Archive index

A11y Slackers Gitter Channel Archive 21st of July 2016

What fresh hell is THIS now? - Patrick Lauke
  1. Job van Achterberg
    @jkva
    Jul 21 06:09
    Morning slackers
  2. Mallory
    @StommePoes
    Jul 21 07:22
    test :)
  3. aha
  4. Job van Achterberg
    @jkva
    Jul 21 07:48
    isn't this redundant?
  5. <header> <div id="top" role="banner" class="header"> [...] </div> </header>
  6. Mallory
    @StommePoes
    Jul 21 07:57
    yeah, if that <header> was a child of the body
  7. Job van Achterberg
    @jkva
    Jul 21 08:13
    Wouldn't it always be? You mean direct child?
  8. Mallory
    @StommePoes
    Jul 21 09:16
    No, you can have <header>s inside <article>s for example
  9. these would not have banner roles
  10. And not direct child since, so far as I know, you could have a div in between
  11. <body><div><header (has role banner)...</header></div>...</body>
  12. But yeah i shouldn't have said child of the body, I should have said "scoped to the body"
  13. so in an article, wouldn't be so scoped, but well a descendant of the body still.
  14. zakim-robot
    @zakim-robot
    Jul 21 09:18
    [michiel] jkva: you can't nest headers.
  15. [michiel] Or was that changed?
  16. Mallory
    @StommePoes
    Jul 21 09:19
    nest?
  17. zakim-robot
    @zakim-robot
    Jul 21 09:19
    [michiel] Was a whole discussion a while back.
  18. [michiel] Kitty nest
  19. Mallory
    @StommePoes
    Jul 21 09:19
    Oh have something with a banner role inside a header?
  20. zakim-robot
    @zakim-robot
    Jul 21 09:19
    [michiel] Yeah
  21. Mallory
    @StommePoes
    Jul 21 09:19
    wouldn't that be nesting banner roles
  22. since that div won't count as a header
  23. zakim-robot
    @zakim-robot
    Jul 21 09:20
    [michiel] Here
  24. [michiel] Content model:
  25. [michiel] Flow content, but with no main element descendants, or header, footer elements that are not descendants of sectioning content which is a descendant of the header.
  26. [michiel] http://w3c.github.io/html/sections.html#the-header-element
  27. [michiel] So, this is invalid: <header><div role=banner>
  28. [michiel] This is valid: <header><div role=group><header>
  29. [michiel] So, jkva's example is invalid :)
  30. [michiel] Besides being redundant :P
  31. Mallory
    @StommePoes
    Jul 21 09:22
    not descendants of sectioning content in the header... meaning, we could <header>... <article><div role="banner"/></article></header> ? Cause that also looks silleh
  32. zakim-robot
    @zakim-robot
    Jul 21 09:22
    [michiel] Yeap
  33. [michiel] Was a request.
  34. Mallory
    @StommePoes
    Jul 21 09:22
    You don't have to always listen to the Dark Side just because they have cookies.
  35. zakim-robot
    @zakim-robot
    Jul 21 09:22
    [michiel] Let me find the issue#
  36. Job van Achterberg
    @jkva
    Jul 21 09:22
    Found it in the source of http://pa11y.org/ :-/
  37. Same thing in the footer
  38. zakim-robot
    @zakim-robot
    Jul 21 09:23
    [michiel] w3c/html#273
  39. Job van Achterberg
    @jkva
    Jul 21 09:23
    https://www.irccloud.com/pastebin/iHnQrZFJ/
  40. :-/
  41. Mallory
    @StommePoes
    Jul 21 09:23
    you shouldn't have something as important as a logo have empty alts either. <img src="images/logo.png" alt="" class="pull-left logo" />
  42. <h1>Pa11y
    <p class="h3">your automated accessibility testing pal</p>
    </h1>
  43. I thought it remained illegal to stuff blocks into headings
  44. did that change?
