Archive index

A11y Slackers Gitter Channel Archive 27th of May 2016

What fresh hell is THIS now? - Patrick Lauke
  1. zakim-robot
    @zakim-robot
    May 27 00:04
    [michiel] toddkloots: I do apologise for the Twitter hiring practices comment; that was a joke based on an article to the same effect.
  2. [michiel] Guess I was in a shitty mood today…
  3. [toddkloots] Wasn't aware you made a comment about Twitter hiring practices
  4. zakim-robot
    @zakim-robot
    May 27 00:28
    [michiel] Then don't read up ;) But nice going on now enabling alt text for the web.
  5. [michiel] Still wish it was on by default.
  6. [michiel] I am not making friends today. Ah well, shit happens.
  7. zakim-robot
    @zakim-robot
    May 27 00:37
    [michiel] As Steve has told me before, I should not be allowed on the Internet.
  8. [nschonni] @michiel: where is your internet minder :p
  9. [michiel] Isn't that your job :P
  10. [michiel] I thought you took over from jkva when he's asleep.
  11. James Nurthen
    @jnurthen
    May 27 00:41
    @MichielBijl coming from @stevefaulkner that is strong.
  12. zakim-robot
    @zakim-robot
    May 27 00:41
    [michiel] garcialo tried to do it for a while, but that didn't really work out; he gave me the aXe.
  13. [nschonni] i think i would have noticed the danger pay on my cheque if that was the case
  14. [michiel] jnurthen: yeah it's saying something…
  15. [michiel] Nick: ha!
  16. [garcialo] @michiel: Sorry. I've been a bit busy uprooting my life.
  17. James Nurthen
    @jnurthen
    May 27 00:44
    @MichielBijl have you gone to the slack side permanently?
  18. zakim-robot
    @zakim-robot
    May 27 00:45
    [nschonni] introducing http://mullig.an/, the right-to-be-forgotten-as-a-service
  19. [michiel] garcialo: I bet.
  20. [michiel] Or bid, I always forget how that site works.
  21. [michiel] jnurthen: not sure yet.
  22. luis garcia
    @garcialo
    May 27 00:47
    Hah, it's bidding, yes.
  23. zakim-robot
    @zakim-robot
    May 27 00:48
    [michiel] Nick: that address doesn't load for me.
  24. [michiel] James: for IRCCloud the Slack bridge is better. It's just that it comes out all crappy on the Gitter IRC bridge.
  25. [michiel] It doesn't really matter for me which I use, it's just a matter of selecting a different tab in the channel list for me…
  26. James Nurthen
    @jnurthen
    May 27 00:49
    And alt text is now available on Twitter.com
  27. zakim-robot
    @zakim-robot
    May 27 00:49
    [michiel] Yeah, pretty nice :)
  28. [nschonni] it's not real. unless you're looking to burn VC, then i'll try and find what country owns ".an"
  29. [michiel] Don't think .an is a real TLD.
  30. zakim-robot
    @zakim-robot
    May 27 00:50
    [nschonni] I think you only need $180K to setup your own now
  31. [michiel] 180k? I would be close to buying my all time favourite watch with that sort of money.
  32. [michiel] <3
  33. [nschonni] Mickey Mouse Tourbillon
  34. Job van Achterberg
    @jkva
    May 27 05:50
    Blurgh. Morning slackers
  35. stevefaulkner
    @stevefaulkner
    May 27 07:44
    Morning @jkva
  36. Job van Achterberg
    @jkva
    May 27 07:44
    Steve! o/
  37. zakim-robot
    @zakim-robot
    May 27 08:38
    [michiel] Morning all!
  38. zakim-robot
    @zakim-robot
    May 27 08:50
    [jitendra] morning @michiel
  39. [michiel] o/
  40. Job van Achterberg
    @jkva
    May 27 08:58
    Morning @Michiel =)
  41. zakim-robot
    @zakim-robot
    May 27 09:05
    [michiel] waves at jkva
  42. jkva @jkva wants to make a joke about Heydon's wedding and :first-child
  43. zakim-robot
    @zakim-robot
    May 27 11:54
    [dean] Twitter ALT text is available on the web - https://www.lireo.com/15-things-to-know-twitter-alternative-text-images/ …. it needs enabling first though :(
  44. powrsurg
    @powrsurg
    May 27 13:32
    So now that Twitter has it will we ever get in on FB?
