Archive index

A11y Slackers
Gitter Channel Archive 7th of August 2015

What fresh hell is THIS now? - Patrick Lauke
  1. zakim-robot
    Aug 07 00:38
    [tim, a11y] Nice article on the empathy of colour blindness. Also includes some plugins and simulators at the end to your designs - http://www.cooper.com/journal/2015/6/the-color-of-empathy-is-not-flat-insights-to-color-blindness-amp-design-nbsp
  2. zakim-robot
    Aug 07 00:48
    [monastic.panic, a11y] ooof chrome does weird things with live regions. anyone have any expertise?
  3. stevefaulkner
    07:08

    A11ySlackers chat log 6th of August 2015 - http://w3c.github.io/a11ySlackers/archives/060815.html

  4. zakim-robot
    07:17
    [Katy Moe, a11y] @monastic.panic: what kind of behaviour are you seeing?
  5. LjWatson
    12:09
    If anyone has time/inclination, thoughts on this proposal would be welcome: http://bkardell.github.io/common-panel/index.src.html
  6. LjWatson
    12:09
    Prototypes (created using Custom Elements) here: http://bkardell.github.io/common-panel/prototype/panelset-element.html
  7. LjWatson
    12:10
    Please file issues (including comments, suggestions, feedback and questions) on Github: https://github.com/bkardell/common-panel
  8. StommePoes
    12:15
    wow. What a cool idea.
  9. StommePoes
    12:15
    me gusta
  10. LjWatson
    12:21
    @StommePoes Thanks :)
  11. StommePoes
    12:22
    so the proposal is that maybe this could become future HTML, or some standard-ish web component set?
  12. StommePoes
    12:23
    (assuming it developers into Stuff)
  13. StommePoes
    12:23
    s/developers/develops
  14. LjWatson
    12:23
    Hopefully it'll be part of HTML 5.1 (or whatever comes after 5.0).
  15. LjWatson
    12:24
    We have a lot of work to do before then - including using the prototypes to test the idea with developers and users.
  16. LjWatson
    12:24
    Plus talking to browser vendors to see whether they think it's viable.
  17. StommePoes
    12:24
    I like the idea too that it can be naturally responsve. And I've seen many people turn typical carousels into tab-panels (I've done it myself)
  18. StommePoes
    12:25
    The Evil Me thinks that browser vendors may only consider it if it's got enough Sexy in it.
  19. LjWatson @LjWatson grins.
  20. StommePoes
    12:26
    Maybe after this is more fleshed out, a separate group of hipster tech-celebrities can for a working group dedicated to making this and some other needed additions/reforms/fixes Sparkley And Sexy.
  21. StommePoes
    12:26
    I'm only half joking.
  22. LjWatson
    12:29
    We have @stevefaulkner on-board... does it get more hipster and sexy than that? ;)
  23. LjWatson
    12:31
    It's a good point though. Perhaps @brucelawson will lend us a bit of sparkle?
  24. StommePoes
    12:34
    Hm... I'm worried a bit of bogan-sparkle would work against it
  25. StommePoes
    12:34
    I'm thinking more the types who get on stage at Sexy Frontend Conferences (like Fronteers) without mankinis
  26. StommePoes
    12:35
    Dress this proposal up in a mankini and you can consider it dead. Or on W3CMemes. Or both.
  27. StommePoes
    12:35
    mankini-panels...
  28. StommePoes
    12:35
    tabkinis.
  29. LjWatson
    12:36
    I'm still trying to get over the image of anyone on stage *without a mankini...
  30. StommePoes
    12:37
    ...and without penises with warts on slides. So that leaves out @stevefaulkner and the very first part leaves out @jaffathecake since his slides pretty much require at least one cock-and-balls. Which I love, but isn't sexy.
  31. MichielBijl
    12:37
    Maybe someone on a toilet…
  32. StommePoes
    12:38
    Nope, no Patrick Lauke either.
  33. MichielBijl
    12:38
    Common Lumberjacks and Viking Panels?
  34. StommePoes
    12:38
    Sorry @LjWatson it looks like the majority at Paciello group won't be able to add any sexy. They have sordid pasts.
  35. StommePoes
    12:38
    Hm, I can't tell yet if that's sexy or just silly.
  36. StommePoes
    12:38
    I'm thinking more, rock stars.