  45. zakim-robot
    @zakim-robot
    Jul 21 09:24
    [michiel] That doesn't look right.
  46. Job van Achterberg
    @jkva
    Jul 21 09:24
    Jeez
  47. zakim-robot
    @zakim-robot
    Jul 21 09:24
    [michiel] Element p not allowed as child of element h1 in this context. (Suppressing further errors from this subtree.)
  48. [rowanmanning] Agh yeah that site’s super old and pretty bad :(
  49. [rowanmanning] (hello :wave: I maintain Pa11y)
  50. Mallory
    @StommePoes
    Jul 21 09:25
    2 h1's
  51. zakim-robot
    @zakim-robot
    Jul 21 09:25
    [michiel] Hello rowanmanning
  52. Mallory
    @StommePoes
    Jul 21 09:25
    You could probably clean all that markup up and still have it stay looking so clean.
  53. zakim-robot
    @zakim-robot
    Jul 21 09:25
    [rowanmanning] We’re working on a new site at the moment, because we’re all embarrassingly aware of how terrible the existing one is.
  54. Job van Achterberg
    @jkva
    Jul 21 09:26
    ++
  55. zakim-robot
    @zakim-robot
    Jul 21 09:26
    [rowanmanning] In the meantime I might get someone to fix that particular issue =D
  56. [michiel] Feelings of embarrassment are a good motivator :P
  57. [rowanmanning] thumbsup emoji
  58. Mallory
    @StommePoes
    Jul 21 09:26
    meh, have someone just run one of the validators through it and fix those, and do the nitpickier stuff later
  59. zakim-robot
    @zakim-robot
    Jul 21 09:26
    [michiel] Intrinsic embarrassment that is.
  60. Mallory
    @StommePoes
    Jul 21 09:27
    Maintains pa11y, but not necessarily maintains the website of it.
  61. zakim-robot
    @zakim-robot
    Jul 21 09:27
    [michiel] Good to hear it's being worked on.
  62. Job van Achterberg
    @jkva
    Jul 21 09:27
    Yeah
  63. zakim-robot
    @zakim-robot
    Jul 21 09:28
    [michiel] StommePoes: I rather meant that I'd go out of my way to fix something if I was really embarrassed about it.
  64. [rowanmanning] The new site’s here if you’re interested http://pa11y.github.io/ – just not styled yet, the content is pretty much all there
  65. Job van Achterberg
    @jkva
    Jul 21 09:28
    And I learned more stuff about HTML syntax
  66. zakim-robot
    @zakim-robot
    Jul 21 09:28
    [michiel] ++
  67. [michiel] Just read about Charles Kinsey…
  68. [michiel] WTF
  69. Mallory
    @StommePoes
    Jul 21 09:29
    An example of a nitpick that should be taken care of, but isn't an error and could wait for a rewtite: <span class="glyphicon glyphicon-heart"></span> I'd ensure no crazy unicode code point names get read out by throwing an aria-hidden on those. It's a good thing to do, but not doing it right now doesn't break the site.
  70. Yeah I saw the video on the twitters, what the freaking foo
  71. Job van Achterberg
    @jkva
    Jul 21 09:29
    h1 followed by h3s?
  72. zakim-robot
    @zakim-robot
    Jul 21 09:29
    [michiel] Bad
  73. Job van Achterberg
    @jkva
    Jul 21 09:29
    Followed by h2
  74. zakim-robot
    @zakim-robot
    Jul 21 09:30
    [michiel] Not illegal, but bad.
  75. Mallory
    @StommePoes
    Jul 21 09:30
    You could if you had some really good reason for it, but shouldn't
  76. Job van Achterberg
    @jkva
    Jul 21 09:30
    Yeah, these are in the new site
  77. Mallory
    @StommePoes
    Jul 21 09:30
    Dunno how any cop could make the claim (if he were to, I didn't hear he did) that he had to shoot because he felt threatened... by an old fart lying on his back with his hands in the air, talking calmly and rationally.
