/irc-logs / mozilla / #accessibility / 2012-12-14 / end

Options:

  1. # Session Start: Fri Dec 14 00:00:00 2012
  2. # Session Ident: #accessibility
  3. # [00:21] * ehsan|brb is now known as ehsan
  4. # [00:43] * Quits: @hub (hub@CAB30FBD.8F96AEA7.2D179A7D.IP) (Ping timeout)
  5. # [01:13] * Quits: Jonadabe (Jonadabe@moz-2FCDBC88.dsl.telepac.pt) (Quit: http://www.visitportugal.com | http://www.youtube.com/visitportugal)
  6. # [01:29] * Quits: satdav (satdav@moz-2B0F19ED.cable.virginmedia.com) (Connection reset by peer)
  7. # [01:55] * Quits: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
  8. # [02:07] * Quits: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net) (Quit: scott_gonzalez)
  9. # [02:10] * Joins: hub (hub@CAB30FBD.8F96AEA7.2D179A7D.IP)
  10. # [02:10] * ChanServ sets mode: +o hub
  11. # [02:26] * Quits: @hub (hub@CAB30FBD.8F96AEA7.2D179A7D.IP) (Ping timeout)
  12. # [02:28] * Quits: nhirata (anonymous@moz-BE33DA21.fw1.sfo1.mozilla.net) (Quit: nhirata)
  13. # [03:00] * khuey is now known as khuey|away
  14. # [04:59] * Joins: ehsan (ehsan@576FC5E5.CD1D2BB8.DC1243F8.IP)
  15. # [05:03] * Quits: ehsan (ehsan@576FC5E5.CD1D2BB8.DC1243F8.IP) (Input/output error)
  16. # [05:04] * Joins: surkov (surkov@1D099191.75504D41.E17943EE.IP)
  17. # [05:04] * ChanServ sets mode: +o surkov
  18. # [05:16] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
  19. # [05:38] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
  20. # [05:59] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
  21. # [07:24] * Quits: @surkov (surkov@1D099191.75504D41.E17943EE.IP) (Quit: surkov)
  22. # [07:24] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Client exited)
  23. # [08:02] * Joins: victorporof (victorporo@87CAE4F1.BEB908B0.79933D60.IP)
  24. # [08:23] * Joins: Stevef (chatzilla@moz-CD0F47B5.cable.virginmedia.com)
  25. # [09:35] * Joins: surkov (surkov@1D099191.75504D41.E17943EE.IP)
  26. # [09:35] * ChanServ sets mode: +o surkov
  27. # [09:44] * Joins: marcoz (marco.zehe@moz-81CEB0D3.dip.t-dialin.net)
  28. # [09:46] * ChanServ sets mode: +o marcoz
  29. # [10:21] * Quits: victorporof (victorporo@87CAE4F1.BEB908B0.79933D60.IP) (Quit: victorporof)
  30. # [10:26] * Quits: @surkov (surkov@1D099191.75504D41.E17943EE.IP) (Quit: surkov)
  31. # [10:51] * Joins: victorporof (victorporo@87CAE4F1.BEB908B0.79933D60.IP)
  32. # [10:59] * Quits: @marcoz (marco.zehe@moz-81CEB0D3.dip.t-dialin.net) (Quit: Leaving.)
  33. # [11:10] * Joins: surkov (surkov@1D099191.75504D41.E17943EE.IP)
  34. # [11:10] * ChanServ sets mode: +o surkov
  35. # [12:04] * Joins: Jonadabe (Jonadabe@moz-38059BFE.dsl.telepac.pt)
  36. # [13:47] * Joins: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net)
  37. # [14:42] * Joins: hub (hub@F2D29657.F60B0462.67AC9B1.IP)
  38. # [14:42] * ChanServ sets mode: +o hub
  39. # [14:52] * Joins: askalski (akuda@moz-137D1D7A.mimuw.edu.pl)
  40. # [14:57] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
  41. # [14:57] * ChanServ sets mode: +qo davidb davidb
  42. # [15:05] <@firebot> jacek@codeweavers.com requested review from trev.saunders@gmail .com for attachment 692259 on bug 819280.
