Archive index

A11y Slackers Gitter Channel Archive 14th of December 2015

What fresh hell is THIS now? - Patrick Lauke
  1. Veyfeyken
    09:16
    I've got a question regarding NVDA
  2. Veyfeyken
    09:16
    http://gijs.anysurfer.be/test/temp.html
  3. Veyfeyken
    09:17
    There's no stop when navigating with the arrow-keys between links. Is this a bug or a feature?
  4. Veyfeyken
    09:19
    NVDA combines the links. This can be confusing. The only way to put focus on the second link is to use the tab key. Jaws and VoiceOver stop at each link. That seems to be the way in my opinion
  5. Veyfeyken
    09:19
    "the way to go"
  6. MichielBijl
    09:21
    That would be a bug imho, but I'm not a NVDA user…
  7. MichielBijl
    09:23
    Maybe it's a setting in: http://www.nvaccess.org/files/nvda/documentation/userGuide.html?#DocumentFormattingSettings
  8. Veyfeyken
    09:27
    Don't think it's a setting. The dialog you refer to handles what is announced.
  9. pkra
    09:55
    @Veyfeyken Do you just mean up/down to read lines? I can still use CTRL+right/left arrow to navigate between words in the line. Doesn't feel like a bug to me.
  10. MichielBijl
    10:01
    @Veyfeyken sorry, I need more coffee, I'll shut up now.
  11. Veyfeyken
    10:10
    @pkra Yes, using down arrow, NVDA reads both links in one go in the first example.
  12. pkra
    10:12
    @Veyfeyken right. NVDA manual says up/down reads the entire line, so that feels expected to me. Otherwise, a sentence with a few links would be constantly disrupted, no?
  13. Veyfeyken
    10:13
    Well yes, that's VoiceOver and Jaws do. A link automatically creates a stop. So it's a bug then?
  14. Veyfeyken
    10:13
    "that's what"
  15. Veyfeyken
    10:13
    I mean, a feature.
  16. Veyfeyken
    10:14
    Let me try that again :-(
  17. Veyfeyken
    10:15
    Other screenreaders automatically stop when they encounter a link in flow content. NVDA doesn't. What is the expected behaviour?
  18. Veyfeyken
    10:17
    When 2 links follow each other directly, this can be problematic. For example: http://www.flanders.be/en
  19. Veyfeyken
    10:18
    After the logo and search, there's a link "home", followed by an icon that opens a dropdown. The hidden link text is "show navigation menu".
  20. Veyfeyken
    10:18
    NVDA reads: link home link show navigation menu and puts the focus on "home".
  21. Veyfeyken
    10:19
    Using down arrow, I can't reach "show navigation menu"
  22. pkra
    10:35
    @Veyfeyken sorry, didn't see these without a mention. I'm afraid I don't know if there's an expected behavior. Just that I can see the benefits of both approaches. I'm also not sure that the example on flanders.be is necessarily problematic. I'm not a screenreader user myself but based on I got from users in the past, it doesn't seem strange to identify two links (and a collapse) in the reading of that structure. But again, I'm not a screenreader user myself.
  23. Veyfeyken
    10:36
    @pkra Thanks for the feedback.
  24. pkra
    10:37
    @Veyfeyken I lost a "feedback" in "based on feedback" there. sry.
  25. Veyfeyken
    10:43
    @pkra It's not a bug. Answer from James Teh: nvaccess/nvda#5605
  26. pkra
    10:44
    @Veyfeyken cool. James is the best.
  27. powrsurg
    14:21
    Has anyone here tested the accuracy in Visual Studios Code's auto-completes for ARIA attributes?
  28. zakim-robot
    15:02
    [jitendra] I have a question.
  29. zakim-robot
    15:03
    [jitendra] in HTML, Forms control are limited. Like this kind of control is not possible with any form element
    http://take.ms/A4Rz3
  30. zakim-robot
    15:04
    [jitendra] So is it ok to use javascript/CSS to build custom control or best practice is still to do only what is possible with HTML forms only?
  31. garcialo
    15:05
    Well, in that case, it looks like the "select" is really just being used to display/not display the checkbox grouping.
  32. garcialo
    15:08
    I don't think there'd be a problem using javascript to do it if it needs to look that way. Just don't think of it as BOTH a select dropdown and group of checkbox.