  78. zakim-robot
    @zakim-robot
    Jul 21 09:32
    [michiel] Yeah if you're threatened by that, you shouldn't be a cop.
  79. Mallory
    @StommePoes
    Jul 21 09:34
    "we got a report someone with a gun was talking about suicide... so when we arrived, we tried to prevent the suicide by just shooting someone ourselves" lawlz
  80. Anyway, I'm done with news for today, if another coup or terror attack happens, tell me tomorrow
  81. Mallory
    @StommePoes
    Jul 21 09:45
    If we're using a button+div as a disclosure widget, since the thingie we're showing is right after (source-wise) the button, does the button really need an aria-owns pointing to that div?
  82. Or is that when the two are not next to each other in source?
  83. Job van Achterberg
    @jkva
    Jul 21 09:48
    [aria-owns] "Identifies an element (or elements) in order to define a visual, functional, or contextual parent/child relationship between DOM elements where the DOM hierarchy cannot be used to represent the relationship"
  84. @StommePoes ^
  85. Sounds to me like hierarchy does indicate that in your example
  86. aria-controls is for instance a hide/show widget where you define what controls which other thing
  87. Mallory
    @StommePoes
    Jul 21 09:50
    well but the div-that-appears is not a child, only an adjascent sibling
  88. Job van Achterberg
    @jkva
    Jul 21 09:50
    aria-owns is to indicate a relationship where this is not clear from the DOM
  89. If it's not a child, i'd just use aria-controls
  90. since it's about disclosure
  91. I feel uncomfortable advising you on this
  92. Mallory
    @StommePoes
    Jul 21 09:50
    Well, I just saw an example of a "button" (anchor) that opens a menu, it has aria-owns
  93. this has been making the rounds https://teachaccess.github.io/tutorial/#/10
  94. Léonie Watson
    @LJWatson
    Jul 21 09:51
    Usually if a control opens something, aria-controls is the right attrib to use.
  95. Mallory
    @StommePoes
    Jul 21 09:51
    and I wondered if we really needed aria-owns in this case
  96. Job van Achterberg
    @jkva
    Jul 21 09:51
    ...it has inline event handlers
  97. Mallory
    @StommePoes
    Jul 21 09:51
    see, no wonder our developers just throw their hands up and say "I'll make it accessible when there's 1 clear way to do it" and walk off
  98. inline event handlers for simplicity, i assume.
  99. Job van Achterberg
    @jkva
    Jul 21 09:51
    I would strongly use aria-controls for that
  100. I would not use aria-owns
  101. Mallory
    @StommePoes
    Jul 21 09:52
    So should I try to hunt these people down and get them to change it?
  102. It's been making the rounds at Pearson
  103. Job van Achterberg
    @jkva
    Jul 21 09:52
    aria-owns does not necessarily imply that it influences control
  104. Léonie Watson
    @LJWatson
    Jul 21 09:52
    aria-owns creates a DOM relationship so it wouldn't be incorrect to use it. aria-controls is intended to signify this kind of relationship though - where activating one thing changes another.
  105. Mallory
    @StommePoes
    Jul 21 09:52
    people will follow it
  106. Léonie Watson
    @LJWatson
    Jul 21 09:52
    jkva+1
  107. Mallory
    @StommePoes
    Jul 21 09:53
    So does anyone know who these teachaccess people are?
  108. Job van Achterberg
    @jkva
    Jul 21 09:53
    @StommePoes agreed that it could be explained better
  109. I think I saw them on twitter
  110. lemme check
  111. Mallory
    @StommePoes
    Jul 21 09:54
    I think I found their site
  112. Job van Achterberg
    @jkva
    Jul 21 09:54
    https://twitter.com/teachaccess
  113. Léonie Watson
    @LJWatson
    Jul 21 09:54
    @Stommpoes never heard of them
  114. Job van Achterberg
    @jkva
    Jul 21 09:54
    They were involved in #AXSchat this week, that's how I recognised the name
  115. Léonie Watson
    @LJWatson
    Jul 21 09:54
    You have a URL for them?