  45. powrsurg
    @powrsurg
    May 27 13:59
    Is there any spec violation for icons not having sufficient contrast? I feel like there should be, but all I've ever seen for color contrast has been in relation to text
  46. Job van Achterberg
    @jkva
    May 27 14:02
    Interesting point, yeah they replace a concept that would otherwise be text
  47. Mallory
    @StommePoes
    May 27 14:08
    If they are interactives/focusables/clickables I think icons cannot fall under the listed exceptions for contrast
  48. but those decorative icons like the warning symbol inside an alert... can be excepted as "decorative"
  49. and if they are part of text content, they ought to fall under text content, which must be contrasty enough
  50. I wonder though... do CAPTCHAs need to be contrasty enough?
  51. since their very nature means to obscure
  52. JP DeVries
    @jpdevries
    May 27 14:10
    Has this accessible component library made the rounds here yet? Looks fantastic to me so far http://frend.co
  53. Mallory
    @StommePoes
    May 27 14:10
    it followed me on twitter
  54. JP DeVries
    @jpdevries
    May 27 14:12
    always good to have a new frend ;)
  55. Mallory
    @StommePoes
    May 27 14:13
    Can Firefox users, esp at least one on Mac, click the audio captcha here and get a sound? (after clicking Register)? I'm thinking no firefoxes can play this sound but don't have enough to check
  56. JP DeVries
    @jpdevries
    May 27 14:15
    Not getting sound in Firefox (or Safari)
  57. Chrome works
  58. Jonathan Neal
    @jonathantneal
    May 27 14:16
    Has anyone here seen responsive font sizes they thought really worked on a site? Particularly, I’m looking for examples that honored the operating system’s current font size for improved accessibility. Here’s a brief introduction I just wrote up: http://codepen.io/jonneal/details/pbzGEp/
  59. James Nurthen
    @jnurthen
    May 27 14:35
    About icons and contrast. It is pretty much recognized that this is a gap in wcag. There is nothing which requires it. Of course to make something accessible you need good contrast.
  60. powrsurg
    @powrsurg
    May 27 14:35
    https://www.davidberman.com/accessibility/web-accessibility-videos/ from the guy that certified that conference software I linked yesterday
  61. low vision simulation glasses seems weird
  62. Mallory
    @StommePoes
    May 27 14:46
    They handed out a bunch of simulation glasses at a meetup I went to once
  63. it reminded us that poor vision can be all sorts of weirdness
  64. but glasses don't work so great because a lot of spots move with your eye... that would be a better simulation, which could prolly be done with VR
  65. VR and eyetracking
  66. powrsurg
    @powrsurg
    May 27 14:47
    My co-workers say I need to up my a11y game now
  67. Job van Achterberg
    @jkva
    May 27 14:47
    wot
  68. powrsurg
    @powrsurg
    May 27 14:48
    Oooh. You just gave me an idea for whenever Samsung gets around to sending me the Gear VR I was supposed to get for buying an S7.
  69. Mallory
    @StommePoes
    May 27 14:48
    I dunno of any VR headsets that do eyetracking but it's prolly not too hard for someone to implement
  70. someone who makes vr headsets I mean
  71. then you can simulare cataracts, tunnel vision, macular degeneration, glaucoma, and a bad case of floaters
  72. or even the weirder ones like snow vision
  73. powrsurg
    @powrsurg
    May 27 14:50
    snow vision?
  74. I once had a "eye migraine" and it's apparently like that except worse and all the time
  75. zakim-robot
    @zakim-robot
    May 27 14:50
    [ellyjones] I found out yesterday that the UI toolkit we're using automatically prevents you from putting text on a colored background if the contrast ratio between them isn't high enough
  76. Mallory
    @StommePoes
    May 27 14:51
    The symptoms are so particular that I could actually find out about "eye migraines" just by googling them