  33. MichielBijl
    15:59
    Looks like a dialog with a checkbox group
  34. garcialo
    16:02
    I was thinking menu button
  35. garcialo
    16:02
    menu button/menu
  36. garcialo
    16:03
    ...which I guess is a select dropdown =p
  37. MichielBijl
    16:34
    @jitendra: can you select multiple items? Or just the one?
  38. MichielBijl
    16:35
    If you can only select one it's just a weird way to display a select box thing =p
  39. zakim-robot
    16:35
    [jitendra] multiple select
  40. MichielBijl
    16:35
    @garcialo but a menu with just one item?
  41. garcialo
    16:36
    I don't know, I'm trying to speak up more in these conversations because it's a weak point for me
  42. MichielBijl
    16:37
    Oh I'm not judging. The only way to find out is to discuss 700 thousand versions ; filter those to the best and test them.
  43. MichielBijl
    16:37
    Or build 700 thousand tests =p
  44. garcialo
    16:39
    Yeah, I'll probably get around to doing that, but I have been busy with work/side project presentation for CSUN/stupid life stuff =p
  45. MichielBijl
    16:40
    You should see my list of “stuff to test”
  46. MichielBijl @MichielBijl will never finish that list
  47. jonathantneal
    16:43
    Anyone remember a blog that recently encouraged people to consider using <button> over type="button" and/or to not <button type="button"> ?
  48. MichielBijl
    16:44
    http://html5doctor.com/html-developers-please-consider/ ??
  49. MichielBijl
    16:45
    Only other article in my reading list is: http://www.karlgroves.com/2013/05/14/links-are-not-buttons-neither-are-divs-and-spans/ But that is not recent (has good points though).
  50. jonathantneal
    16:47
    @MichielBijl, thank you, that was it!
  51. MichielBijl
    16:47
    Glad I could help =)
  52. jonathantneal
    16:48
    Did I just correctly learn that the default type for button is submit?
  53. MichielBijl
    16:51
    IIRC it is, yes.
  54. zakim-robot
    16:51
    [jitendra] So what would be the best solution for my answer?
  55. MichielBijl
    16:52
    “The missing value default is the Submit Button state.”—http://www.w3.org/TR/html5/forms.html#attr-button-type
  56. alexlande
    16:54
    For safety, it's generally a good idea to specify <button type="button">, lest your button winds up in a form one day and accidentally starts submitting it
  57. MichielBijl
    16:54
    @jonathantneal I don't see the HTML5doctor article suggesting you shouldn't use type="button" though?
  58. jonathantneal
    16:56
    From my original comment, I remembered it incorrectly. Since it doesn’t say anything, that’s when I researched type="button" and learned that second tidbit somewhere else on the internets. :)
  59. alexlande
    16:56
    I think using role="button" is generally a smell, but type should be safe
  60. pkra
    17:48
    Broad question: Has anyone ever tried provisioning vagrant boxes from modern.ie to simplify testing of e.g., NVDA, JAWS, Window Eyes test version? I've looked at it sporadically and it still seems like both the VMs from Microsoft as well as AT installations are a bit of a mess (as in: I could never get it to work).
  61. pkra
    17:48
    Related question / business idea: something like browserstack, saucelabs etc for AT testing.
  62. MichielBijl
    17:54
    @pkra what is your host OS?
  63. pkra
    17:54
    Linux (ubuntu)
  64. pkra
    17:55
    @MichielBijl (didn't tag you there)
  65. MichielBijl
    17:55
    Hmm. Marco Zehe wrote about problems on OS X (capslock doesn't work as modifier key).
  66. MichielBijl
    17:55
    Not sure if that is related to the issues you're experiencing.
  67. MichielBijl
    17:56
    Here is the article for anyone interested: https://www.marcozehe.de/2015/06/07/how-to-map-your-macs-capslock-key-to-a-nvda-or-jaws-key-in-a-windows-virtual-machine/
  68. pkra
    17:58
    Ah, no, I'm talking about something different
  69. pkra
    17:58
    @MichielBijl I'm looking for building a vagrant file that will allow me to start a clean VM that on first load installs one or more screenreader demos.
  70. MichielBijl
    17:59
    Ah, excuse me.
  71. pkra
    17:59
    @MichielBijl I don't have a problem using the AT :)
  72. pkra
    17:59
    @MichielBijl no worries :-)
  73. MichielBijl
    17:59
    Can't you make a snapshot after you installed the demo's?