  43. # [15:05] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=819280 nor, --, mozilla20, jacek, REOP, Fix GCC warnings in Windows accessibility code
  44. # [15:06] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
  45. # [15:17] * Quits: askalski (akuda@moz-137D1D7A.mimuw.edu.pl) (Ping timeout)
  46. # [15:18] * Joins: askalski (akuda@moz-137D1D7A.mimuw.edu.pl)
  47. # [15:28] * Quits: Wolf (chatzilla@moz-F8C1AF63.carolina.res.rr.com) (Connection reset by peer)
  48. # [15:29] * Joins: Wolf (chatzilla@moz-F8C1AF63.carolina.res.rr.com)
  49. # [15:49] <@davidb> heyo!
  50. # [15:52] * Joins: habber (habber@moz-8654C0B7.nyc.res.rr.com)
  51. # [15:56] <@davidb> surkov: warning, I'll probably want you to join my w3c calls when you're in Toronto :)
  52. # [15:58] <@firebot> surkov.alexander@gmail.com changed the Status on bug 821269 from UNCONFIRMED to NEW.
  53. # [15:58] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=821269 nor, --, ---, nobody, NEW, Fix build warnings in accessible/src/html
  54. # [15:59] * Quits: askalski (akuda@moz-137D1D7A.mimuw.edu.pl) (Quit: Wychodzi)
  55. # [16:04] <@firebot> jacek@codeweavers.com changed the Resolution on bug 819280 from FIXED to ---.
  56. # [16:04] <@firebot> jacek@codeweavers.com changed the Status on bug 819280 from RESOLVED to REOPENED.
  57. # [16:04] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=819280 nor, --, mozilla20, jacek, REOP, Fix GCC warnings in Windows accessibility code
  58. # [16:18] * Quits: @surkov (surkov@1D099191.75504D41.E17943EE.IP) (Quit: surkov)
  59. # [16:34] <Stevef> davidb: why are you going to punish surkov?
  60. # [16:36] <@davidb> Stevef: shhhh! it is so much fun!
  61. # [16:38] <Stevef> davidb: surkov should be allowed to join w3c calls where ever his not only while in toronto :-)
  62. # [16:38] <@davidb> yeah the timing is difficult
  63. # [16:38] <@davidb> (he is "allowed" of course)
  64. # [16:42] <Stevef> he'll leave toronto after a few calls am sure, ensure he no longer has fone access....
  65. # [16:46] * clown notes that surkov has suspiciously left the building.
  66. # [16:49] <@hub> clown: it is very late for him
  67. # [16:50] <clown> hub, indeed. it might already by Sat for him.
  68. # [16:50] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
  69. # [16:56] <@davidb> it is
  70. # [16:57] <clown> davidb, lucky him; it's already the weekend.
  71. # [16:58] <@davidb> i like fridays
  72. # [16:58] <clown> me too at about 5pm...
  73. # [17:11] * khuey|away is now known as khuey
  74. # [17:35] * habber is now known as habber-away
  75. # [17:36] <clown> davidb, I came across a picture of our building: https://maps.google.com/maps?q=205+Richmond+Street+West,+Toronto,+ON,+Canada&layer=c&z=17&iwloc=A&sll=43.649461,-79.388749&cbp=13,158.3,0,0,0&cbll=43.649581,-79.388815&hl=en&ved=0CAoQ2wU&sa=X&ei=3VPLUMGTI4WsxQHY1ICQBg
  76. # [17:36] <clown> that's me in the truck.
  77. # [17:36] <clown> not.