  116. Job van Achterberg
    @jkva
    Jul 21 09:55
    http://teachaccess.org/ I believe
  117. Mallory
    @StommePoes
    Jul 21 09:55
    http://teachaccess.org/
  118. Léonie Watson
    @LJWatson
    Jul 21 09:55
    Thanks both :)
  119. Job van Achterberg
    @jkva
    Jul 21 09:55
    Yep, same people
  120. Léonie Watson
    @LJWatson
    Jul 21 09:55
    Oh. I'm really not paying attention.
  121. Mallory
    @StommePoes
    Jul 21 09:55
    Yeah I got the link from our Accessibility User Tester at work, Chris Langston. He was sending it to all the teams
  122. Léonie Watson
    @LJWatson
    Jul 21 09:56
    I do know them. It's a collaboration between a bunch of a11y orgs (including TPG).
  123. Job van Achterberg
    @jkva
    Jul 21 09:56
    Aha
  124. Léonie Watson
    @LJWatson
    Jul 21 09:56
    If there are issues with the content, file issues and/or let me know - I'll find the right people and let them know.
  125. Mallory
    @StommePoes
    Jul 21 09:56
    I can't ever get AXSchat to "run" or "play" in my browsers so I never listen
  126. I found a contact form on the website and had just sent a short message
  127. just saying "the experts in the room suggested aria-controls over aria-owns " for the menu example
  128. Léonie Watson
    @LJWatson
    Jul 21 09:57
    https://www.paciellogroup.com/blog/2016/07/announcing-the-teach-access-tutorial/
  129. Job van Achterberg
    @jkva
    Jul 21 09:57
    Small world!
  130. Léonie Watson
    @LJWatson
    Jul 21 09:57
    @Stommepoes - yes, agree about that. I'll see if I can find smeone who can do something about that.
  131. Job van Achterberg
    @jkva
    Jul 21 09:58
    Thanks @LJWatson
  132. Léonie Watson
    @LJWatson
    Jul 21 09:58
    Np.
  133. Job van Achterberg
    @jkva
    Jul 21 10:01
    @StommePoes I never watch #AXSChat since it's always during fire training, but the Twitter feed backlog is interesting
  134. Mallory
    @StommePoes
    Jul 21 10:04
    Ug our crappy datepicker is not only utter hell and a complete steaming pile of crap in 3 screen readers but IE11 decides to ignore some code for closing the thing once the date is picked. Without seeing the code I have no idea, but since I've seen most of the rest of the code, I can probably assume it's the fact that the code is a steaming pile
  135. And we still have aria-hidden crap being read out in FF, I assume because gridcell roles are "interactives" :(
  136. Job van Achterberg
    @jkva
    Jul 21 10:05
    That sounds like a bug
  137. Mallory
    @StommePoes
    Jul 21 10:05
    It's bugs all the way down
  138. And with the 1 crappy datepicker, it's not tied to an input... so if you can't get the calendar to work, you can't pick ANY dates
  139. Job van Achterberg
    @jkva
    Jul 21 10:05
    At least that's consistent
  140. Léonie Watson
    @LJWatson
    Jul 21 10:06
    There is a ug in FF where content hidden with aria-hidden is recognised when it's inside a <button>.
