/irc-logs / mozilla / #accessibility / 2011-11-28 / end

Options:

  1. # Session Start: Mon Nov 28 00:00:00 2011
  2. # Session Ident: #accessibility
  3. # [01:55] * Quits: silvia (Adium@moz-B7928F7E.dyn.iinet.net.au) (Quit: Leaving.)
  4. # [02:04] * Joins: mdcurran (mick@6D209346.96206780.9ADEC69.IP)
  5. # [02:14] * eeejay is now known as eeejay_is_afk
  6. # [02:17] * Joins: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP)
  7. # [02:38] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  8. # [02:39] * Joins: tty234 (telex@moz-F9058B8A.net)
  9. # [03:01] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  10. # [03:01] * Quits: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP) (Input/output error)
  11. # [03:02] * Joins: tty234 (telex@moz-F9058B8A.net)
  12. # [03:24] * Joins: surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  13. # [03:24] * ChanServ sets mode: +o surkov
  14. # [03:52] <@firebot> surkov.alexander@gmail.com granted review for attachment 577065 on bug 705436.
  15. # [03:52] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=705436 nor, --, ---, Ms2ger, ASSI, Remove dead code: getDocumentTypeCB
  16. # [03:53] <@firebot> surkov.alexander@gmail.com granted review for attachment 577066 on bug 705437.
  17. # [03:53] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=705437 nor, --, ---, Ms2ger, ASSI, Remove dead code: nsCoreUtils::GetLastSensibleColumn
  18. # [04:06] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
  19. # [04:06] * ChanServ sets mode: +o jprmc
  20. # [04:16] * Quits: @surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Ping timeout)
  21. # [04:16] * Joins: surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  22. # [04:16] * ChanServ sets mode: +o surkov
  23. # [04:26] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  24. # [04:26] * Joins: tty234 (telex@moz-F9058B8A.net)
  25. # [04:36] * Quits: @surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Ping timeout)
  26. # [04:37] * Joins: surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  27. # [04:37] * ChanServ sets mode: +o surkov
  28. # [04:42] * Joins: surkov_ (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  29. # [04:42] * Quits: @surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Connection reset by peer)
  30. # [04:42] * surkov_ is now known as surkov
  31. # [04:44] * Joins: surkov_ (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  32. # [04:44] * Quits: surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Connection reset by peer)
  33. # [04:44] * surkov_ is now known as surkov
  34. # [04:46] * Quits: @davidb (davidb@31ABB91E.F7EBA23.C3C3A9D0.IP) (Quit: bye)
  35. # [06:18] * eeejay_is_afk is now known as eeejay
  36. # [06:47] * Quits: mdcurran (mick@6D209346.96206780.9ADEC69.IP) (Quit: leaving)
  37. # [06:47] * Joins: mdcurran (mick@6D209346.96206780.9ADEC69.IP)
  38. # [07:04] * Quits: mdcurran (mick@6D209346.96206780.9ADEC69.IP) (Connection reset by peer)
  39. # [07:11] * Joins: Jamie (jamie@moz-CA26021.jantrid.net)
  40. # [07:12] <Jamie> surkov: afternoon.
  41. # [07:12] <surkov> hello, Jamie!
  42. # [07:12] <Jamie> surkov: I'm not really sure how ot interpret the data in about:memory, but this does look odd to me: 1,085.67 MB (88.52%) -- heap-unclassified
  43. # [07:12] <Jamie> that's with two tabs open
  44. # [07:12] <Jamie> and one of those is about:memory :)
  45. # [07:13] <surkov> sounds like a problem, running GC and etc doesn't help, right?
  46. # [07:14] <Jamie> I just hit all of the buttons several times, and now I have 1,078.46 MB (90.16%) -- heap-unclassified
  47. # [07:15] <Jamie> surkov: the problem is that reproducing this is complicated and I don't know how ot reproduce it quickly; takes usage time
  48. # [07:15] <Jamie> there are also several variables at play, but I can't eliminate all of them.