  78. # [17:42] <@davidb> heh
  79. # [18:12] * Joins: nhirata (anonymous@moz-6D599AA6.hsd1.ca.comcast.net)
  80. # [18:20] * Quits: Stevef (chatzilla@moz-CD0F47B5.cable.virginmedia.com) (Ping timeout)
  81. # [18:50] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
  82. # [19:04] <@davidb> clown: your tweet to me looks suspicious did you send it? (haven't clicked it yet)
  83. # [19:09] <clown> davidb, yep that's from me.
  84. # [19:09] <clown> what makes it look suspicious?
  85. # [19:09] <@davidb> heh
  86. # [19:09] <@davidb> hard to find humour right now...
  87. # [19:09] <@davidb> clown: http://www2.wbtw.com/news/2012/dec/14/5/school-shooting-reported-newtown-conn-ar-5169555/
  88. # [19:10] <clown> crap
  89. # [19:22] <scott_gonzalez> davidb clown: Did either of you see my questions about indeterminate progressbars yesterday?
  90. # [19:22] <clown> scott_gonzalez: no, where did you post it?
  91. # [19:22] <scott_gonzalez> In here.
  92. # [19:22] * clown trys to remember if he was here yesterday...
  93. # [19:23] <@davidb> sorry I'm barely here ATM
  94. # [19:23] <clown> scott_gonzalez: I found this in the logs: "In jQuery UI, we're using role="progressbar" aria-valuemin="0" for indeterminate progress indicators, but NVDA and JAWS don't annouce anything. Is Firefox exposing anything in that case?"
  95. # [19:23] <scott_gonzalez> After the discussion yesterday, my first question would be: Is there anything in the OS accessibility layers for indeterminate progressbars.
  96. # [19:24] <clown> scott_gonzalez: good question. *looks*
  97. # [19:24] <scott_gonzalez> tbsaunde said he wasn't sure that there was anything.
  98. # [19:24] * ehsan is now known as ehsan|away
  99. # [19:24] <scott_gonzalez> Firefox just exposes an empty valuetext for the ARIA properties we're using.
  100. # [19:24] <scott_gonzalez> Which, I suppose is reasonable, if there's nothing else to expose.
  101. # [19:25] <scott_gonzalez> VoiceOver does announce it propertly, even in Firefox.
  102. # [19:27] <clown> scott_gonzalez: ATSPI has a "ATSPI_STATE_INDETERMINATE". It's documented as being used for (1) tri-state checkboxes (checked, unchecked, mixed).
  103. # [19:28] <clown> if you want, I can look at what FF outputs to the a11y layer for your case — do you have an url?
  104. # [19:28] * @davidb notes FF should expose STATE_MIXED and no value now or value text
  105. # [19:28] <scott_gonzalez> http://view.jqueryui.com/master/demos/progressbar/indeterminate.html
  106. # [19:28] <scott_gonzalez> Though the progressbar doesn't have a tabindex, so you can't focus it.
  107. # [19:30] <clown> I don't need to (I think). I just want to look at the a11y tree and see what states/properties it has.
  108. # [19:33] <clown> scott_gonzalez: on Linux/GNOME, FF is outputting an ATSPI_STATE_INDETERMINATE state for that progress bar. Also, its own internal "common a11y layer" is showing "mixed" as one of the states.
  109. # [19:34] <clown> but, I have no idea what FF is doing on OS X.
  110. # [19:34] <@firebot> bugzilla@mversen.de changed the Resolution on bug 776272 from --- to WORKSFORME.
  111. # [19:34] <@firebot> bugzilla@mversen.de changed the Status on bug 776272 from UNCONFIRMED to RESOLVED.
  112. # [19:34] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=776272 nor, --, ---, nobody, RESO WORKSFORME, Wikimedia keyboard shortcuts not working
  113. # [19:35] <clown> FF 17.0 if it matters...
  114. # [19:35] <scott_gonzalez> clown: How I can check that level of detail on OS X?
  115. # [19:35] * clown wonders what orca says...
  116. # [19:35] <scott_gonzalez> I installed the DOM Inspector extension yesterday.
  117. # [19:35] <clown> scott_gonzalez: alas I don't know the details. marcoz might. Is he here?