  141. Mallory
    @StommePoes
    Jul 21 10:06
    NVDA just randomly starts reading out date numbers when you switch months... because this is React, the live regions don't work because React rebuilds All The Things so a live region doesn't notice any change because, hey, the live region is new too
  142. Léonie Watson
    @LJWatson
    Jul 21 10:06
    *bug
  143. Mallory
    @StommePoes
    Jul 21 10:06
    Yeah I saw that one Leonie but we're not using buttons, but they are interactives
  144. and thought "maybe the same issue"
  145. since in that bug report someone claims to have fixed specifically your use case
  146. tho I dunno if it's actually fixed
  147. Léonie Watson
    @LJWatson
    Jul 21 10:06
    Hmm. Possible yes. When I get a mo I'll experiment to see if it extends beyond buttons.
  148. Mallory
    @StommePoes
    Jul 21 10:07
    and I never run bleeding edge FF
  149. Léonie Watson
    @LJWatson
    Jul 21 10:07
    Haven't checked. It def isn't fixed in 47.
  150. Mallory
    @StommePoes
    Jul 21 10:08
    I wanted to have an example of our grid/gridcells calendar so others could check it but it's such a steaming pile, I could never get a clean and non-proprietary version out there :( and the basis we're using doesn't use any of the roles correctly so the one on github is a bit worthless to test against
  151. someone put a role="widget" on the div around the whole thing, I wasted a day looking for this ghost role
  152. Léonie Watson
    @LJWatson
    Jul 21 10:08
    (facepalm)
  153. Mallory
    @StommePoes
    Jul 21 10:09
    Probably by the time I figure out React enough to rewrite this calendar to not suck, FF will have fixed the bug anyway.
  154. Léonie Watson
    @LJWatson
    Jul 21 10:10
    I was mistaken initially - CSS generated is a red herring.
  155. Mallory
    @StommePoes
    Jul 21 10:10
    we happen to also be using css generated
  156. Léonie Watson
    @LJWatson
    Jul 21 10:10
    But would be good to get the bug fixed.
  157. Mallory
    @StommePoes
    Jul 21 10:10
    but we did accidentally put the aria-hidden on the sr-only text while testing and yeah, it also read out :)
  158. Léonie Watson
    @LJWatson
    Jul 21 10:10
    Will go take a look in a while. Need more tea first.
  159. Mallory
    @StommePoes
    Jul 21 10:11
    Well so until then, kids get to hear about the wonderful circle-bullet on their days with assignments due.
  160. Lawlz.
  161. Léonie Watson
    @LJWatson
    Jul 21 10:11
    Ha.
  162. Mallory
    @StommePoes
    Jul 21 10:11
    And we added to the list, adding a div outside the calendar for a live region so we can state the new month name when one clicks next/previous buttons
  163. becuase React rebuilding all the things means the usual version doesn't work. So instead you click "next month" and start hearing random (perhaps how React renders the rows?) dates read out. Annoying.
  164. If I were a user of this, I'd just skip the damn calendar and go for an assignment list.
  165. Job van Achterberg
    @jkva
    Jul 21 10:13
    KISS
  166. Mallory
    @StommePoes
    Jul 21 10:13
    I wish
  167. but we have designers
  168. Léonie Watson
    @LJWatson
    Jul 21 10:13
    +1 I always prefer an edit field instead of a date picker.
  169. Mallory
    @StommePoes
    Jul 21 10:13
    They've managed to come up with the most amazing Rube Goldberg devices for the web
  170. At least our "normal" date pickers are tied to an input
  171. Léonie Watson
    @LJWatson
    Jul 21 10:14
    Rube Goldberg?
  172. Mallory
    @StommePoes
    Jul 21 10:14
    this thing is a special weird thing they've added for just this one place, to see assignments due
  173. Léonie Watson
    @LJWatson
    Jul 21 10:14
    Ah.
  174. Mallory
    @StommePoes
    Jul 21 10:14
    Yeah he was well-known for "inventing" these very complex, convoluted contraptions
  175. similar to the board game mouse trap
  176. he has a wiki page
  177. Job van Achterberg
    @jkva
    Jul 21 10:14
    Should've called him Rube Goldplate
  178. Mallory
    @StommePoes
    Jul 21 10:15
    Well the deisngers are asking, about our more "normal" date picker, if keyboarders can't get into the calendar but can type into the input, isn't that accessible?