  37. MichielBijl
    12:39
    Get Brian Cox to do a documentary about it.
  38. StommePoes
    12:39
    Browser rock stars, evangelists with hair, these people have at times helped push some stuff into the vendors
  39. MichielBijl
    12:39
    Mr Redhair himself Heilmann?
  40. StommePoes
    12:39
    Imagine if a collection of Javascript warriors, Paul Irish, Divya, Heilmann, etc were to band together and say common-panels was awesome.
  41. StommePoes
    12:39
    On stage.
  42. MichielBijl
    12:40
    We should make a music video!
  43. LjWatson
    12:40
    It would be easy enough to reach out to any/all of those people.
  44. MichielBijl
    12:40
    It’s the best idea ever!
  45. StommePoes
    12:40
    Oh, and Sebly Delilse (or however that's spelled)
  46. MichielBijl
    12:40
    Get Bruce to write a song about it
  47. MichielBijl
    12:40
    Let The Hoff sing it
  48. StommePoes
    12:40
    It would include mankinis, you know this
  49. MichielBijl
    12:40
    Maybe add some lolcats
  50. StommePoes
    12:41
    No nyans.
  51. LjWatson
    12:41
    For now we're just getting it out there bit by bit. I talked about it at Open Web Camp on Saturday, pinging it here seemed like a good thing to do.
  52. LjWatson
    12:41
    Can I has panels pls?
  53. MichielBijl
    12:41
    :D
  54. StommePoes
    12:41
    That does sound good.
  55. StommePoes
    12:41
    NO no no, the spelling's all wrong for that @ljWatson, it's
  56. StommePoes
    12:41
    "I can haz panelz??"
  57. LjWatson
    12:41
    I'm sorry. I'm British, I can't bring myself to that level of spelling blasphemy.
  58. LjWatson
    12:42
    Could I possibly have a panel please, if it wouldn't be too much trouble...
  59. StommePoes
    12:42
    It's too late to add to Fronteers, but perhaps by next year there's enough filled in that perhaps a talk on this (maybe with some other ideas) would be pretty cool? Get developers salivating over something and they'll be bitching about every vendor who doesn't comply
  60. LjWatson
    12:43
    Good idea.
  61. StommePoes
    12:44
    yeah, you shoulda seen it when @jaffathecake talks about webfonts, every thing he mentioned some Browser X didn't do, people were tweeting about it for days
  62. StommePoes
    12:44
    or Divya or Lea Verou talking about CSS secrets and they'd be like "except this one doesn't work in X", devs drooled
  63. StommePoes
    12:45
    And because this sounds like a good idea (developers should not be reinventing this wheel over and over, like popup/modal dialogs), with a good set of ideal standards this could get some real attention.
  64. LjWatson
    12:46
    THat's what we're hoping - to make things easier for devs and users.
  65. StommePoes
    12:46
    Devs are lazy, both in the Larry Wall sense and ... not. So it's win-win
  66. StommePoes
    12:47
    Every time a dev creates another badly-implemented tabby-thingie, a lolKitten dies.
  67. MichielBijl
    12:48
    That’s why they’re not that hip anymore
  68. rodneyrehm
    12:48
    what’s a tabby thingie?
  69. StommePoes
    12:48
    any thingie with tab-like features
  70. StommePoes
    12:48
    carousels with the little balls
  71. StommePoes
    12:48
    tab panels
  72. StommePoes
    12:48
    accordions
  73. rodneyrehm
    12:48
    ah, got it
  74. StommePoes
    12:48
    thingie where you click a visible thingie from a visible set of thingies to show a hidden thingie
  75. StommePoes
    12:48
    <thingie>, I got dibs on that web component
  76. LjWatson
    12:49
    @rodneyrehm Links to a proposed new set of HTML elements and attributes posted ^^
  77. rodneyrehm
    12:49
    my current keyboard fetish has me linking anything containing “tab” to that particular key ;)
  78. StommePoes
    12:49
    I personally still like tabbing between actual tabs, but that goes against the aria spec. So...
  79. StommePoes
    12:49
    it has to be uniformly implemented with arrows, I'm just still not sure up.down arrows should count
  80. LjWatson
    12:50
    @StommePoes that's definitely worth thinking about. We're currently revising the ARIA design patterns too - for the 1.1 authoring practices guide, so I'll take that back to them too.