  49. # [07:16] <Jamie> in short, I'm pretty sure it relates to twitter.com when used with my Greasemonkey script, which implements some a11y fixes. The script changes the aria role of some nodes and gives them a tabindex to make them focusable
  50. # [07:16] * Jamie isn't sure if the grasemonkey script is at fault, though, because i can't reproduce the issue without the a11y fixes
  51. # [07:16] <surkov> ok, Jamie, could you answer to email saying this, maybe David can cc someone who can give a better hint how to narrow down the problem
  52. # [07:16] <Jamie> i.e. I can't access it, so don't knwo what I'm doing
  53. # [07:17] <Jamie> sure
  54. # [07:17] <surkov> Jamie: it's quite possible that a11y impl is guilty because I don't recall anything similar when there's no a11y
  55. # [07:17] <Jamie> surkov: can you try something for me quickly with twitter.com?
  56. # [07:17] <surkov> a11y memory is not reported by about:memory still
  57. # [07:18] <surkov> Jamie: sure
  58. # [07:18] <Jamie> surkov: access a timeline, then press j lots of times; maybe just hold it down for a while :)
  59. # [07:18] <Jamie> that should move you through the tweets
  60. # [07:18] * Jamie isn't sure how it is indicated visually
  61. # [07:18] <surkov> Jamie: with or without NVDA running
  62. # [07:18] <surkov> ?
  63. # [07:18] <Jamie> in theory, that's all that's required to cause the leak
  64. # [07:18] <surkov> ok, I'll try
  65. # [07:18] <Jamie> surkov: I'm curious to know whether you get any leaking without, but then with as well
  66. # [07:19] <surkov> Jamie: what is 'j' supposed to do?
  67. # [07:19] <Jamie> surkov: moves to previous tweet
  68. # [07:19] <surkov> it's twitter shortcut, right?
  69. # [07:19] <Jamie> twitter sort of focuses it visually, even though it doesn't use real focus. (the gm script makes it real focus)
  70. # [07:19] <surkov> what Firefox do you use?
  71. # [07:19] <Jamie> yes
  72. # [07:20] <Jamie> Mozilla/5.0 (Windows NT 6.1; WOW64; rv:11.0a1) Gecko/20111116 Firefox/11.0a1
  73. # [07:20] <Jamie> hmm, that needs an update :)
  74. # [07:20] <Jamie> but apparently it also occurs in firefox 8
  75. # [07:32] <surkov> Jamie: heap-unclassified doesn't grow but total memory grows, after opening twitter, 'j' it and then close the tab
  76. # [07:33] <Jamie> so closing tab doesn't release memory for you?
  77. # [07:36] <surkov> not all memory
  78. # [07:38] <Jamie> any idea what heap-unclassified means?
  79. # [07:44] * Quits: JulienP (julien.pic@moz-DFEC0675.opera.com) (Ping timeout)
  80. # [07:46] * Joins: JulienP (julien.pic@moz-DFEC0675.opera.com)
  81. # [07:47] <surkov> Jamie: I think a11y memory should follow into this category since we're not integrated with about:memory stills and other random stuffs
  82. # [07:47] <surkov> after all operations task manager shows 300mb or memory, about:memory explicit allocation is 124mb
  83. # [07:48] <surkov> maybe this is memory defragmentation
  84. # [07:48] <Jamie> perhaps, though mine will get to 700+mb and then gb of memory in an hour or so
  85. # [07:48] <Jamie> surkov: was that with NVDa running or not running?
  86. # [07:49] <surkov> latest release
  87. # [07:49] <surkov> 2011.3
  88. # [07:49] * Jamie nods
  89. # [07:49] <surkov> do you have anything installed like extensions?