  179. Léonie Watson
    @LJWatson
    Jul 21 10:15
    ah. We use Heath Robinson instead... though that signifies things tat are held together with sticky tape and hope, and really should never work.
  180. Mallory
    @StommePoes
    Jul 21 10:15
    and I had to say, "technically, yes. It means whatever benefits you wanted to offer by showing a calendar is only given to sightie mouse users though"
  181. aha yeah Heath is mentioned
  182. "[Rube G] is best known for a series of popular cartoons depicting complicated gadgets that perform simple tasks in indirect, convoluted ways, similar to Heath Robinson devices in the UK, as well as the Storm P devices in Denmark."
  183. Léonie Watson
    @LJWatson
    Jul 21 10:16
    I18n \o/
  184. Mallory
    @StommePoes
    Jul 21 10:17
    oh and we had a lot of fun with that too... someone wanted to use React Intl, meant in IE we get zero text for attributes
  185. So FF users get the names of the buttons, next month, previous month... IE users get "unlabelled button 0, unlabelled button 1"... le sigh.
  186. Job van Achterberg
    @jkva
    Jul 21 10:17
    I tried to capture it in a Tweet: https://twitter.com/detonite/status/756069329154744320
  187. But I really need to write good exampled
  188. *examples
  189. Mallory
    @StommePoes
    Jul 21 10:18
    Well the button showing/hiding text underneath it is one for aria-controls
  190. Job van Achterberg
    @jkva
    Jul 21 10:19
    Yes, A influences state of B
  191. Mallory
    @StommePoes
    Jul 21 10:19
    Can't think of one off the bat for aria-owns, though I remember seeing them before
  192. maybe form controls outside a form (though we have the form="" attribute for that)
  193. Job van Achterberg
    @jkva
    Jul 21 10:19
    Or details further down the page
  194. Like the address of a person that you're looking at. Gotta think of a more concrete example
  195. Mallory
    @StommePoes
    Jul 21 10:21
    Hm, those are more content-related, I thought aria-owns was more about the DOM.
  196. Job van Achterberg
    @jkva
    Jul 21 10:22
    Ha, http://tink.uk/using-the-aria-owns-attribute/
  197. But if the address isn't in the expected spot in the DOM (but is on the screen) then you'd use aria-owns
  198. Mallory
    @StommePoes
    Jul 21 10:23
    Ah, linked lists.
  199. sophieschoice
    @sophieschoice
    Jul 21 12:22
    Quick question: I want to let users download a file on my website: what is better, a link or a button?
  200. Job van Achterberg
    @jkva
    Jul 21 12:23
    I would use a link - the resource remains available
  201. anchor goes somewhere
  202. button performs an action
  203. If the button goes to an XHR call which provides dynamically-generated content, use a button
  204. But it could also be a GET to a dynamically generated resource, and then it can be a link
  205. GAH IT DEPENDS
  206. Léonie Watson
    @LJWatson
    Jul 21 12:24
    @sophies use a link - they're designed for fetching resources.
  207. sophieschoice
    @sophieschoice
    Jul 21 12:24
    it will be a static link @jkva ;)
  208. @LJWatson Thanks, Léonie! :D
  209. Job van Achterberg
    @jkva
    Jul 21 12:25
    =)
  210. sophieschoice
    @sophieschoice
    Jul 21 12:30
    But it is bad to use CSS to make the link look like a button, right?