  118. # [19:36] <scott_gonzalez> I do see that it says mixed state.
  119. # [19:36] <clown> scott_gonzalez: that helps (the DOM inspector) — does that show the a11y info on OS X? It does on windows and gnome.
  120. # [19:36] <scott_gonzalez> I didn't notice that yesterday, it's in a separate section from the rest of the attributes.
  121. # [19:37] <clown> yes, it's in the a11y object attributes.
  122. # [19:37] <scott_gonzalez> http://cl.ly/image/2c1r1r2v0n0j
  123. # [19:37] <scott_gonzalez> So it sounds like Firefox is exposing what it should.
  124. # [19:37] <clown> whoops. No, I mean it's in the a11y states section, not the object attributes.
  125. # [19:37] <scott_gonzalez> But NVDA and JAWS are probably just reading the valuetext, which is blank, and then not announcing anything.
  126. # [19:37] <clown> your screen shot looks a lot like mine on Fedora
  127. # [19:38] * clown congratulates *someone* for getting dom inspector to show a11y info on os x. don't know who to congratulate though...
  128. # [19:39] <@tbsaunde> clown: why? that part is just the same code as linux / windows
  129. # [19:39] <clown> scott_gonzalez: Chrome/Chromevox says "indeterminate" on OS X
  130. # [19:40] <clown> tbsaunde: there was a time when the dom inspector showed nothing when running FF on OS X. I don't know why not...
  131. # [19:40] <@tbsaunde> clown: for a while we didn't build a11y at all on firefox
  132. # [19:41] <@tbsaunde> now we build it, but only voice over and xpcom can turn it on
  133. # [19:41] <clown> "at all" == even for windows?
  134. # [19:41] <clown> aha! I have to actually start voiceover to make it kick in.
  135. # [19:41] <clown> that's useful info.
  136. # [19:41] <clown> thanks tbsaunde.
  137. # [19:42] <scott_gonzalez> Yeah, turning on VoiceOver changes the behavior or some stuff.
  138. # [19:42] <scott_gonzalez> Even native things like the behavior of tabs in system preferences.
  139. # [19:43] <@tbsaunde> clown: I meant at all on osx
  140. # [19:43] <@tbsaunde> clown: no, something like dom inspector should do it too
  141. # [19:43] <clown> scott_gonzalez: I may have been too quick with what I said about Chromvox. It might be just speaking a title or something when it says intedeterminate.
  142. # [19:44] <scott_gonzalez> clown: http://jsfiddle.net/xhXm3/1/show/
  143. # [19:44] <scott_gonzalez> That's a bare minimum test case.
  144. # [19:44] <scott_gonzalez> With a focusable progressbar.
  145. # [19:45] <clown> scott_gonzalez: that one speaks, "progress bar min zero"
  146. # [19:45] <clown> with chromevox
  147. # [19:45] <scott_gonzalez> Ok, so that's not too useful.
  148. # [19:49] <scott_gonzalez> clown: It seems like Firefox is exposing everything it should though, right?
  149. # [19:49] <scott_gonzalez> So this is a problem with the screen readers?
  150. # [19:50] <clown> scott_gonzalez: looks that way.
  151. # [19:50] <scott_gonzalez> Ok, thanks.
  152. # [19:50] <clown> I still haven't tried orca.
  153. # [19:51] * habber-away is now known as habber
  154. # [19:53] <clown> yeesh. orca is completely silent when on the progress bar.
  155. # [19:54] <scott_gonzalez> clown: Do you get anything different for http://jsfiddle.net/xhXm3/2/show/ ?
  156. # [19:54] * clown checking...
  157. # [19:54] <scott_gonzalez> Yeah, Orca is probably just reading valuetext, which is blank.
  158. # [19:55] <clown> scott_gonzalez: btw, safari+voiceover speaks "indeterminate progress bar" for the first jsfiddle link you gave above.