  81. rodneyrehm
    12:50
    ok, “some form of content disclosure widget”…
  82. StommePoes
    12:50
    Because a mega-menu or nav menu with submenus could be seen as a tabby-thingie, if you think about it, and I still personally like tabbing between main menu items but arrowing around the submenu links
  83. MichielBijl
    12:50
    @StommePoes as long as I can have <ka-jigger>
  84. StommePoes
    12:51
    @MIchiel since i dunno what a kajigger is, you can haz it. Be aware it rhymes with a swear word tho.
  85. MichielBijl
    12:51
    It’s used by an old lady in Futurama; she referes to everything with it.
  86. MichielBijl
    12:51
    Well, mostly electrical stuff
  87. StommePoes
    12:52
    @LjWatson yeah it was a real estate site where I noticed these long tab-panel panels with home information, I'd be focussed on a tab to show the right panel but the arrowing down to see more of it would switch me to the next one... got old fast.
  88. StommePoes
    12:52
    @Michiel you mean Mom?
  89. LjWatson
    12:53
    @StommePoes using the down arrow ke scrolls the page, is that right?
  90. StommePoes
    12:54
    Yeah, whether I'm "focussed" on just the body or on anything else. Except in FF if you've got caret navigation, which then of course moves the cursor.
  91. LjWatson
    12:55
    Thanks. So if the down arrow is hijacked for tab navigation/selection, then you have to tab into the panel content and then use the down arrow to scroll, right?
  92. StommePoes
    12:55
    The exceptions of things like, you're focussed on something and the down arrow opens something, those aren't too bad really, usually you can back up once and get your page scroll back.
  93. StommePoes
    12:56
    yeah, I'm thinking even if the panel itself is either focussed or has that fake-focus-pointer bit you get when you click an in-page link that goes to a non-focusable element, the down arrow in most browsers (I can't vouch for IE) still lets me move the screen up and down, yeah.
  94. LjWatson @LjWatson nods. Ok. Definitely needs thinking about.
  95. StommePoes
    12:58
    Poll some keyboarders though, I can't claim to be representative.
  96. MichielBijl
    12:58
    @StommePoes no Hattie: http://futurama.wikia.com/wiki/Hattie_McDoogal
  97. LjWatson
    12:59
    @StommePoes Will do - for panels at least it's part of the reason we created the prototypes.
  98. rodneyrehm
    13:00
    “fake-focus-pointer bit you get when you click an in-page link that goes to a non-focusable element” => “sequential focus navigation starting point"
  99. StommePoes
    13:00
    yeah, that thingie
  100. StommePoes
    13:00
    : )
  101. rodneyrehm
    13:00
    (I have to look that up every time I need to name it)
  102. StommePoes
    13:01
    I do remember someone calling it a pointer, like as in a C/memory pointer.
  103. rodneyrehm
    13:02
    I assume it is nothing but a reference, a pointer if you will.
  104. StommePoes
    13:02
    @Michiel oh cat lady
  105. StommePoes
    13:05
    @ljWatson I'm actually curious, Birkir's version has aria-owns on the li, but if the li's role is presentation, wouldn't that orphan any aria-ownses it might have anyway? (instead of if you put it on the anchor)
  106. StommePoes
    13:06
    looks like my account got disconnected for a moment, old stuff:
  107. StommePoes
    13:06
    (03:04:02 PM) Stomme poes: @LjWatson thanks for jumping in on that schema github issue
  108. StommePoes
    13:06
    (03:04:20 PM) Stomme poes: I was actually back in my mail looking at the latest tab-panel debate
  109. MichielBijl
    13:16
    @LjWatson @StommePoes it’s a first draft, but if we can get somebody cool and sexy to sing this: https://gist.github.com/MichielBijl/3c0d82d8ebfdfe5a5464
  110. MichielBijl
    13:16
    Raw version displays better: https://gist.githubusercontent.com/MichielBijl/3c0d82d8ebfdfe5a5464/raw/8c32c0dfc9c590fed3b81adc00d6610850890e13/Native%2520Element.md
  111. MichielBijl @MichielBijl is not used to gist’s
  112. StommePoes @StommePoes cringes
  113. StommePoes
    13:19
    I gotta say, it sounds like something an accessibility person would sing :)
  114. MichielBijl
    13:20
    :D
  115. StommePoes
    13:20
    Eric Egger did an accessibility rap song at one of the Fronteers Jamsessions once
  116. MichielBijl
    13:20
    Than I did my job
  117. StommePoes
    13:20
    "We need common panels To believe in Burning hearts and a brand new feeling Panels!"