  90. # [07:49] <Jamie> I do have a few, including greasemonkey with ascript for twitter. as I say, it's useless to me without that gm script, so I can't tell whether I see the issue without it
  91. # [07:50] <Jamie> surkov: however, I can give you the URL to the gm script if you want to have a play
  92. # [07:50] <surkov> sure
  93. # [07:50] <surkov> because heap-unclassified is pretty low
  94. # [07:50] <Jamie> http://dl.dropbox.com/u/28976681/TwitterA11yFixes.user.js
  95. # [07:50] <surkov> is not too high when twitter is open and low when it's closed
  96. # [07:51] <surkov> Jamie: how to install it?
  97. # [07:51] <Jamie> surkov: install Greasemonkey (greasespot.net), then just point Firefox at that URL I gave above
  98. # [07:57] <surkov> Jamie: I don't see a difference with script installed
  99. # [08:00] <Jamie> weird
  100. # [08:00] <Jamie> surkov: as I say, you'll only get huge memory bloat with extended usage
  101. # [08:00] <Jamie> but ... hmm
  102. # [08:01] * Jamie is frustrated. Keep finding bugs that are difficult to reproduce on demand
  103. # [08:01] <surkov> I see memory grows in tasks manager, but nothing suspicious for unclassified heap
  104. # [08:02] * Jamie has 65 mb heap-unclassified atm
  105. # [08:02] <Jamie> and just restarted a little while ago
  106. # [08:05] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  107. # [08:05] * Joins: tty234 (telex@moz-F9058B8A.net)
  108. # [08:19] <surkov> Jamie: that could be a11y and could not mean a leak
  109. # [08:44] * Quits: @firebot (firebot@moz-67BF9FE6.carolina.res.rr.com) (Ping timeout)
  110. # [09:09] * Quits: Jamie (jamie@moz-CA26021.jantrid.net) (Quit: leaving)
  111. # [09:17] * Joins: firebot (firebot@moz-67BF9FE6.carolina.res.rr.com)
  112. # [09:17] * ChanServ sets mode: +o firebot
  113. # [09:47] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  114. # [09:47] * Joins: tty234 (telex@moz-F9058B8A.net)
  115. # [10:45] <@firebot> surkov.alexander@gmail.com requested superreview from neil@httl.net for attachment 576519 on bug 704852.
  116. # [10:45] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=704852 nor, --, ---, surkov.alexander, ASSI, Calculate compatibility mode when accessibility starts
  117. # [11:24] <@firebot> atulagrwl@gmail.com requested review from jst@mozilla.org for attachment 577211 on bug 682770.
  118. # [11:24] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=682770 nor, --, ---, atulagrwl, NEW, nsIAccessibleSelectable::GetSelectionCount should be unsigned long
  119. # [11:37] * Joins: Flyingmana (Mana@moz-978A58E4.adsl.alicedsl.de)
  120. # [12:12] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 577212 on bug 699017.
  121. # [12:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=699017 nor, --, ---, surkov.alexander, ASSI, aria-required attribute on file input not read by JAWS
  122. # [12:13] <@firebot> surkov.alexander@gmail.com requested review from roc@ocallahan.org for attachment 577212 on bug 699017.
  123. # [12:38] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 577213 on bug 703202.
  124. # [12:38] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=703202 nor, --, ---, surkov.alexander, ASSI, ARIA comboboxes don't fire value change events
  125. # [12:41] <@firebot> surkov.alexander@gmail.com changed the Component on bug 693948 from Disability Access to Developer Tools: Console.
  126. # [12:41] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=693948 nor, --, Firefox 10, trev.saunders, NEW, expose layout-guess: true object attribute on CSS table accessible
  127. # [13:02] * khuey|away is now known as khuey
  128. # [14:40] * Quits: victorporof (victorporo@95A92D4E.9F23B8E6.79933D60.IP) (Quit: victorporof)
  129. # [14:45] <@firebot> surkov.alexander@gmail.com requested review from bolterbugz@gmail.co m for attachment 577231 on bug 689847.