  77. Wow, which toolkit is that?
  78. zakim-robot
    @zakim-robot
    May 27 14:51
    [ellyjones] it's Views, the UI toolkit Chrome/Chromium use
  79. Mallory
    @StommePoes
    May 27 14:51
    never heard of it (tho I don't use UI toolkits), but that sounds very nice
  80. zakim-robot
    @zakim-robot
    May 27 14:51
    [ellyjones] if you try to use a text color that isn't readable on a background color, it changes it to either flat black or flat white as appropriate
  81. [ellyjones] it caught a mistake I made yesterday, and now I'm idly considering if Chrome should do this (or offer this) for web content
  82. Mallory
    @StommePoes
    May 27 14:52
    oooh, something you could just turn on in your settings like the font size
  83. zakim-robot
    @zakim-robot
    May 27 14:52
    [ellyjones] yeah :) "force readable text colors"
  84. Mallory
    @StommePoes
    May 27 14:53
    "Stop designers from being so designy"
  85. :P
  86. I like. Along with built-in landmark keyboard navigation
  87. zakim-robot
    @zakim-robot
    May 27 14:54
    [ellyjones] ":ballot_box_with_check: yellow on white? really?"
  88. Job van Achterberg
    @jkva
    May 27 14:55
    Client says placeholders are enough, function as label. Sigh
  89. Tempted to add labels and hide offscreen
  90. zakim-robot
    @zakim-robot
    May 27 14:55
    [ellyjones] aria labels! \o/
  91. Mallory
    @StommePoes
    May 27 14:56
    jkva do it, I always did
  92. Job van Achterberg
    @jkva
    May 27 14:56
    Yeah
  93. Mallory
    @StommePoes
    May 27 14:56
    I always had a label
  94. Job van Achterberg
    @jkva
    May 27 14:56
    Apparently thinking about semantics is "ideological"
  95. jkva @jkva runs out of sighs
  96. zakim-robot
    @zakim-robot
    May 27 14:57
    [ellyjones] is the ideology there "correct design"?
  97. Job van Achterberg
    @jkva
    May 27 14:59
    Yep
  98. Mallory
    @StommePoes
    May 27 14:59
    semantics sounds to a lot of people as "doing something just because someone made a rule about it"
  99. but then, you could say the same for writing documentation, writing tests, and modularising instead of spaghetti-ing your code
  100. zakim-robot
    @zakim-robot
    May 27 14:59
    [ellyjones] bah, one of my coworkers points out that implementing that feature ("readable text") in the UI toolkit is very easy, because the text elements just contain a flat background color, but very hard on the web because there could be 900 billion semi-transparent CSS layers and transforms and whatnot causing a cat picture to appear immediately behind the right edge of the text
  101. Mallory
    @StommePoes
    May 27 15:00
    the goal then is to convince people that semantics is more than just stopping at a red light for 5 minutes while zero traffic or pedestrians go by.
  102. zakim-robot
    @zakim-robot
    May 27 15:00
    [ellyjones] @jkva you mean semantics of your UI?
  103. Mallory
    @StommePoes
    May 27 15:00
    Elly: there is a browser plugin for chrome made by university of N carolina that can take a picture of your page and show you if contrast fails.
  104. Job van Achterberg
    @jkva
    May 27 15:00
    @ellyjones: The semantics of my HTML.
  105. Mallory
    @StommePoes
    May 27 15:01
    It's slow so the only thing I use it for is these times when images are behind text
  106. Job van Achterberg
    @jkva
    May 27 15:01
    Which is the UI, yeah
  107. zakim-robot
    @zakim-robot
    May 27 15:01
    [ellyjones] yeah, "so slow" is kind of a problem :(
  108. Mallory
    @StommePoes
    May 27 15:01
    Well, I speed it up by only having it check a part of the page
  109. I suspect being on Linux has something to do with the speed tho
  110. Job van Achterberg
    @jkva
    May 27 15:01
    http://fortune.com/2016/05/27/cellular-radiation-cancer/ HOW, the amplitude is lower than visible light
  111. I must misunderstand something fundamentally
  112. Mallory
    @StommePoes
    May 27 15:04
    Aren't microwaves also lower than visible light?
  113. Yet they can heat cells using brownian movement. Heat can change DNA in cells who are ready to split
  114. I could see it working, though not at very high numbers of cells
  115. actually, it's not brownian movement they use, sorry.
  116. just wiggling might be the technical term
  117. Job van Achterberg
    @jkva
    May 27 15:06
    @stommepoes isn't the amplitude way higher in a microwave?
  118. Mallory
    @StommePoes
    May 27 15:08
    I dunno, doesn't that depend on your power settings?