  159. # [19:56] <clown> whoops: "indeterminate progress indicator"
  160. # [19:57] <scott_gonzalez> clown: Yeah, VoiceOver gets it right in Firefox and Chrome.
  161. # [19:57] <scott_gonzalez> Using VoiceOver in Opera is just painful.
  162. # [19:57] <clown> scott_gonzalez: for the second example (http://jsfiddle.net/xhXm3/2/show/), chromevox says "progress bar"
  163. # [19:58] <scott_gonzalez> Ok, so ChromeVox is probably just announcing the individual properties.
  164. # [19:58] <scott_gonzalez> For some reason I can't get ChromeVox to install.
  165. # [19:58] <scott_gonzalez> Then again, my Chrome install is kinda messed up. I'm on 21 and it thinks its up-to-date.
  166. # [19:59] <clown> and, orca is again completely silent when focus goes to the progress bar.
  167. # [19:59] * clown phone ringing...
  168. # [20:13] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
  169. # [20:16] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
  170. # [20:40] * habber is now known as habber-away
  171. # [21:00] * khuey is now known as khuey|away
  172. # [21:04] <@firebot> josh@joshmatthews.net changed the Resolution on bug 755048 from --- to FIXED.
  173. # [21:04] <@firebot> josh@joshmatthews.net changed the Status on bug 755048 from NEW to RESOLVED.
  174. # [21:04] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=755048 nor, --, ---, josh, RESO FIXED, Fix build warnings under accessibility/
  175. # [21:28] * Joins: Stevef (chatzilla@moz-CD0F47B5.cable.virginmedia.com)
  176. # [21:35] <@davidb> Hixie: bug 820508 seems to be meandering off track
  177. # [21:35] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=820508 nor, --, ---, nobody, NEW, Add support for <main> element
  178. # [21:35] <@davidb> (at least to me)
  179. # [21:39] <@tbsaunde> davidb: which part?
  180. # [21:40] <@davidb> anything that seems more about people than about the bug
  181. # [21:40] * ehsan|away is now known as ehsan
  182. # [21:40] <@davidb> unless the people part is relevant, e.g. web devs
  183. # [21:44] <Hixie> davidb: what is the track?
  184. # [21:45] <@davidb> Hixie: about <main> and not about forecasting flaws in people not admitting mistakes in the future.
  185. # [21:45] <@davidb> I worry this will get heated again.
  186. # [21:45] <Hixie> davidb: yeah that was just an appendix to my comment, not on track at all
  187. # [21:45] <@davidb> yeah, the rest was good
  188. # [21:46] <Hixie> davidb: <main> hasn't been particualrly heated, compared to other discussions ;-)
  189. # [21:46] <@davidb> i wish bugzilla allowed an aux section or something
  190. # [21:46] <@davidb> heh
  191. # [21:47] <@tbsaunde> Hixie: I really liked the then I wouldn't have to bother with this part :)
  192. # [21:51] <Hixie> heh
  193. # [21:51] <Hixie> <main> is a classic example of something i think is dumb but that i really have no interest in
  194. # [21:51] <Hixie> it's not like those things that i have a passionate interest in stopping or adding
  195. # [21:52] <Hixie> where i am eager to keep fighting :-P
  196. # [21:52] <@davidb> Same. There are a lot of other things I give more attention than <main>.
  197. # [21:53] <@davidb> I don't really have an entrenched opinion but I am hopeful to see use cases outside a11y because that can really help it fly.
  198. # [21:53] <eeejay> <main/> makes sense to me, but i don't care enough, and nobody cares if i care :)
  199. # [21:54] <@davidb> Henri's bug summary makes sense to me.
  200. # [21:54] * khuey|away is now known as khuey
  201. # [21:58] <@davidb> eeejay: I care.
  202. # [21:58] <@davidb> but nobody cares if i care if you care
  203. # [21:58] <Stevef> comparing <main> to longdesc is FUD nothing more
  204. # [21:58] <Stevef> I wouldn't mind hixie continuing to argue if he actually brought something new to the table
  205. # [21:59] <@davidb> Hey now.