  130. # [14:46] <@firebot> surkov.alexander@gmail.com changed the Status on bug 689847 from NEW to ASSIGNED.
  131. # [14:46] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=689847 nor, --, ---, surkov.alexander, ASSI, Expose active state on current item of selectable widgets
  132. # [14:51] * Joins: surkov_ (surkov@464FB5EC.C2B307F0.34044A7F.IP)
  133. # [14:51] * Quits: surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Ping timeout)
  134. # [14:51] * surkov_ is now known as surkov
  135. # [14:52] * Joins: MarcoZ (marco.zehe@moz-703E4635.dip.t-dialin.net)
  136. # [14:52] * ChanServ sets mode: +o MarcoZ
  137. # [15:06] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
  138. # [15:06] * ChanServ sets mode: +qo davidb davidb
  139. # [15:06] <@davidb> hi all!
  140. # [15:08] <@tbsaunde> hi davidb
  141. # [15:08] <@davidb> hi tbsaunde, how's skule?
  142. # [15:09] <@tbsaunde> davidb: ok, be nice when its done, or atleast break
  143. # [15:10] <@davidb> yeah
  144. # [15:12] <@firebot> surkov.alexander@gmail.com cancelled review?(bolterbugz@gmail.co m) for attachment 577231 on bug 689847.
  145. # [15:12] <@firebot> surkov.alexander@gmail.com requested review from the wind for attachment 577236 on bug 689847.
  146. # [15:13] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=689847 nor, --, ---, surkov.alexander, ASSI, Expose active state on current item of selectable widgets
  147. # [15:13] <@firebot> surkov.alexander@gmail.com requested review from bolterbugz@gmail.co m for attachment 577236 on bug 689847.
  148. # [15:13] <@davidb> there are some gems here http://www.dodgycoder.net/2011/11/yoda-conditions-pokemon-exception.html
  149. # [15:15] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
  150. # [15:15] <@firebot> neil@httl.net denied superreview for attachment 576519 on bug 704852.
  151. # [15:15] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=704852 nor, --, ---, surkov.alexander, ASSI, Calculate compatibility mode when accessibility starts
  152. # [15:16] <@MarcoZ> Hi all!
  153. # [15:16] <@davidb> heyo
  154. # [15:16] <clown> yo
  155. # [15:21] <@firebot> bolterbugz@gmail.com granted review for attachment 577236 on bug 689847.
  156. # [15:22] <@davidb> clown: have you thought about ARIA extensibility lately?
  157. # [15:22] <clown> davidb, no.
  158. # [15:22] <@davidb> fair enuff
  159. # [15:23] <clown> have you?
  160. # [15:24] <@davidb> clown: a little
  161. # [15:24] <@davidb> it is something i come full circle on occasionally
  162. # [15:25] <@davidb> but this time I hope to cause some change
  163. # [15:25] <clown> have you filed issues against aria 1.1 or aria 2.0?
  164. # [15:26] <clown> or is not steeped enough?
  165. # [15:27] <@davidb> clown: where are aria 2.0 issues happening?
  166. # [15:27] <clown> in the usual place ;-)
  167. # [15:27] <clown> the PF/W3 issue tracker.
  168. # [15:27] <@davidb> yeah
  169. # [15:27] <@davidb> figures
  170. # [15:28] * @davidb tires of w3c
  171. # [15:30] <@firebot> bolterbugz@gmail.com cancelled review?(bolterbugz@gmail.co m) for attachment 576519 on bug 704852.
  172. # [15:30] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=704852 nor, --, ---, surkov.alexander, ASSI, Calculate compatibility mode when accessibility starts
  173. # [15:32] <@davidb> oh hi surkov
  174. # [15:32] <surkov> hi, davidb
  175. # [15:33] * davidb sets mode: +o surkov
  176. # [15:33] <@firebot> marco.zehe@googlemail.com granted review for attachment 577212 on bug 699017.