  119. Job van Achterberg
    @jkva
    May 27 15:09
    Even at the lowest, the wattage of a microwave is way higher than the transmitter of a cellphone
  120. I think a cellphone doesn't even use a whole watt to get to the nearest repeater
  121. Mallory
    @StommePoes
    May 27 15:10
    The details of the study would have to say the strength of the signals, the distances from skin, and any measurable tempurature differences (although I'm betting you can get a few tumours without regularly measuring a significant difference in heat)
  122. bleh I hate misspelling temperature.
  123. Job van Achterberg
    @jkva
    May 27 15:11
    I mean shit, my entire ham radio club should have cancer by now
  124. IF it were ionising radiation, which it's not
  125. Mallory
    @StommePoes
    May 27 15:11
    Incidences were lower for female rats \o/
  126. it doesn't have to be ionising radiation
  127. All that means is radiation that can knock an electron out of orbit, nothing more
  128. that's not the sole cause of cell damage
  129. heat also causes cell damage
  130. it's why we don't put our pets in the microwave.
  131. Job van Achterberg
    @jkva
    May 27 15:13
    Right - but the intensity / amplitude of the signal is lower than visible light, which means that going outside (ew gross where is my internet) is more damaging
  132. Mallory
    @StommePoes
    May 27 15:13
    fun fact: you can't microwave a mummy for the same reasons you can't MRI it.
  133. again we need to see where they kept these rats and how far away the source was. It says they started bombarding them in the womb.
  134. I assume they were kept in little rooms with close sources.
  135. Job van Achterberg
    @jkva
    May 27 15:15
    Maybe the rats smoked pipes
  136. Maybe they visited geocities often
  137. Mallory
    @StommePoes
    May 27 15:17
    So ES6 "classes" are just objects stuffed with the two function types we normally may use to give a new thing someone else's prototype. Sheesh.
  138. "As it happens, the control group rats actually had lower survival rates at the end of the two-year study than those exposed to the radiation—the researchers said it was possible this could have had an effect on the comparison between exposed and non-exposed rats, if the tumors in question are late-developing."
  139. zakim-robot
    @zakim-robot
    May 27 15:25
    [ellyjones] agh, someone did this bit of UI ignoring our normal UI toolkits, but instead by just drawing the pixels directly... so as a result it's totally invisible to VoiceOver and keyboard access
  140. [ellyjones] "Why??" "didn't want the overhead of making a full control"
  141. James Nurthen
    @jnurthen
    May 27 15:33
    @jkva rather than hiding the label offscreen just put aria-label on it.
  142. Job van Achterberg
    @jkva
    May 27 15:33
    @jnurthen why is that better?
  143. James Nurthen
    @jnurthen
    May 27 15:34
    simpler
  144. Mallory
    @StommePoes
    May 27 15:34
    won't work on older UAs
  145. will work with Dragon 13 tho
  146. Job van Achterberg
    @jkva
    May 27 15:34
    Oh, ok. I already had a .offscreen utility anyway
  147. *utility class
  148. James Nurthen
    @jnurthen
    May 27 15:34
    how far back do you need to go to find UA which doesn't work with aria-label?
  149. Mallory
    @StommePoes
    May 27 15:35
    a few years
  150. Job van Achterberg
    @jkva
    May 27 15:35
    But we don't have those users using our site
  151. sorry I channeled client for a second there
  152. Mallory
    @StommePoes
    May 27 15:35
    that's exactly what that ecommerce company said to me about disabled people
  153. James Nurthen
    @jnurthen
    May 27 15:36
    now browsers auto update old user agents are becoming less of a problem
  154. Mallory
    @StommePoes
    May 27 15:36
    I think it's silly for example that my company wants to start some initiative "for low-income people" yet only want to support latest-and-greatest UAs
  155. unless, like me, you turned off auto-updates because it was not compatible with the AT
  156. After I coughed up 90 euros, I could then safely update from IE9 to IE11, for example
  157. James Nurthen
    @jnurthen
    May 27 15:37
    but even IE9 supports aria-label
  158. Mallory
    @StommePoes
    May 27 15:37
    I do feel sorry for those folks who were given JAWS long ago and never heard of NVDA
  159. James Nurthen
    @jnurthen
    May 27 15:38
    yeah
  160. powrsurg
    @powrsurg
    May 27 15:38
    alternate browsers don't tend to actually keep updated too
  161. Job van Achterberg
    @jkva
    May 27 15:38
    I support all the way down to elinks / w3m on another project. I do sometimes wonder why I bother
  162. James Nurthen
    @jnurthen
    May 27 15:38
    because it makes you feel good?