  74. MichielBijl
    17:59
    And just restore on either boot or close?
  75. MichielBijl
    17:59
    That's what I do with my test VM's.
  76. pkra
    18:00
    yes, but that doesn't help my colleagues, hundreds of miles away.
  77. pkra
    18:00
    vagrant files would allow us to have reproducible environments.
  78. pkra
    18:00
    Also, I don't have space for one VM for each windows version and screenreader combo I'm testing :)
  79. pkra
    18:00
    But I noticed that differetn AT can lead to problems.
  80. pkra
    18:01
    different AT on the same machine that is.
  81. pkra
    18:01
    Anyway, for reference, there's https://joecod.es/modernie-for-vagrant-reloaded/
  82. pkra
    18:02
    One day, I'll get it to work and learn how to write powershell scripts that work...
  83. pkra
    18:02
    Or to put it differently, if anyone is interested in banging heads together over this, I'm game...
  84. zakim-robot
    18:38
    [callumacrae] Are there any stats on the number of people using screenreaders?
  85. powrsurg
    18:39
    Maybe this is incorrect to assume, but was anyone here affiliated with the W3C's Dev Ave/online courses? Is there anything accessibility related that is of concern with creating / serving badges?
  86. garcialo
    18:40
    @callumacrae I don't know of any; I only know of the WebAIM survey which surveys screen reader users.
  87. alexlande
    18:43
    Yeah, I've never seen any stats on number of readers. There isn't a reliable way to detect screen readers for analytics (http://www.karlgroves.com/2013/05/13/can-we-track-how-many-users-with-disabilities-access-our-site/)
  88. powrsurg
    19:58
    haven't some browsers added (or started to add) media queries that are for some thigns you can detect?
  89. deborahgu
    19:58
    callumacrae: the problem with AT stat gathering is that a lot of us don't even want reliable ways to do detection, because they would likely be misused. E.g. only enabling "accessibility mode" if a screen reader is detected.
  90. deborahgu
    19:58
    Not to mention privacy.
  91. fstorr
    20:45
    @stevefaulkner (pinging you as you're the only person I'm aware of that's working on W3C standards). The ARIA 1.0 Authoring practices document recommends multiple <h1> elements per page which goes against the HTML5.1 spec. Relevant content: http://www.w3.org/WAI/PF/aria-practices/#kbd_layout_header
  92. stevefaulkner
    20:52
    @fstorr thanks will check it out
  93. stevefaulkner
    20:55
    @fstorr note the WD you pointed to is stale here is latest http://w3c.github.io/aria/practices/aria-practices.html
  94. stevefaulkner
    20:56
    May be same in up to date version as well
  95. MichielBijl
    21:02
    @fstorr you can file bugs on github: https://github.com/w3c/aria/issues
  96. MichielBijl
    21:02
    Would be much appreciated :D
  97. MichielBijl
    21:03
    @stevefaulkner yes, same in current WD.
  98. MichielBijl
    21:05
    Here are the minutes for todays APG call for anyone interested: http://www.w3.org/2015/12/14-aria-apg-minutes.html
  99. MichielBijl
    21:06
    TL;DR: We have a new radio group code example: https://cdn.rawgit.com/w3c/aria/master/practices/examples/radio/radio.html. And we need to create some sort of tooltip/bubble help/popover help design pattern section.
  100. StommePoes
    21:10
    I haz a question for the group, as I have no idea what others are doing
  101. StommePoes
    21:10
    in the case of a web page where a user may toggle the human language
  102. StommePoes
    21:10
    and the whole freakin page is written in le javascreeept
  103. StommePoes
    21:11
    the content everywhere changes from say French to Russian when the user toggles
  104. StommePoes
    21:11
    without a page refresh, is there any sane way to ensure the AT is brought up to date on anything it has made copies of?
  105. StommePoes
    21:11
    With Orca this isn't an issue as it doens't buffer
  106. StommePoes
    21:11
    VO I heard was similar
  107. garcialo
    21:12
    changing the lang attribute doesn't work?
  108. StommePoes
    21:12
    I'm not sure
  109. MichielBijl
    21:12
    @StommePoes we can use GitHub to collab on a test =p
  110. StommePoes
    21:12
    I don't have this application with me
  111. garcialo
    21:12
    I'll mock it up
  112. StommePoes
    21:12
    I was asked as a "how would this work"
  113. StommePoes
    21:13
    oh cool
  114. MichielBijl
    21:13
    Nice!