  118. StommePoes
    13:20
    oh man
  119. MichielBijl
    13:20
    There is a return between those
  120. StommePoes
    13:20
    maybe if you get that frozen chick to sing it
  121. StommePoes
    13:21
    yeah that return didn't wanna get copy-pasted, blame opera
  122. MichielBijl
    13:21
    I can add punctuation…
  123. MichielBijl
    13:23
    Added punctuation.
  124. MichielBijl
    13:24
    Maybe @Heydon is willing to produce this gem :p
  125. zakim-robot
    13:37
    [monastic.panic, a11y] @katy: hi, sorry if you are still here, chrome doesn't seem to read out most changes in an live, assertive span i have. FF reads it fine
  126. zakim-robot
    13:38
    [monastic.panic, a11y] I tried just changing text, making each msg a node, all all combo's of that with aria-relevant the only time it works sort of consistently is when its a <ul> but that also adds noise
  127. zakim-robot
    13:39
    [Katy Moe, a11y] interesting. I’ve had similar markup that works fine - maybe make a quick demo so we can help debug @monastic.panic ?
  128. zakim-robot
    13:42
    [monastic.panic, a11y] ya let me try and get something together
  129. zakim-robot
    13:55
    [monastic.panic, a11y] @katy: sorry for the React, but thats what I am using, if all the aria looks good we probably can assume its related to how React updates the node value (it uses innerHTML).
  130. zakim-robot
    13:55
    [monastic.panic, a11y] t works greta for annoucing additions
  131. zakim-robot
    13:55
    [monastic.panic, a11y] but not deletions
  132. zakim-robot
    13:55
    [monastic.panic, a11y] http://jsfiddle.net/69z2wepo/13470/
  133. zakim-robot
    13:59
    [monastic.panic, a11y] some more context, i am testing with nvda
  134. LjWatson
    14:06
    DS: Sent an email to the list, this time of the week is difficult for me, are people open to trying another day/time?
  135. LjWatson
    14:06
    Ignore that.
  136. zakim-robot
    14:40
    [Katy Moe, a11y] @monastic.panic: give me a moment and I’ll take a look at that.
  137. zakim-robot
    14:46
    [Katy Moe, a11y] okay, so it works for me on VoiceOver. will have a look on NVDA
  138. zakim-robot
    14:46
    [Katy Moe, a11y] this is interesting as we have something very similar in our React app
  139. zakim-robot
    14:46
    [Katy Moe, a11y] but we haven’t found any problems so far
  140. zakim-robot
    14:50
    [monastic.panic, a11y] yeah it's odd, and only on Chrome
  141. zakim-robot
    14:50
    [monastic.panic, a11y] It's driving me crazy :P
  142. zakim-robot
    14:51
    [Katy Moe, a11y] what version of Chrome?
  143. zakim-robot
    15:06
    [Katy Moe, a11y] I have repro on Windows 10, Chrome 44.0.2403.130, NVDA 2015.2
  144. LjWatson
    15:07
    @KatyMoe What's the problem you're experiencing?
  145. zakim-robot
    15:10
    [Katy Moe, a11y] on the demo page that monastic.panic posted, there is a live region in a span that doesn't get announced on updates
  146. zakim-robot
    15:12
    [Katy Moe, a11y] sorry, JSFiddle is pretty difficult to navigate - here's the demo page with just the field on it https://jsfiddle.net/69z2wepo/13470/embedded/result/
  147. zakim-robot
    15:12
    [Katy Moe, a11y] What should happen is that when you type into the input, 'current value: [inputvalue]' gets announced as it's a live region with role status, aria-live of assertive, and aria-relevant set
  148. LjWatson
    15:15
    Hmm. I'm not getting anything when I follow that URI. Could you repost the original JSBin link?