  177. # [15:33] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=699017 nor, --, ---, surkov.alexander, ASSI, aria-required attribute on file input not read by JAWS
  178. # [15:38] <@davidb> surkov: I think IsVisibleConsideringAncestors is supposedly expensive, which led to the original design (re: bug 591363)
  179. # [15:38] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=591363 nor, --, ---, bolterbugz, NEW, (in)visible state is not always correct?
  180. # [15:39] <@surkov> davidb: and?
  181. # [15:39] <@davidb> surkov: if i understand your suggestion, you are suggesting to call that function every time
  182. # [15:40] <@surkov> davidb: don't you do that without my suggestion?
  183. # [15:40] <@davidb> surkov: there are early bailouts
  184. # [15:40] <@davidb> if i understand you, you want me to move those after that expensive call?
  185. # [15:40] <@davidb> (inside the block)
  186. # [15:40] <@surkov> davidb: for text nodes without text which doesn't happen normally and for inlines where we call expensive GetBounds()?
  187. # [15:41] <@davidb> surkov: right
  188. # [15:41] <@davidb> i'm not sure which is better
  189. # [15:42] <@davidb> surkov: anyways, am I understanding your suggestion correctly?
  190. # [15:42] <@surkov> I think yes
  191. # [15:42] <@davidb> ok, let's try it.
  192. # [15:43] <@surkov> I think current algorithm hits IsVisibleConsideringAncestors 90% of time, for others 10% (inlines with empty rect) it hits expensive GetBounds()
  193. # [15:44] <@surkov> and currently it hits expensive GetBounds and IsVisibleConsideringAncestors both for all inlines
  194. # [15:45] <@davidb> i have the feeling IsVisibleConsideringAncestors is most expensive
  195. # [15:46] <@davidb> surkov: if I was writing this method from scratch I would probably have just used style visibility :)
  196. # [15:47] <@davidb> we take 'visible' very literally.
  197. # [15:51] * bwinton_away is now known as bwinton
  198. # [15:54] <@davidb> the specs are a mess here.
  199. # [15:56] <@davidb> surkov: NVDA doesn't like us exposing background tab content as invisible
  200. # [15:57] <@surkov> davidb: yes, about mess spec, see a comment, we decided to not follow MSAA
  201. # [15:57] <@davidb> sort of
  202. # [15:58] <@davidb> there are 26 comments BTW :)
  203. # [15:58] <@davidb> in this bug (there are related bugs)
  204. # [16:07] * clown is now known as clown_mtg
  205. # [16:15] * khuey is now known as khuey|away
  206. # [16:43] * khuey|away is now known as khuey
  207. # [16:47] <@firebot> marco.zehe@googlemail.com changed the Status on bug 704416 from RESOLVED to VERIFIED.
  208. # [16:47] <@firebot> marco.zehe@googlemail.com granted in-testsuite on bug 704416.
  209. # [16:47] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=704416 nor, --, mozilla11, surkov.alexander, VERI FIXED, HTML acronym and abbr names should be provided by @title
  210. # [16:49] <@firebot> marco.zehe@googlemail.com changed the Status on bug 704754 from RESOLVED to VERIFIED.
  211. # [16:49] <@firebot> marco.zehe@googlemail.com granted in-testsuite on bug 704754.
  212. # [16:49] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=704754 nor, --, mozilla11, surkov.alexander, VERI FIXED, expose abbr object attribute on header cells
  213. # [16:50] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
  214. # [16:53] <@firebot> marco.zehe@googlemail.com changed the Status on bug 287740 from RESOLVED to VERIFIED.
  215. # [16:53] <@firebot> marco.zehe@googlemail.com granted in-testsuite on bug 287740.