  211. Job van Achterberg
    @jkva
    Jul 21 12:32
    @sophieschoice I would see it as bad UI, since the user gets navigated away while perhaps not expecting it, leading to bad UX
  212. I'd make the link look like a link =)
  213. Mallory
    @StommePoes
    Jul 21 12:59
    @ljWatson I noticed that I couldn't do private rooms via IRC either, thought it was just me
  214. Job van Achterberg
    @jkva
    Jul 21 13:00
    This message was deleted
  215. This message was deleted
  216. This message was deleted
  217. You can open a private room via IRC in the gitter interface if I recall correctly
  218. Lemme check
  219. Hmm I can't remember how I did that
  220. Job van Achterberg
    @jkva
    Jul 21 13:05
    /msg @sophieschoice
  221. Nope, that doesn't work either
  222. Hmm clicking the avatar brings up a dialog
  223. But that's not very accessible
  224. Job van Achterberg
    @jkva
    Jul 21 13:11
    That's weird, messages I wrote in gitter.IM don't show up here via the IRCCloud bridge
  225. yrdy
  226. *test
  227. But vice-versa works
  228. powrsurg
    @powrsurg
    Jul 21 13:23
    Do messages from other gitter.im users show up? (If you see this and you're on irccloud then yes they do)
  229. Job van Achterberg
    @jkva
    Jul 21 13:24
    YEah
  230. Likely then because it's my own account.
  231. powrsurg
    @powrsurg
    Jul 21 13:24
    Also, to go back to @sophieschoice's point the link is to download something (which I assume brings up the file download prompt) so I would say that styling it to look like a button is okay. But it should still be a link since they should be able to perform standard link actions like right click
  232. I'd include the download attribute on the link if I was designing it.
  233. But normally, yes, I would say styling a link to look like a button isn't ideal
  234. sophieschoice
    @sophieschoice
    Jul 21 13:26
    @powrsurg yes, the link would bring up the file download prompt.
  235. @powrsurg good point on the download attribute, will add it
  236. Job van Achterberg
    @jkva
    Jul 21 13:27
    @powrsurg ++ I forgot about that
  237. Mallory
    @StommePoes
    Jul 21 13:29
    That's what I thought she was planning, download attr.
  238. zakim-robot
    @zakim-robot
    Jul 21 14:39
    [car] Does anybody have a feeling for when html 5.x <menu> and <menuitem> will be fully supported across browsers so that we can stop using <div><ul><li> for menus and menubars?
    http://w3c.github.io/html/interactive-elements.html#the-menu-element
    http://w3c.github.io/html/interactive-elements.html#the-menuitem-element
    Firefox had support for context <menu> years ago (slightly different from the current spec, but close), and Chrome has supported (context only?) <menu> behind an "experimental" flag for a while now. I'm not sure what the issues are, or why it is taking so long to become mainstream. If anyone knows what's really going on with this, I'd love to know.
  239. powrsurg
    @powrsurg
    Jul 21 14:44
    I'd imagine since there are other ways of doing it it gets less developer resources than new hotness stuff like webRTC and webGL. That said, I'm surprised Apple didn't push for it to be used with Force Touch/3D Touch/whatever they called it.
  240. Seems like we'll probably get it in context menus long before we get it for regular toolbars.
  241. Also, I imagine love for toolbars is going away since they're not very mobile friendly.
  242. zakim-robot
    @zakim-robot
    Jul 21 15:01
    [car] Good point about toolbar menus and mobile. Context menu support is definitely more important so apps can merge their menu items with the browser's menu items. I've seen a couple of pretty hacky work-arounds for this: (1) right-click (or shift+F10) twice to get app context menu, then browser context menu; or (2) last menu item in app context menu is "More..." which opens browser context menu. :stuck_out_tongue_closed_eyes:
  243. powrsurg
    @powrsurg
    Jul 21 15:34
    What situations are actually good situations to really be adding things to the Context menu? I've been doing web dev for 14 years (only full time for 10 later this year), and have done a lot of work in front/back end web dev, server hosting, web performance, analytic/conversion tracking, WYSIWYG editors, mobile, accessibility things, audio playback, video streaming, a little gaming, a little smart watch dev, and am starting on VR ... but I've never had to add something to the context menu.