  149. zakim-robot
    15:15
    [Katy Moe, a11y] http://jsfiddle.net/69z2wepo/13470/
  150. zakim-robot
    15:17
    [monastic.panic, a11y] i am on chrome Version 44.0.2403.130
  151. zakim-robot
    15:18
    [monastic.panic, a11y] lets see what version of NVDA
  152. StommePoes
    15:19
    http://jsfiddle.net/69z2wepo/13470/
  153. zakim-robot
    15:19
    [monastic.panic, a11y] same 2015.2
  154. StommePoes
    15:20
    ah on IRC the two links get smudged together, the embedded one is really https://jsfiddle.net/69z2wepo/13470/embedded/result/
  155. StommePoes
    15:20
    Links prolly work better on the web interface...
  156. LjWatson
    15:22
    @StommePoes I think it's when links are posted in the Slack channel and pied through to the Gitter channel that they get mudged.
  157. LjWatson
    15:22
    FWIW the live region doesn't work in Chrome with Jaws on Windows either.
  158. StommePoes
    15:22
    is this one of those weird setups where, to get the live announcement, a container element is needed with the aria-live stuff and a child element is needed?
  159. zakim-robot
    15:23
    [monastic.panic, a11y] i've tried it with children as well with no luck
  160. LjWatson
    15:23
    @StommePoes Possibly - but whatever it is, it's starting to look like a Chrome thing... the example works in FF with both Jaws and NVDA.
  161. StommePoes
    15:23
    Do people use non-Chromevox on chrome yet these days?
  162. zakim-robot
    15:23
    [monastic.panic, a11y] the only thing that sort of worked was using a list and the value as an <li>
  163. zakim-robot
    15:24
    [monastic.panic, a11y] but that didn't work great on FF
  164. LjWatson
    15:24
    @StommePoes I don't know any blind people that use ChromeVox at all :) Some use NVDA with Chrome.
  165. StommePoes
    15:24
    Same here
  166. StommePoes
    15:25
    Well, no
  167. goetsu
    15:25
    hi
  168. StommePoes
    15:25
    There's a guy on the Orca list who's used ChromeVox+chrome because he got a chromebook for the lawlz
  169. goetsu
    15:25
    I think chromeVox are used if you have a computer with chrome OS
  170. StommePoes
    15:25
    and his work sometimes used something that only really worked in Chrome
  171. LjWatson
    15:25
    :)
  172. StommePoes
    15:25
    @goetsu yeah
  173. LjWatson
    15:26
    @Monastic.Panic Have you tried removing all the ARIA except role="status" and aria-live="assertive"?
  174. StommePoes
    15:26
    I think that guy didn't need to use his Chromebox much after Joanie fixed a lot of live region stuff for Orca and he could use all the Googley things easier
  175. LjWatson
    15:26
    Wondering whether aria-relevant might be interfering.
  176. zakim-robot
    15:28
    [Katy Moe, a11y] I tried removing aria-relevant and it didn’t help
  177. zakim-robot
    15:28
    [monastic.panic, a11y] ya i started without it
  178. zakim-robot
    15:28
    [monastic.panic, a11y] tried it with just aria-live as well
  179. zakim-robot
    15:30
    [Katy Moe, a11y] This is an interesting problem because I’m not quite sure whether it’s a bug in React, Chrome, or the screen readers.
  180. zakim-robot
    15:31
    [monastic.panic, a11y] i don't think its react I also tried to just set innerHTML manually in componentDidUpdate and that didn't yield better results
  181. goetsu
    15:31
    http://fr.slideshare.net/maxdesign/arialive-the-good-the-bad-and-the-ugly
  182. StommePoes
    15:32
    yeah that's mostly pointing how how terrible relevant, removals and additions still is
  183. StommePoes
    15:32
    So a guy I follow is testing out React:
  184. StommePoes
    15:32
    "Learning React. Not sure if it's just me, but the apps I'm building seem to have a higher baseline accessibility level than elsewhere."
  185. StommePoes
    15:32
    "Like, when I delete an item in the sample todo list I'm building, I land on the text of the next list item rather than who the hell knows."
  186. StommePoes
    15:32
    "Think this is because it doesn't just delete from the DOM, it modifies it in place, so focus isn't lost. Interesting implications."
  187. StommePoes
    15:32
    I dunno if that last one is true
  188. StommePoes
    15:32
    or if built-in, focus gets moved.