  206. # [21:59] <@tbsaunde> davidb: Henry's description doesn't make sense to me unless we're discussing whether its better to solve the wrong problem explicitly or implicitly
  207. # [21:59] <eeejay> in a previous incarnation of accessfu, i made it a convention if a scroll command is given, to either scroll the window, or scroll role="main.
  208. # [21:59] <eeejay> so, <main/> makes sens in the case
  209. # [22:01] <@davidb> it is hard to predict the future.
  210. # [22:01] <eeejay> so while scoobi do is good for capturing non-contiguous main content, it does not demarcate an authoritative block that is main
  211. # [22:01] <@davidb> i don't know that scooby doo is about non-contiguous
  212. # [22:02] <@davidb> aroo?
  213. # [22:02] <eeejay> davidb, it collects all containers that are not children of other known landmarks
  214. # [22:03] * @davidb nods
  215. # [22:03] <@davidb> ok my <main> attention is waning again
  216. # [22:03] <@tbsaunde> eeejay: but it is impossible to demercaate one block that is all and only all the main content
  217. # [22:04] <eeejay> i think the debate would be, could main content be considered as a single landmark, or is it defined by negation (ie. everything that is not already a landmark)
  218. # [22:04] <@tbsaunde> eeejay: I don't understand why you'd prefer to scroll to the start of the main content instead of changing views to a view of the website that only has main content and everything else stripped out
  219. # [22:04] * @davidb stifles asking why we have body tags… why not just assume not header
  220. # [22:04] * @davidb err head
  221. # [22:04] <eeejay> tbsaunde, that isn't the use case
  222. # [22:05] * @davidb is destined to see the same debates repeated 3+ times
  223. # [22:05] <@tbsaunde> davidb: other than that's not what body does because its already there and we can't remove it seems like a good enough reason
  224. # [22:05] <@tbsaunde> eeejay: why not?
  225. # [22:05] <@davidb> tbsaunde: there is a prime directive about not messing with web content
  226. # [22:06] <@davidb> tbsaunde: that is add-on territoryy
  227. # [22:06] <eeejay> tbsaunde, i thought you were referring to the accessfu use case i mentioned
  228. # [22:06] <@tbsaunde> eeejay: I ... sort of am
  229. # [22:06] <@davidb> tbsaunde: would you advocate against <body> today?
  230. # [22:06] * @davidb goes for water
  231. # [22:07] <@tbsaunde> davidb: maybe? I don't see why it can't all just live in <html>
  232. # [22:07] <@tbsaunde> eeejay: but if the use case for swiping to move to main content is so that you can read the main content then I don't know what it is
  233. # [22:07] <eeejay> tbsaunde, in the accessfu use case, a user does a "scroll" gesture. First we try to scroll the window, if there is content offscreen. If the window is not scrollable, we check to see if the main block is scrollable, and scroll that
  234. # [22:07] <eeejay> actually, we first check for the main block and only then window
  235. # [22:08] <@tbsaunde> eeejay: your describing an algorithm, not telling me about a use case
  236. # [22:08] <eeejay> tbsaunde, i'm telling you about an algorithm that needs to know what is the main block
  237. # [22:09] <eeejay> emphasis on "the", not a collection, but the single main block of a document
  238. # [22:09] <eeejay> so we were using role="main"
  239. # [22:10] <@tbsaunde> eeejay: first, there could easily be multiple ones
  240. # [22:11] <@tbsaunde> eeejay: also I still don't understand what problem you want to solve
  241. # [22:11] * Quits: Stevef (chatzilla@moz-CD0F47B5.cable.virginmedia.com) (Ping timeout)
  242. # [22:11] <eeejay> tbsaunde, imagine a document viewer app
  243. # [22:12] <@tbsaunde> eeejay: sure
  244. # [22:12] <eeejay> tbsaunde, the app itself is not scrollable, but the content in it has css overflow: scroll