  163. Job van Achterberg
    @jkva
    May 27 15:38
    And client wants
  164. Mallory
    @StommePoes
    May 27 15:38
    Because if something text-based doesn't work on a freaking text browser, you're doing something wrong
  165. it's not playing Crysis on an Atari
  166. Job van Achterberg
    @jkva
    May 27 15:39
    Yep
  167. powrsurg
    @powrsurg
    May 27 15:39
    older mobile devices that use custom browsers for things like samsung, htc, etc rarely get updated. But newer mobile devices are supposed to be updating the webview from the play store. That said ... I am not seeing that personally with my Samsung "Internet browser for Android", though others seem to be on higher branches
  168. Mallory
    @StommePoes
    May 27 15:40
    Our architect asked 'well how often do people try to use our stuff with JS off?" and I'm like, nobody! Because all that text-based coursework and text-based submit-to-server forms are completely broken without Javascript! because you deliberately coded it to break
  169. Pearson's also ignoring all the mobile browsers in the "growth" parts of the world
  170. their loss, I guess
  171. Job van Achterberg
    @jkva
    May 27 15:41
    "We'll be showing text, better demand JS"
  172. Mallory
    @StommePoes
    May 27 15:41
    Americans also don't believe in people using internet on trains
  173. Because Americans don't use trains, I guess.
  174. All Developers Have Fat Pipes.
  175. zakim-robot
    @zakim-robot
    May 27 15:42
    [ellyjones] sometimes when I deliberately swap onto simulated 2g or 3g, I get to watch various websites completely fail
  176. Job van Achterberg
    @jkva
    May 27 15:43
    I think the fast internet comes for free with the whole freedom-liberty-independence package
  177. Mallory
    @StommePoes
    May 27 15:44
    now, Americans DO have data plans
  178. James Nurthen
    @jnurthen
    May 27 15:47
  179. Job van Achterberg
    @jkva
    May 27 15:48
    I used internet on the train every day, for almost 2 years, with 3 long tunnels in the route. No fun with slow sites
  180. But I did eventually really time my usage around the tunnel entry/exit timestamps
  181. zakim-robot
    @zakim-robot
    May 27 15:49
    [ellyjones] @jnurthen when I worked on networking for chromium we routinely did those kinds of experiments, with some success
  182. Mallory
    @StommePoes
    May 27 15:50
    Someone, I think it might have been jake Archibald, did a talk on how even "having 3g" doesn't mean you even mostly load things at 3g speeds
  183. zakim-robot
    @zakim-robot
    May 27 15:50
    [ellyjones] even when you have a certain amount of throughput your latency is often much higher than developers expect, yeah
  184. James Nurthen
    @jnurthen
    May 27 15:52
    it is nice that chrome (at least) has this built into the dev tools now
  185. zakim-robot
    @zakim-robot
    May 27 15:52
    [ellyjones] yeah :)
  186. James Nurthen
    @jnurthen
    May 27 17:40
    Has anyone ever investigated the accessibility of web notifications?
  187. My completely non-thorough testing just showed the notification got read with JAWS on FF 45, but didn't get read by JAWS with FF38 or Chrome
  188. same results with NVDA
  189. IE doesn't support so can't test there at all - and don't have win10 so can't test edge
  190. zakim-robot
    @zakim-robot
    May 27 18:42
    [ellyjones] @jnurthen chrome has "accessibility of web notifications" as a current problem area on Mac, at least
  191. [ellyjones] (in that they more or less aren't)
  192. James Nurthen
    @jnurthen
    May 27 18:43
    windows too by the looks of my testing
  193. Safari on OSX w/ VO seemed to work nicely
  194. @stevefaulker any thoughts on adding things like web notifications to html5accessibility.com
  195. zakim-robot
    @zakim-robot
    May 27 19:22
    [michiel] Jnurthen: free to open PR ;)
  196. James Nurthen
    @jnurthen
    May 27 21:39
    @michiel Not sure it is appropriate at html5accessibility which is why I asked.
  197. zakim-robot
    @zakim-robot
    May 27 21:49
    [michiel] Fair enough. I think it can use an HTML 5 API section.
  198. [michiel] But might be worth waiting for the design refresh.
  199. [michiel] I don't know what other goodness dstorey has in store for us.