  189. zakim-robot
    15:33
    [Katy Moe, a11y] I think it depends whether you’re juggling IDs or not (tried to start a discussion in @react about this…)
  190. zakim-robot
    15:33
    [monastic.panic, a11y] yeah, i think some of that is accurate, there are also intersting things that can be done for a11y since there is an intermeddiate layer
  191. zakim-robot
    15:34
    [monastic.panic, a11y] like this experiment: https://github.com/rackt/react-a11y
  192. StommePoes
    15:36
    Hey I sent him that link, he said "Sweet, haven't even looked for a11y-specific stuff, just wrapping my head around it thus far."
  193. goetsu
    15:45

    quick #survey how do you determine what is large text for testing SC 1.4.3 contrast ratio ?

  194. StommePoes
    15:46
    I thought it said somewhere a minimum of 18 pt or 14 pt bold
  195. StommePoes
    15:46
    then again, we don't use pt for various reasons...
  196. goetsu
    15:47
    yes but in reality it's even more complicated than that
  197. StommePoes
    15:47
    apparently it can differ based on ppi of the device, ug
  198. garcialo
    15:47
    @goetsu I use FireEyes; it makes the determination for me. I had asked a few years ago how it got calculated, but I've since forgotten. Perhaps @dylanb can provide some insight.
  199. goetsu
    15:47
    because 18/14 is the value for a default text of 12pt
  200. StommePoes
    15:48
    What I do when I test is generally count "body text" as small, and large headings as "large". Stuff in between, I've been making judgement calls.
  201. goetsu
    15:48
    fireeyes like every other tools are wrong on this one
  202. StommePoes
    15:48
    @goetsu yeah and one never knows any user's default text size
  203. goetsu
    15:48
    they do not take consideration of the size of the defaut text
  204. garcialo
    15:48
    Ah, I see.
  205. goetsu
    15:48
    it's not user default it's based on author default
  206. StommePoes
    15:48
    My default text isn't that 16 px we keep hearing, because I've set all my desktop application base sizes larger
  207. StommePoes
    15:49
    But as an author you can set it to the user's: body {font-size: 100%;}
  208. StommePoes
    15:49
    should be... whatever the user's default settings are.
  209. StommePoes
    15:49
    and my browsers, they grab that from my application settings, dunno what all computers do, but what happens in Gnome on Linux anyway.
  210. StommePoes
    15:50
    I think my base size is probably closer to ~20px rather than 16 if I open firefox to a page that doesn't state font sizes anywhere.
  211. StommePoes
    15:51
    So what is the author's default if the author says "100%" and uses em's or % from there?
  212. goetsu
    15:51
    yes in that case if author define body 100%, large text is > 1.2em bold or 1.5em
  213. goetsu
    15:55
    What I do is, I take the font size of the paragraphe in an article for example
  214. goetsu
    15:55
    I considere it as the default font size
  215. goetsu
    15:55
    I multiply it by 1.2 / 1.5
  216. goetsu
    15:56
    it give me the value for large text
  217. StommePoes
    15:56
    and if the developer was evil
  218. StommePoes
    15:56
    and did what my clients do
  219. StommePoes
    15:57
    with body {font: 11px arial,sans-serif;}
  220. StommePoes
    15:57
    do you increase that?
  221. StommePoes
    15:57
    or do you assume, all users have the power to turn that 12px into whatever default size they need for body text because all browsers can enlarge fonts?
  222. goetsu
    15:57
    nope in that case I considere large text as 11px X 1.2
  223. StommePoes
    15:57
    whether ctrl++ing or something else
  224. goetsu
    15:57
    and 11px X 1.5
  225. goetsu
    15:58
    wcag say Note 2: Font size is the size when the content is delivered. It does not include resizing that may be done by a user.
  226. StommePoes
    15:59
    I can see this letting quite smallish text passing at 3.1 minimum because someone liked 9px base size, blerg
  227. StommePoes
    15:59
    but then I suppose that's what report comments are for
  228. StommePoes
    16:01
    also, thin fonts suck
  229. goetsu
    16:02
    yes but anyway it's better than using this 14 / 18 pt that mean nothing in web environnement even more with responsive website where text size can also change depending of the size of the viewport
  230. StommePoes
    16:03
    for every instance my marketer has decided he wants roboto-thin at 1em, I dock him extra points than I would if it was a readable font
  231. goetsu
    16:20
    Note 1: Fonts with extraordinarily thin strokes or unusual features and characteristics that reduce the familiarity of their letter forms are harder to read, especially at lower contrast levels.
