steve. You could have a special 508 thing for the US, rather than WCAG 1.
  32. Muhannad
    @Muhnad_a_twitter
    Dec 05 14:25
    @StommePoes I can't understand you, can you please explain to me what you mean about 508 thing and why US ?
  33. Mallory
    @StommePoes
    Dec 05 14:27
    Well
  34. You say you are using WCAG1 and WCAG2 for your checker
  35. Muhannad
    @Muhnad_a_twitter
    Dec 05 14:27
    @stevefaulkner I'm not depend on WCAG 1.0 fully , but I think something like alt in image and title in frames recommended from WCAG 1.0 ?
  36. Mallory
    @StommePoes
    Dec 05 14:27
    however, WCAG 1 is out of date. It should not be used.
  37. Muhannad
    @Muhnad_a_twitter
    Dec 05 14:27
    I know is out of date :D
  38. Mallory
    @StommePoes
    Dec 05 14:28
    In the United States though, instead of using WCAg2, they still use something derived from WCAG1 encoded in a section added to an accessibkility law called Section 508
  39. so there are a lot of accessibility checkers who have like a "508 section"
  40. which just check 508-related rules.
  41. stevefaulkner
    @stevefaulkner
    Dec 05 14:28

    however, WCAG 1 is out of date. It should not be used.

    :+1:

  42. Mallory
    @StommePoes
    Dec 05 14:28
    We have them, since my company has US clients.
  43. alt in image is covered in wcag2 so that's covered
  44. I knew about the title for iframes but I can't remember where I got that one from actually
  45. stevefaulkner
    @stevefaulkner
    Dec 05 14:29
    that will be covered on name,role,state
  46. Mallory
    @StommePoes
    Dec 05 14:29
    it's part of our internal guidelines but like I said, as a company with US clients those could be coming from 508 rules.
  47. lemme check, I didn't think it was that specific about iframes
  48. lol looks like people argue about whether or not an iframe is considered an interface
  49. if I can focus to it, I'll consider it an interface and so it needs a name :)
  50. stevefaulkner
    @stevefaulkner
    Dec 05 14:30
    its specific about accessible name
  51. Muhannad
    @Muhnad_a_twitter
    Dec 05 14:34
    @StommePoes Thanks, I understand that :D
  52. now I can say I'm depend on wcag2
  53. Mallory
    @StommePoes
    Dec 05 14:41
    Muhannad: cool!
  54. Mallory
    @StommePoes
    Dec 05 15:30
    Surprised to discover a label can be inside a label.
  55. stevefaulkner
    @stevefaulkner
    Dec 05 16:00

    @StommePoes no it cant
    <label> can contain:

    Phrasing content, but with no descendant labelable elements unless it is the element’s labeled control, and no descendant label elements.

    https://w3c.github.io/html/sec-forms.html#the-label-element

  56. Mallory
    @StommePoes
    Dec 05 16:21
    Well I found an instance of one and neither nu nor w3 caught it. However it was an Angular page with all the Angular crap making errors and I wonder if the validators just gave up part way through.
  57. @stevefaulkner I will go ding them on that then.
  58. stevefaulkner
    @stevefaulkner
    Dec 05 16:23

    Well I found an instance of one and neither nu nor w3 caught it. However it was an Angular page with all the Angular crap making errors and I wonder if the validators just gave up part way through.

    tool fail

  59. stevefaulkner
    @stevefaulkner
    Dec 05 16:47

    I will go ding them on that then.

    its a parsing issue (elements not nesting according to spec)

  60. Mallory
    @StommePoes
    Dec 05 16:49
    yeah but it's also just weird. I can't imagine anyone did it on purpose
  61. so they get dinged
  62. Mallory
    @StommePoes
    Dec 05 17:07
    yup it was a "whoops-- I don't even remember writing that code!" from them :P
  63. zakim-robot
    @zakim-robot
    Dec 05 20:16
    [cameron] hello friends! I’m seeing performance issues with NVDA on VirtualBox for OS X
  64. [cameron] anyone experienced issues with this combo?
  65. [cameron] I’m also going to play with VMware Fusion
  66. [cameron] but would prefer the opensource option with VirtualBox
  67. [cameron] any tips are welcome, thanks!
  68. zakim-robot
    @zakim-robot
    Dec 05 20:24
    [marcysutton] I experienced that but it really depended on the page being used.
  69. [marcysutton] I had to bail completely on a demo of some NYTimes pages because of the lag when navigating content...which I'm guessing was caused by a bunch of heavy, leaky scripts and a large DOM
  70. zakim-robot
    @zakim-robot
    Dec 05 20:32
    [cameron] @marcysutton have you tried vmware or dual-booting with any success?
  71. [marcysutton] I haven't. I liked VMWare Fusion when I worked with it in the past though, a lot more than Parallels
  72. [cameron] interesting, I’ll try it and post notes
  73. [cameron] thanks for your thoughts Marcy!
  74. zakim-robot
    @zakim-robot
    Dec 05 21:06
    [cameron] holy cow, VMware Fusion is so much better with NVDA/FF
  75. [cameron] cc @marcysutton
  76. [cameron] the only caveat is I bumped the default RAM for 1gb to 2gb and set it to use 2 cores
  77. [cameron] 1gb/1 core was a little laggy
  78. [marcysutton] That's what I was about to ask–maybe the default VirtualBox images aren't powerful enough?
  79. [marcysutton] (the ones from modern.ie)
  80. [cameron] yes, I increased the memory/processor similarly for the same image for virtualbox
  81. [cameron] didn’t help much
  82. [cameron] :shrugs:
  83. Mallory
    @StommePoes
    Dec 05 23:07
    https://www.viget.com/articles/the-case-against-progressive-enhancements-flimsy-moral-foundation

  84. "Honda needs to consider a universal audience for its marketing websites even though... (b) only certain people can access or afford the cars being advertised."
    Not to make any statements about Honda or any other particular company and how they choose to stuff their ads in my face, but this is exactly the argument that stops a lot of people I know with disabilities from doing simple things on the web.
    I wouldn't believe it if it didn't keep hearing it but it's an argument I'm pretty sick of hearing from developers over and over and over again: We don't need captions because we decided whether or not deaf people know what music is (and we know they have no idea! so why waste our time?). We don't need an accessible form because we're a vehicle insurance site and blind people don't drive. We don't need to think about accessibility because we're a gaming site and only abled people with 2 arms play games..." etc etc etc.

    This is just Yet Another Article using that argument like it's smart or good or something anyone should be doing.

  85. I'm just ignoring alll the other arguments because this one just had to be included and this one kills me
  86. zakim-robot
    @zakim-robot
    Dec 05 23:11
    [karlgroves] +100
  87. Mallory
    @StommePoes
    Dec 05 23:12
    I would post it but I ain't making a fcking discus account to post a fcking comment. Christ.
  88. zakim-robot
    @zakim-robot
    Dec 05 23:24
    [caesar] I'd reply, but the debate is a little too heavy for me
  89. [caesar] In response to the Honda thing, just because a blind person can't drive a car, doesn't mean that they wouldn't own one
  90. [caesar] That's the message we tend to use, anyway
  91. zakim-robot
    @zakim-robot
    Dec 05 23:46
    [nschonni] Welp, that was an infuriating read