  115. garcialo
    21:13
    but I would expect that as long as you change the lang attribute on the html element, it should work out
  116. MichielBijl
    21:14
    To the science lab!
  117. StommePoes
    21:14
    so we've got developers who've read this page https://developer.mozilla.org/en-US/docs/Web/CSS/font-weight?redirectlocale=en-US&redirectslug=CSS%2Ffont-weight
  118. StommePoes
    21:14
    and now they think those numbers matter
  119. MichielBijl
    21:15
    100-900?
  120. StommePoes
    21:15
    they were heavily debating whether font size 100 should be bumped up to 200 because it looked maybe a little bad on retina
  121. StommePoes
    21:15
    wtf
  122. MichielBijl
    21:15
    I'd expect designers to worry about that.
  123. StommePoes
    21:15
    can retina screeens suddenly fix the fact that browsers have two settings: normal and bold?? Or, did browsers suddenly start supporting multiple font weights??
  124. MichielBijl
    21:15
    But I guess, designers are developers and vice versa.
  125. StommePoes
    21:16
    Last I heard the only support for font weights was safari-with-multiple-@font-face-webfonts
  126. MichielBijl
    21:16
    @StommePoes most browsers do support multiple font weights?
  127. StommePoes
    21:16
    these are developers
  128. StommePoes
    21:16
    Do they??
  129. MichielBijl
    21:16
    AFAIK
  130. StommePoes
    21:16
    I have only seen normal and bold, with all the numbers mapping to either 1 or the other
  131. MichielBijl
    21:16
    It's the fonts that need support too.
  132. StommePoes
    21:16
    with the one exception of safari if a font already on a mac happened to be so large as to actually have multiple font weights
  133. StommePoes
    21:17
    well here they are discussing helvetica and arial
  134. StommePoes
    21:17
    and we all know on Windows the user will always get arial in that case
  135. StommePoes
    21:17
    so does arial on the web ever have multiple font weights??
  136. MichielBijl
    21:17
    If I feed Safari, Chrome or Firefox with 3MB of webfonts that support multiple font weights, it'll work.
  137. MichielBijl
    21:18
    Not sure about Arial, Helvetica for sure.
  138. MichielBijl
    21:18
    Just use Helvetica Neue, that is not mapped to Arial on Windows.
  139. StommePoes
    21:18
    @fontface you mean
  140. StommePoes
    21:19
    but with "helvetica neue",helvetica,arial,san-serif there should be two: normal and bold
  141. MichielBijl
    21:19
    In that case font-weight: 200 should just fallback to 400 right?
  142. StommePoes
    21:19
    Does a mac, with helvetica, show three different things for 400 and 600 and 800?
  143. MichielBijl
    21:19
    Let me try.
  144. StommePoes
    21:19
    yeah either 400 or 500 was the tipping point from normal to bold
  145. StommePoes
    21:20
    •If a weight greater than 500 is given, the closest available darker weight is used (or, if there is none, the closest available lighter weight).
    •If a weight less than 400 is given, the closest available lighter weight is used (or, if there is none, the closest available darker weight).
  146. MichielBijl
    21:24
    Test file: http://dir.agosto.nl/dump/font-weight.html, result in Safari 9: http://dir.agosto.nl/dump/font-weight.png alt= list that shows all different font-weights (100-900) for Helvetica Neue in Safari 9.
  147. MichielBijl
    21:26
    FYI: result is the same in Chrome on OS X.
  148. StommePoes
    21:27
    wow
  149. MichielBijl
    21:27
    IT'S SO PRETTY!
  150. StommePoes
    21:29
    https://pbs.twimg.com/media/CWN2IsJWUAAFbqD.png:large
  151. MichielBijl
    21:29
    No luck on Windows I see.
  152. MichielBijl
    21:29
    But that is probably because it now falls back to Arial.
  153. MichielBijl
    21:30
    And Arial doesn't support it apparently.
  154. StommePoes
    21:31
    http://snook.ca/archives/html_and_css/windows-subs-helvetica-arial lawlz
  155. MichielBijl
    21:31
    Yeah, it's because of Arial not supporting more font-weights than 400 & 700.
  156. MichielBijl
    21:31
    If I test it with Arial as the font, it shows the same result as on your Windows thing.