  232. StommePoes
    16:21
    damn right
  233. goetsu
    16:21
    you can use this to justify an higher ratio I presume
  234. StommePoes
    16:21
    I cannot read our own company web site. Ug
  235. StommePoes
    16:21
    I do whether I can or not :)
  236. StommePoes
    16:22
    So far, I haven' done any web testing where someone had to meet some very specific government regulation. But if/when that happens, I will have to be more strict.
  237. StommePoes
    16:23
    But for now, roboto-thin @ ~12px with a ratio of 4.5:1 is still a smack with a wet newspaper roll
  238. rodneyrehm
    16:42
    https://twitter.com/rodneyrehm/status/629694003383009280
  239. rodneyrehm
    16:42
    I’d love your opinion…
  240. zakim-robot
    16:51
    [Karl Groves, a11y] Rodney, “inactive” and “disabled” state have important distinctions.
  241. zakim-robot
    16:52
    [Karl Groves, a11y] “Inactive” means that the control could be interacted with but is not currently being acted on
    “Disabled” means that the control cannot be interacted with (unless some other event causes the control to become enabled)
  242. zakim-robot
    16:53
    [neilmilliken, a11y] Not strictly a11y but hopeful of interest https://www.eventbrite.co.uk/e/axschat-with-kurt-yaeger-tickets-18071117198
  243. rodneyrehm
    16:55
    Karl, is that defined somewhere? I couldn’t find anything. I just put inactive in there to have a third option, to be honest :)
  244. zakim-robot
    16:59

    [Karl Groves, a11y] Rodney, this sort of stuff used to be well defined in the Microsoft User Interface Styleguide (or somesuch name). Even if you’re not an MS fan its worth mentioning that OSX/ KDE/ Android/ iOS styleguides are largely similar in theme. Microsoft’s guide was hands-down the best in terms of explaining things.

    Unfortunately I can’t find the URL right now. I might have a PDF somewhere in an external drive.

    Other than that, the ARIA spec explains some of these same principles WRT disabled

    http://www.w3.org/TR/wai-aria/states_and_properties#aria-disabled

  245. rodneyrehm
    17:01
    thx!
  246. zakim-robot
    17:07
    [Karl Groves, a11y] Rodney: http://www.microsoft.com/en-us/download/details.aspx?id=2695
  247. MichielBijl
    17:08
    Good to see Gitter folk work on a11y: https://github.com/gitterHQ/gitter/issues/771#issuecomment-128749278
  248. MichielBijl
    17:08
    With thanks to @stevefaulkner
  249. rodneyrehm
    17:12
    Karl, searching that doc for “inactive” does not say anything regarding “currently not interacted with"
  250. zakim-robot
    17:13
    [Karl Groves, a11y] Theere’s no list of states?
  251. rodneyrehm
    17:15
    if there is, “inactive” is not one of them
  252. rodneyrehm
    17:16
    at least preview app did not show anything like that when searching for "inactive"
  253. rodneyrehm
    17:17
    it looks like it’s going to be “disabled” anyway
  254. StommePoes
    17:36
    still going through this myself, but Detlev would like any comments/questions/concerns/input/hate mail about this draft http://www.incobs.de/articles/items/diagram-a11y.html
  255. StommePoes
    17:36
    ok not the hatemail
  256. StommePoes
    17:37
    can tweet to @wcagtest if twotting
  257. zakim-robot
    17:37
    [Katy Moe, a11y] @tim ^ the above link may be relevant for your a11y graphic design work
  258. rodneyrehm
    17:39
    http://www.urbandictionary.com/define.php?term=Twotting - “twotting”… :D
  259. StommePoes
    17:39
    tweet, twot, twit
  260. StommePoes
    17:40
    Dutch actually incorporated it as a verb where past tense becomes "getwittert"
  261. StommePoes
    17:40
    possibly with a d instead of a t on the end, seeing's how everyone mixes those two up anyway
  262. rodneyrehm
    17:40
    yeah, same in german ;)
  263. StommePoes
    17:40
    geupload, gegoogled, gemaild
  264. StommePoes
    17:41
    lol
  265. StommePoes
    17:41
    but I like to do the indo-european vowel-shift thing with tweet
  266. rodneyrehm
    17:41
    you’re from the NL, I presume?