  216. # [16:53] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=287740 nor, --, mozilla11, surkov.alexander, RESO FIXED, Expose advanced table semantics
  217. # [17:00] * Joins: victorporof (victorporo@95A92D4E.9F23B8E6.79933D60.IP)
  218. # [17:00] * Joins: nhirata (nhirata.bu@moz-2A9C9106.hsd1.ca.comcast.net)
  219. # [17:25] * Joins: aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com)
  220. # [17:25] * ChanServ sets mode: +o aaronlev
  221. # [17:26] * Quits: nhirata (nhirata.bu@moz-2A9C9106.hsd1.ca.comcast.net) (Quit: nhirata)
  222. # [17:53] * eeejay is now known as eeejay_is_afk
  223. # [17:55] * Joins: Flyingmana_ (Mana@moz-821D4244.adsl.alicedsl.de)
  224. # [17:56] * Quits: Flyingmana (Mana@moz-978A58E4.adsl.alicedsl.de) (Ping timeout)
  225. # [18:02] <@davidb> clown_mtg: how is/was meeting?
  226. # [18:02] * eeejay_is_afk is now known as eeejay
  227. # [18:06] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
  228. # [18:10] * @eeejay waves with vigor
  229. # [18:16] <@davidb> eeejay: heyo
  230. # [18:17] <@eeejay> hey davidb
  231. # [18:17] <@davidb> eeejay: i didn't look at your patches yet, but let me know when/if i should.
  232. # [18:17] <@davidb> eeejay: also, today is a special day :)
  233. # [18:18] <@davidb> congrats
  234. # [18:18] <@eeejay> davidb, is that you volunteering to review, cool!
  235. # [18:18] <@eeejay> davidb, 'tis!
  236. # [18:18] <@eeejay> weird, it doesn't feel different
  237. # [18:19] <@davidb> heh
  238. # [18:19] <@davidb> good
  239. # [18:19] <@eeejay> davidb, i feel like surkov and i are in a circular discussion.
  240. # [18:20] <@eeejay> i feel pretty good about the current API proposal
  241. # [18:21] <@davidb> i have had similar discussions but they always end well
  242. # [18:22] <@davidb> the trick is to avoid too many rotations somehow
  243. # [18:22] <@eeejay> too late :)
  244. # [18:22] <@eeejay> i'm glad we threw out the previous proposals.
  245. # [18:23] <@davidb> good
  246. # [18:23] <@eeejay> this proposal still feels good, it survived the weekend test
  247. # [18:23] <@davidb> haha
  248. # [18:23] <@surkov> eeejay: sorry for that, I just need to make sure the way we take is nice and the best, at least we should try to make it that way
  249. # [18:23] <@eeejay> surkov, yeah. i think we could afford to get some stuff wrong
  250. # [18:24] <@surkov> yes but not on propose
  251. # [18:24] <@eeejay> surkov, since it will be mostly "internal" use. when it is tested and good we could think of making this a more public feature
  252. # [18:24] <@davidb> don't make me quote Voltaire
  253. # [18:24] <@davidb> :)
  254. # [18:25] <@eeejay> hah
  255. # [18:25] <@davidb> http://en.wikipedia.org/wiki/Perfect_is_the_enemy_of_good
  256. # [18:25] <@davidb> But I'll accept perfection :)
  257. # [18:25] <@surkov> the current version is not suitable for public I think
  258. # [18:25] <@surkov> davidb: don't push me :)
  259. # [18:26] <@davidb> heheh
  260. # [18:26] <@eeejay> surkov, what is missing?
  261. # [18:26] <@eeejay> i am itching to move on to the rest of our mobile "solution"
  262. # [18:26] * @davidb bows and exits backwards… lunch!
  263. # [18:26] <@surkov> mapping vc to existing APIs, btw that was in previous versions
  264. # [18:27] <@eeejay> surkov, i got rid of it, because i thought you didn't like it
  265. # [18:27] <@surkov> eeejay: I have a feeling you treat me wrong, I said I don't like that for pivots, but that's right for vc
  266. # [18:28] <@eeejay> surkov, sorry if i treat you wrong!