  157. MichielBijl
    21:32
    But this would work if we'd load up a webfont that supported it.
  158. powrsurg
    21:33
    there are very few fonts that actually use all 100 - 900
  159. powrsurg
    21:33
    well, I guess there are probably a lot over the whole course of the web, but it's very uncommon for it
  160. StommePoes
    21:34
    well, since it seems everyone but me on the dev team is a mac user, I guess they can worry about font weights
  161. powrsurg
    21:34
    I think the most I ever was called to use was a font with 200, 400, 600, 700, 900
  162. StommePoes
    21:34
    it'll affect them if not most of the users
  163. MichielBijl
    21:34
    @StommePoes what do you get with this: http://dir.agosto.nl/dump/font-weight-webfont.html
  164. StommePoes
    21:34
    @michielBijl that one does the multiple font weights
  165. StommePoes
    21:35
    but I've done that with for example raleway from font squirrel
  166. StommePoes
    21:35
    but I had to set a different font per "weight"
  167. StommePoes
    21:35
    rather than the weight itself
  168. MichielBijl
    21:35
    I loaded up Open Sans in that. But as @powrsurg noted, not 100-900 only 200, 400, 600, and 800.
  169. powrsurg
    21:35
    honestly, you should be testing your design with at the very least the fonts you fall back to in your font-family rule
  170. StommePoes
    21:35
    open sans, cool
  171. StommePoes
    21:35
    they're using helvetica and arial, which means on windows always arial
  172. MichielBijl
    21:36
    Uhm @StommePoes yes. You need multiple @font-face's if that's what you're getting at.
  173. StommePoes
    21:36
    yeah so how many mb was that?
  174. powrsurg
    21:36
    surprised they aren't trying to use "helvetica neue". Mac designers used to love it ...
  175. StommePoes
    21:36
    they do put that in the fonrt
  176. StommePoes
    21:36
    front
  177. StommePoes
    21:37
    it is indeed a bit nicer than helvetica
  178. StommePoes
    21:37
    and these are developers
  179. powrsurg
    21:37
    which of course isn't a free font so you'd need to license it, which is annoying since there are like 70 versions of it on fonts.com
  180. MichielBijl
    21:37
    @StommePoes Open Sans is either really small or Chrome is lying, it says 50kb.
  181. StommePoes
    21:37
    does it come with any mac software?
  182. MichielBijl
    21:37
    Helvetica Neue comes with OS X.
  183. MichielBijl
    21:37
    And iOS.
  184. StommePoes
    21:37
    hm 50kb is amazing for 6 weights
  185. powrsurg
    21:38
    it does, but don't you want to use the same font for everyone that uses your site?
  186. powrsurg
    21:38
    or at least try to
  187. StommePoes
    21:38
    No
  188. powrsurg
    21:38
    that means paying the license
  189. StommePoes
    21:38
    Well, they want mac people to have helvetica and they want windows people to have helvetica
  190. StommePoes
    21:38
    but windows people will not get helvetica
  191. StommePoes
    21:38
    and anyways, both fonts suck the biggest balls for Il letters
  192. MichielBijl
    21:39
    Doesn't Lucida Grande come with Windows?
  193. StommePoes
    21:39
    I feel neither font is appropriate for learning or reading text where it matters
  194. StommePoes
    21:39
    I dunno, it was on Linuz
  195. StommePoes
    21:40
    http://www.apaddedcell.com/web-fonts
  196. StommePoes
    21:40
    old but I remember back when I used to use that for font stacks
  197. MichielBijl
    21:40
    Hmm, Lucida Grande only support 400/800.
  198. StommePoes
    21:41
    I'm going to tell them to either use Verdana or other font that sufficiently shows differences between I,l,and 1
  199. StommePoes
    21:41
    and maybe a few other ones
  200. StommePoes
    21:42
    Verdana is close and on many OSes, but designers hate it because it has a large x-height
  201. MichielBijl
    21:42
    http://dir.agosto.nl/dump/Screen%20Shot%202015-12-14%20at%2022.38.03.png alt=Web Inspector on Network tab shows 6 web font loaded (Open Sans in multiple weights) weighing in at 50KB.
  202. StommePoes
    21:42
    10k for a font
  203. StommePoes
    21:42
    unbelievable
  204. StommePoes
    21:42
    maybe they have compression in a special way
  205. MichielBijl
    21:43
    I guess so.
  206. StommePoes
    21:47
    "The name is commonly misspelled Levan polkka, due to the similarity of lower-case L and upper-case i in sans-serif fonts." lawlzy lawlz
  207. StommePoes
    21:48
    not that verdana is not immune to bad letters http://typography.guru/journal/letters-symbols-misrecognition
  208. garcialo
    21:53
    @StommePoes Changing the lang attribute (along with all the other text) seems to work just fine with NVDA+Firefox on Windows 10. http://codepen.io/garcialo/pen/GooKPp/
  209. garcialo
    21:55
    ...I should really update that to announce the language change =p
  210. MichielBijl
    21:55
    What is expected to happen.
  211. MichielBijl
    21:56
    I don't hear a change in pronunciation, but I don't think I have any Spanish languages installed.
  212. garcialo
    21:56
    the SR will dynamically adjust to the changed lang attribute when the entire page language changes
  213. MichielBijl
    21:56
    I'll fork it and try it with Dutch.
  214. garcialo
    21:56
    sure
  215. garcialo
    21:58
    Dutch is one of the languages I never learned that phrase in
  216. MichielBijl
    21:58
    Dutch version: http://codepen.io/anon/pen/eJJOqj
  217. MichielBijl
    21:59
    It doesn't dynamically load a different Voice for me, but it does announce the right content…
  218. garcialo
    21:59
    it changes the pronunciation for me
  219. garcialo
    22:00
    as it should
  220. garcialo
    22:00
    even with the Dutch
  221. MichielBijl
    22:00
    VoiceOver doesn't for me =(
  222. garcialo
    22:00
    it's very similar to the German
  223. MichielBijl
    22:00
    Same language group
  224. garcialo
    22:01
    yeah
  225. MichielBijl
    22:02
    Maybe VO is just lazy…
  226. garcialo
    22:02
    maybe, I'll try JAWS
  227. MichielBijl
    22:03
    Maybe we should document this :P
  228. garcialo
    22:03
    oh geez; yeah, it doesn't work with JAWS+IE
  229. MichielBijl
    22:04
    Can you create a repository? I don't want to take credit for your code ;)
  230. MichielBijl
    22:04
    We can use layout from my font-awesome project.
  231. garcialo
    22:04
    as awesome as the code is, I'm fine with you taking credit for it
  232. MichielBijl
    22:05
    I'll create a repo and will credit you anyway.
  233. garcialo
    22:05
    or putting it in your repo =p
  234. garcialo
    22:05
    kk
  235. MichielBijl @MichielBijl was going to sleep early today… WHO NEEDS SLEEP ANYWAY?!
  236. garcialo
    22:05
    add me as editor and we can eventually put a proper language switcher on it
  237. garcialo
    22:05
    sleep is for the weak
  238. fstorr
    22:29
    @MichielBijl will file a bug.
  239. MichielBijl
    22:40
    @garcialo low and behold: https://github.com/MichielBijl/language-switch
  240. MichielBijl
    22:40
    @fstorr thank you!
  241. garcialo
    22:40
    it is a thing of beauty
  242. MichielBijl
    22:40
    :D
  243. MichielBijl
    22:41
    And for 35 minutes work; not bad.
  244. garcialo
    22:41
    hah
  245. MichielBijl
    22:42
    Need to fix the title of index :')
  246. garcialo
    22:42
    yup
  247. MichielBijl
    22:42
    Done.
  248. MichielBijl
    22:43
    And I'm pretty sure you twitter handle isn't @StommePoes :')
  249. garcialo
    22:44
    there's no way of knowing for sure
  250. MichielBijl
    22:44
    Haha
  251. MichielBijl
    23:06
    Wish I could go to CSUN16 =(
  252. garcialo
    23:06
    send me a GoPro and I'll livestream my experience for you
  253. MichielBijl
    23:07
    Haha, that is nearly half the money I'd need to go :P
  254. garcialo
    23:07
    all of the excitement for half the cost!
  255. MichielBijl
    23:07
    Hah
  256. garcialo
    23:07
    and none of the walking!
  257. MichielBijl
    23:07
    And flying
  258. MichielBijl
    23:08
    But also less talking with people =(
  259. garcialo
    23:08
    yeah, it probably would have been delayed anyways, and you'd end up stranded somewhere in the U.S.
  260. MichielBijl
    23:10
    Anyway, time for bed. God nat!
  261. garcialo
    23:10
    night