  244. Mallory
    @StommePoes
    Jul 21 15:35
    A game or an app with special commands is all I can think of
  245. like that crappy i hate it so much icomoon.app
  246. icomoon.io/app
  247. powrsurg
    @powrsurg
    Jul 21 15:35
    Only thing I can thing that springs to mind is in an admin when you have a row that has a button to select to add to a list you could add a menu option to select it as well
  248. Mallory
    @StommePoes
    Jul 21 15:36
    So the stuff under the crazy icon-pics look like context-menu material
  249. powrsurg
    @powrsurg
    Jul 21 15:37
    The "Add icons from library"? I'm not sure what I'm doing on that page.
  250. I feel like this is both important and I can't remember a time I've ever done it
  251. The most I've seen was when I focus on WYSIWYG stuff, but then it was the library I used (TinyMCE) that did the heavy lifting so I didn't need to touch that myself
  252. (Also, is it bad I had to track down a copy of my resume to find out how long I've been doing this?)
  253. Job van Achterberg
    @jkva
    Jul 21 15:44
    Haha
  254. You've seen too much, man
  255. powrsurg
    @powrsurg
    Jul 21 15:57
    Yeah, I had to live and breath this stuff to get by, especially with my old job. Was always on Twitter to keep up with the latest dev, or responding to emails. I had to respond to my boss and client emails while running a marathon (because heaven forbid they give me a break).
  256. zakim-robot
    @zakim-robot
    Jul 21 19:54
    [car] Our app-that-uses-menus is an IDE. Not sure if you know of the Eclipse IDE? We have a webby-cloudy version of Eclipse called Orion. Lots of stuff on the context menu, i.e. refactoring code things like "move the selected code into a function" or "rename the selected function"; code-based searchy things like "open the selected function's definition" or "find references to the selected function"; source code control stuff like "git-blame" and "show differences annotations"; and all the usual code editor stuff like Undo, Redo, Goto line#, etc. Everything has a keyboard shortcut as well. That's the code editor context menu - the file navigator context menu has other stuff. (And she cringes we also have a toolbar menu. <grin>)
  257. Job van Achterberg
    @jkva
    Jul 21 19:58
    Hello I am testing Dragon NaturallySpeaking 13 for Windows
  258. It works in Internet Explorer but does not work in chrome and or Firefox, it actually broke Google Chrome which now no longer works
  259. This is actually pretty awesome I can now click links and browse the web using my voice. Really impressed :-)
  260. sophieschoice
    @sophieschoice
    Jul 21 20:32
    lazy :P
  261. but seriously: cool technology!
  262. powrsurg
    @powrsurg
    Jul 21 20:46
    What do you guys think may be an easier sell to my boss for testing software: Jaws (most used screen reader, but pretty expensive and one would hope that things that were good in VO and NVDA would be okay in Jaws too) or Dragon (nothing really like it that is freely available).
  263. zakim-robot
    @zakim-robot
    Jul 21 21:07
    [michiel] A keyboard
  264. [michiel] Is this in addition to anything?
  265. powrsurg
    @powrsurg
    Jul 21 21:37
    I just meant for getting AT software that costs money that we should get for testing purposes, what would be an easier sell: the most used screen reader, of software to text voice commands.
  266. zakim-robot
    @zakim-robot
    Jul 21 22:56
    [marcysutton] I don’t think they want us to do this but the JAWS demo mode is free and lasts 40 minutes at a time
  267. [marcysutton] It is definitely worth testing since so many people still use it.
  268. James Nurthen
    @jnurthen
    Jul 21 23:10
    @marcysutton using JAWS like that violates their license
  269. zakim-robot
    @zakim-robot
    Jul 21 23:10
    [marcysutton] ahh.
  270. [marcysutton] I take that back then :)
  271. James Nurthen
    @jnurthen
    Jul 21 23:30
    it is why we have site licenses for JAWS