  245. # [22:13] <@tbsaunde> davidb: as for messing with content we should introde into the documents sand box, but you don't need to do that to do what I'm suggesting, you just treat a bunch of stuff as if it was display: hidden
  246. # [22:13] <eeejay> tbsaunde, so a sighted user sees a "chrome", and an obvious scrollbar for the content embedded in the chrome
  247. # [22:13] <eeejay> but a screen reader, when it needs to scroll, needs to know what the focal point of the app is that needs to be scrolled
  248. # [22:13] <eeejay> ie. the main content
  249. # [22:14] <eeejay> in the doc view web app example it would be the embedded document
  250. # [22:14] <@tbsaunde> eeejay: needing to scroll isn't a use case, its a tool for acomplishing something
  251. # [22:15] <@tbsaunde> pressumably making it so you can read what your interested in
  252. # [22:15] <eeejay> tbsaunde, not sure what you mean when you say it isn't a use case
  253. # [22:17] <eeejay> tbsaunde, i agree that one use case is, remove all crap and only give me the main content to read. but that is not what i am talking about, i am giving another case
  254. # [22:17] <@tbsaunde> eeejay: do you decide "I'm a want to scroll something into view" or "I would like to read X therefore I'm going to scroll X into view so I can read X"
  255. # [22:18] <eeejay> tbsaunde, and about multiple main contents, aria defines the main role as "the author SHOULD mark no more than one element with the main role."
  256. # [22:18] <eeejay> tbsaunde, i am not talking about scroll into view
  257. # [22:18] <@tbsaunde> eeejay: I'm not talking about what the spec says, I can easily image documents that in reality should have two different sections of main content
  258. # [22:19] <eeejay> true
  259. # [22:19] <eeejay> so that doc would not use main at all, but article instead
  260. # [22:19] <@tbsaunde> eeejay: your use case far as I can make it out is that your trying to solve the I want to only read main content and nothing else use case accept that you've decided to solve it in a silly way
  261. # [22:20] <eeejay> tbsaunde, so you didn't understand what i said
  262. # [22:21] <@tbsaunde> eeejay: you are free to convince me of that :)
  263. # [22:21] <eeejay> tbsaunde, i am offering a totally different use case.. i don't actually care anymore since we already changed how this is done in accessfu
  264. # [22:22] <eeejay> but it is not unlikely that we will need it again
  265. # [22:23] <@tbsaunde> eeejay: ok, I'm still unconvinced there is a problem users acutally want to solve other than get rid of all the crap so I can read the main stuff but if you don't care about convincing me that's fine
  266. # [22:25] <eeejay> tbsaunde, yeah, this is not about reading. it is more about web apps then documents, and it is about determining the main "control"/"panel"/"content" of the app taht should receive global commands
  267. # [22:28] <@davidb> holy scrollback batman
  268. # [22:28] <@davidb> ok i gotta relocate...
  269. # [22:29] <@davidb> have a great weekend!
  270. # [22:29] <@davidb> hub: i'm coming over to give you a handshake
  271. # [22:29] <@tbsaunde> eeejay: I'm not really sure what your doing on a web page other than reading, but ok
  272. # [22:29] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
  273. # [22:40] * Parts: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
  274. # [23:11] * habber-away is now known as habber
  275. # [23:12] * Quits: victorporof (victorporo@87CAE4F1.BEB908B0.79933D60.IP) (Quit: victorporof)
  276. # [23:14] * Quits: @hub (hub@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
  277. # [23:19] * Joins: Stevef (chatzilla@moz-CD0F47B5.cable.virginmedia.com)
  278. # [23:26] * Quits: Wolf (chatzilla@moz-F8C1AF63.carolina.res.rr.com) (Connection reset by peer)
  279. # [23:26] * Joins: Wolf (chatzilla@moz-F8C1AF63.carolina.res.rr.com)
  280. # Session Close: Sat Dec 15 00:00:00 2012

The end :)