  267. # [18:28] <@surkov> that's right for vs at least because no way to map observer stuffs to things like MSAA
  268. # [18:28] <@eeejay> surkov, i don't always 100% understand, that is true.
  269. # [18:29] <@surkov> vs -> vc
  270. # [18:29] <@eeejay> surkov, i agree
  271. # [18:29] <@surkov> that's my english I guess
  272. # [18:29] <@eeejay> surkov, vc changes should fire an a11y event
  273. # [18:29] <@surkov> yes
  274. # [18:30] <@surkov> basically I want to see vc which implementation is based on nice pivots concept
  275. # [18:30] <@surkov> if you want me then I'll try to put my wishes into API tomorrow
  276. # [18:30] <@surkov> wishes -> concerns
  277. # [18:31] <@surkov> I still don't have clever idea how it should look in the end
  278. # [18:31] <@eeejay> surkov, ok. but one thing:
  279. # [18:31] <@eeejay> i don't think vc should have any traversal functionality
  280. # [18:32] <@surkov> you introduce cool and nice feature that's likely is going to be reused as part of AT APIs specs, therefore I don't consider this as pure internal thing and want to make sure it's flexible
  281. # [18:32] <@surkov> why?
  282. # [18:33] <@eeejay> surkov, m/v/c
  283. # [18:33] <@surkov> what does it mean? sorry
  284. # [18:34] <@eeejay> surkov, i think the vc should only deal with state (model), the traversal logic belongs in the controller
  285. # [18:35] <clown_mtg> davidb: re: how is/was meeting — accomplished a few things.
  286. # [18:35] <@surkov> yep, but you control the state (accessible point or range) so why don't you want to add some handy to do that?
  287. # [18:35] * clown_mtg is now known as clown
  288. # [18:35] <@eeejay> surkov, you convinced me that is best (maybe by mistake)
  289. # [18:35] <@surkov> I think I talked about implementation
  290. # [18:35] <@surkov> to keep things separately
  291. # [18:36] <clown> so, no gantt charts this time, davidb ...
  292. # [18:36] <@eeejay> surkov, for example, you brought up the case of multiple ATs
  293. # [18:36] <@eeejay> surkov, each AT has a different assumption of the vc's behavior, and each tries to set it to a new behavior
  294. # [18:37] <@eeejay> solution: vc traversal behavior is not stored in the vc!
  295. # [18:37] <@surkov> but that could happen without traversal logic, for example, when different ATs set vc on different accessibles
  296. # [18:37] <@eeejay> surkov, so besides implementation, i think the API could be separate too
  297. # [18:38] <@surkov> ok, that doesn't look bad
  298. # [18:38] <@eeejay> surkov, i am assuming in the multi-AT case that the user input is what changes the vc, and that the user is aware as to what input should do what
  299. # [18:39] <@eeejay> surkov, i already have a working example
  300. # [18:39] * @eeejay needs to push it to github..
  301. # [18:39] <@surkov> ok, makes sense
  302. # [18:40] * @eeejay is looking for a url...
  303. # [18:41] <@eeejay> surkov, https://github.com/eeejay/Soft-Focus-Demo/blob/master/content/SoftFocusController.js#L138
  304. # [18:46] <@surkov> I see
  305. # [18:46] <@eeejay> surkov, the other advantage of keeping traversal and vc seperate is more flexibility for the future
  306. # [18:46] <@surkov> eeejay: but why do you want to deal with ranges?
  307. # [18:47] <@eeejay> surkov, ranges? where?
  308. # [18:47] <@surkov> startOffset/endOffset
  309. # [18:48] <@eeejay> surkov, for line/word/character navigation
  310. # [18:48] <@surkov> eeejay: I thought about vc states managed by AT, so you set a current state (that could be a pivot or trewalker for example) and call next/prev on vc directly
  311. # [18:48] <@surkov> if another AT wants to use vc then it sets different state
  312. # [18:48] <@surkov> eeejay: why DOM selection doesn't suite this?