  267. StommePoes
    17:41
    so sometimes I say I twotted, and even more incorrectly Ik heb getwot
  268. StommePoes
    17:42
    I'm living there
  269. StommePoes
    17:42
    not originally-originally, but Dutch now
  270. rodneyrehm
    17:42
    originally from? are you attending fronteers in oct?
  271. StommePoes
    17:42
    yes! You?
  272. rodneyrehm
    17:42
    of course I am ;)
  273. StommePoes
    17:42
    (I couldn't get a ticket, but I used to be a member and luckily someone from the group offered me one from those who didn't bother to pay on time :D )
  274. StommePoes
    17:43
    wanbetalers ++
  275. StommePoes
    17:43
    last year was the first one I'd missed since almost the beginning, in order to save money for my first-ever CSUN
  276. StommePoes
    17:43
    which I thought would be my last
  277. StommePoes
    17:43
    but now I'm switching jobs to a company who does the CSUN thing by default, which is awesome
  278. StommePoes
    17:44
    I'll see you in October then
  279. StommePoes
    17:44
    I used to have cool awesome strange hair colours (white in the front, orange in the back) but now that I'm growing hair to donate to cancer I can't colour it anymore :(
  280. StommePoes
    17:44
    so... I'll probably have cartoons on my shoes and pigtails.
  281. rodneyrehm
    17:45
    :D
  282. StommePoes
    17:45
    too bad @MichielBijl won't be there
  283. zakim-robot
    18:28
    [Andres Lopez, a11y] I'm looking into good choices on organizations for accessibility certification. Any recommendations?
  284. MichielBijl
    18:40
    @Andres Lopez: Siteimprove has some certificates, but those don't seem to be all that official
  285. MichielBijl
    18:47
    @StommePoes I'll see if I can make it to the jam sessions
  286. MichielBijl
    18:48
    Can't promise anything for next year, as I hope to have moved to Denmark by then
  287. zakim-robot
    19:00
    [Justin Stockton, a11y] @drelopez: I haven’t looked in a few years, but I don’t there is an officially recognized one yet. That’s why it’s one of the goals of the IAAP
  288. zakim-robot
    19:02
    [Justin Stockton, a11y] though if you’re a U.S. Federal employee or contractor, there is DHS’ Trusted Tester certification
  289. rodneyrehm
    19:02
    sometimes I wonder why I have “manager” in my title. I’ve been working on ally.js for nearly a year now. I’ve already had to move from a11y.js to ally.js because I didn’t grab the name on npm right away. Now I find allyjs.com has been taken… wtf is wrong with me?
  290. zakim-robot
    19:06
    [Andres Lopez, a11y] Thanks for the tips Poorgeek and michiebijl
  291. zakim-robot
    19:06
    [Andres Lopez, a11y] I'll look inot them
  292. zakim-robot
    19:07
    [Andres Lopez, a11y] into*
  293. zakim-robot
    20:51
    [Marcy Sutton, a11y] I wish this content still existed..was so useful for explaining text alternatives beyond <img> http://www.w3.org/TR/html-alt-techniques/
  294. zakim-robot
    20:53
    [Justin Stockton, a11y] it still exists. just go to the Previous Version http://www.w3.org/TR/2014/WD-html-alt-techniques-20141023/
  295. zakim-robot
    20:55
    [Marcy Sutton, a11y] yay thanks
  296. zakim-robot
    20:56
    [Marcy Sutton, a11y] Just waiting to get a “don’t use that!” admonition :simple_smile:
  297. zakim-robot
    20:57
    [Justin Stockton, a11y] it appears that they’re just not maintaining as a seperate spec. it’s been rolled into full spec for describing the alt attribute on <img> http://www.w3.org/TR/html51/semantics.html#alt
  298. zakim-robot
    21:06
    [Marcy Sutton, a11y] For some reason I thought the original document included other kinds of text alternatives. I see <img> and <figure>. Would be most helpful to have visually-hidden spans and aria-label listed too. Sounds like a blog post is in order :simple_smile:
  299. stevefaulkner
    21:09
    @marcysutton the alt tech spec has only ever been about using alt (and friends) and Justin is correct it got folded into HTML5 before it went to Rec. I maintain and develop the alt techniques within the html spec now
  300. stevefaulkner
    21:12
    @marcysutton the one Justin linked to from October last year is still ok
  301. zakim-robot
    21:28
    [Marcy Sutton, a11y] Ok, good to know! Thanks Steve :simple_smile: