Archive index

A11y Slackers Gitter Channel Archive 15th of August 2016

What fresh hell is THIS now? - Patrick Lauke
  1. Job van Achterberg
    @jkva
    Aug 15 08:23
    _>
  2. zakim-robot
    @zakim-robot
    Aug 15 12:51
    [michiel] Morning slackers!
  3. jkva @jkva looks at his watch
  4. Job van Achterberg
    @jkva
    Aug 15 12:54
    I didn't know Oxford was that far off ;)
  5. zakim-robot
    @zakim-robot
    Aug 15 12:54
    [michiel] Afternoon* slackers!
  6. Job van Achterberg
    @jkva
    Aug 15 12:54
    \o/
  7. Afternoon Michiel!
  8. zakim-robot
    @zakim-robot
    Aug 15 12:55
    [michiel] Sometimes I have to work with Billy so I'm on EST :P
  9. zakim-robot
    @zakim-robot
    Aug 15 13:12
    [stanislauzubovich] Morning, do someone have statistics on UK users with disabilities?
  10. [michiel] No one has reliable ones ;)
  11. [michiel] If we count people with glasses it's about 40-50%
  12. zakim-robot
    @zakim-robot
    Aug 15 13:18
    [michiel] The one I always use: whichever way you look at it, it's higher than the percentage of IE users.
  13. [stanislauzubovich] I know there is statistics on US people from 2010s or smth like this
  14. [michiel] I'll stop being unhelpful, but also can't be helpful with this one I'm afraid.
  15. zakim-robot
    @zakim-robot
    Aug 15 14:21
    [scottgemmell] Hello, a client wants the h1 on a page to be 12px (same as body copy and most other text). Any good reasons I can reject this would be useful.
  16. Job van Achterberg
    @jkva
    Aug 15 14:23
    @scottgemell: Because it'd make no visual sense because it does not distinguish the site heading from body copy? Semantically it's fine. Has the client motivated this requirement?
    • @scottgemmell ^
  17. zakim-robot
    @zakim-robot
    Aug 15 14:25
    [scottgemmell] Yes. I explained headings would not be distinguishable.
  18. Job van Achterberg
    @jkva
    Aug 15 14:26
    @scottgemmell: You could argue that it make the content hard to understand and visually unclear, which might make visitors frustrated and cause them to leave. But that's hard to back up without A/B testing.
  19. But as long as they're aware of the risks... as long as the content is readable and understandable, it's not a problem per se
  20. zakim-robot
    @zakim-robot
    Aug 15 14:27
    [scottgemmell] Bother. Thanks for feedback. I guess I will lose this one. :(
  21. OceanWolf
    @OceanWolf
    Aug 15 14:36
    Give the client what they want, you can show them what it looks like, give them an alternative, if they still want it, then fine, their fault for loosing visitors, so long as you get paid
  22. zakim-robot
    @zakim-robot
    Aug 15 16:42
    [simong] marcysutton: Are you still looking on running axe within Node? If you are, I’d be happy to help in any way I can. We’re currently running it in phantom, but it’s just too slow for what we want to do (There’s a 5second penalty on doing phantom.exit(0) -_-).
  23. [marcysutton] We do run it in node, there are some problems specific to JSDOM though
  24. zakim-robot
    @zakim-robot
    Aug 15 16:47
    [simong] oh ok
  25. [marcysutton] I guess a better way of putting it is we handle axe in Node, but it does require a rendered DOM to run checks against
  26. [simong] Are you running an older version? I’d be interested in hearing how you’re doing it currently as we’re looking to process a few million HTML files maybe
  27. [marcysutton] An older version of Node?
  28. [marcysutton] I'm in 4x
  29. [simong] No sorry, I misunderstood, I thought you were successfully using axe + jsdom and that it was working ok on an older version of axe
  30. [marcysutton] The newer version of axe handles window better than it used to, you still need an HTML Document context in some form
  31. [simong] ok, I’ll give HEAD a try
  32. [marcysutton] the latest public release has those fixes (2.0.5)
  33. [marcysutton] you would notice it right away if it choked on window
  34. [simong] Hmm ok. Do you think it’s feasible to get it to work with jsdom?
  35. [marcysutton] I just did a talk on part of this problem
  36. [marcysutton] The JSDOM problem I'm still looking into, it needs a bit more support for some reason
  37. [marcysutton] But I am definitely working on it =D
  38. [simong] Thanks! I’ll poke around a bit as well and report back if I get something working
  39. zakim-robot
    @zakim-robot
    Aug 15 17:19
    [simong] This gets me the violations that don’t rely on document.createRange: http://pastebin.com/rSX8GuR4
  40. zakim-robot
    @zakim-robot
    Aug 15 18:04
    [marcysutton] Yep this is because JSDOM and Cheerio are both partial implementations
  41. [marcysutton] I have filed a bug internally and will investigate our options
  42. [marcysutton] There are tradeoffs for that speed you need :)
  43. [simong] Supposedly, someone is working on the createRange method for jsdom (tmpvar/jsdom#317), so maybe soon ..
  44. [simong] Yeah for sure
  45. [marcysutton] sweet!
  46. [marcysutton] sees date of that issue was filed in 2011 :(
  47. [simong] The last comment was March 18 (2016) though!
  48. [marcysutton] That is an appallingly slow pace though. :snail:
  49. [marcysutton] Websites and apps have been built, launched and retired in that amount of time
  50. zakim-robot
    @zakim-robot
    Aug 15 18:10
    [marcysutton] Although the web standards track is slower for good reason, tooling needs to evolve quicker than that
  51. zakim-robot
    @zakim-robot
    Aug 15 18:19
    [scottohara] Does anyone have/have seen an example of a properly implemented carousel that doesn’t just have abysmal a11y support?
  52. [scottohara] The main one that I have heard before is the Slick slider, but I’ve seen some pretty bad implementations of that one too
  53. [marcysutton] I think @estelle was working on one
  54. [marcysutton] But I have a hard time recommending sliders... /shrug
  55. zakim-robot
    @zakim-robot
    Aug 15 18:42
    [scottohara] heh. i have no desire to recommend one. as i’m currently working on a site where i’m recommending to remove them. but i thought i’d see if there were better options, so if it has to be kept, i can at least compare to a carousel that isn’t a hot garbage fire on wheels
  56. [scottohara] :)
  57. zakim-robot
    @zakim-robot
    Aug 15 20:14
    [aarongustafson] Ok folks, test your mettle: https://a-k-apart.com/
  58. [aarongustafson] @scottohara: Have your tried https://github.com/filamentgroup/responsive-carousel
  59. Job van Achterberg
    @jkva
    Aug 15 20:15
    Wow, cool challenge @aarongustafson
  60. zakim-robot
    @zakim-robot
    Aug 15 20:31
    [scottohara] Oh thanks. I'll give that a look
  61. zakim-robot
    @zakim-robot
    Aug 15 21:27
    [estelle] I have a "merry-go-round" script. It looks like a carousel, and to the sighted user behaves like a carousel, but to the screen reader is an unordered list of options, fully accessible.
  62. [estelle] @marcysutton and @scottohara (forgot to mention you before answering you)
  63. [estelle] The reason it's called merry-go-round is because carousels are expected to work and fail in specific ways. This provides all options up front without requiring aria-live to update the AT that the screen has changed.
  64. [estelle] So it's calling it "carousel" was confusing to those testing it with screen readers
  65. zakim-robot
    @zakim-robot
    Aug 15 22:04
    [jiatyan] So, how do we submit a broken link report on an official WCAG page? eg. https://www.w3.org/TR/UNDERSTANDING-WCAG20/navigation-mechanisms-title.html <-- "Writing Better Web Page Titles" is broken.
  66. zakim-robot
    @zakim-robot
    Aug 15 22:14
    [michiel] jiatyan: GitHub would do :)
  67. zakim-robot
    @zakim-robot
    Aug 15 22:20
    [jiatyan] dank je @michiel
  68. The first one has education levels, employment, heads of households, etc. Digested figures, not the base data. I got the base data of the US census and it's fun to poke through the ginourmous endless scrolling tables (per State) but not fun to find info in.
  69. The second one just has a lot of somewhat random statistics mixed in everywhere.
  70. gov.uk has some stats for them somewhere