  313. # [18:49] <@surkov> you can do line/word and etc navigation by point (single offset)
  314. # [18:49] <@surkov> so why do you need a range?
  315. # [18:50] <@eeejay> surkov, the point would be the offset for the beginning for the word/line?
  316. # [18:50] <@surkov> yep, for example
  317. # [18:51] <@surkov> you set an accessible and offset so you have a point
  318. # [18:51] <@surkov> and move it
  319. # [18:51] <@surkov> by words, lines or whatever you want
  320. # [18:52] <@eeejay> surkov, but what would that offset indicate? does the screen reader speak the next word? character? from that point to the end of the block?
  321. # [18:53] <@firebot> New Firefox - Disability Access bug 705744 filed by bugzilla@reqr.net.
  322. # [18:53] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=705744 nor, --, ---, nobody, UNCO, Pop-up window closes on Accessibility API call
  323. # [18:54] <@surkov> usually screen reader knows what to talk, maybe not passive one like android talkback
  324. # [18:54] <@eeejay> right
  325. # [18:54] <@eeejay> surkov, i think this is how vc is different from a conventional caret
  326. # [18:55] <@surkov> ok, so vc needs a selection
  327. # [18:55] <@eeejay> basically :)
  328. # [18:56] <@surkov> should traversal API include capabilities to move by words and etc?
  329. # [18:57] <@eeejay> surkov, i want to add to the pivot interface a helper method that gets the bounds of the current accessible/range
  330. # [18:57] <@eeejay> good q
  331. # [18:57] <@eeejay> question
  332. # [18:57] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  333. # [18:58] <@eeejay> surkov, that would actually work well in the pivot API
  334. # [18:58] <@eeejay> surkov, nextWord(), nextLine()
  335. # [18:58] <@surkov> yes, if we want to add some sugar
  336. # [18:59] * @tbsaunde tries to stay far away
  337. # [19:00] <@eeejay> haha
  338. # [19:01] <@tbsaunde> eeejay: I really hope the way you implement the focus ring was just a quick hack not the way you actually intend to do that
  339. # [19:01] <@eeejay> tbsaunde, in that extension? yeah
  340. # [19:02] * bwinton is now known as bwinton_away
  341. # [19:02] <@eeejay> tbsaunde, what is your concern?
  342. # [19:03] <@tbsaunde> eeejay: your mutating the DOM that belongs to the web page in ways they have no reason to expect
  343. # [19:03] <@eeejay> tbsaunde, it is not the content dom
  344. # [19:03] <@eeejay> tbsaunde, it is in the xul
  345. # [19:04] <@eeejay> tbsaunde, i toyed with the idea of doing it with a canvas element because of its inaccessible properties
  346. # [19:04] <@tbsaunde> (I'm also remembering a discussion with sicking and davidb about how chrome vox worked and disaproving of doing stuff in the web pages DOM / js context
  347. # [19:04] <@tbsaunde> eeejay: oh, huh
  348. # [19:04] * bwinton_away is now known as bwinton
  349. # [19:04] * khuey is now known as khuey|away
  350. # [19:04] <@eeejay> tbsaunde, i know. that is why it is not in the dom
  351. # [19:05] <@tbsaunde> sorry, I didn't understand what was going on there
  352. # [19:05] <@tbsaunde> btw what is the point of let in js? it seems like it does the same thing as var foo = blah
  353. # [19:07] <@eeejay> tbsaunde, let actually does block scopes correctly
  354. # [19:08] * Quits: @surkov (surkov@464FB5EC.C2B307F0.34044A7F.IP) (Quit: surkov)
  355. # [19:10] <@tbsaunde> eeejay: correctly?
  356. # [19:10] <@tbsaunde> what does var do? O.O
  357. # [19:13] <@eeejay> tbsaunde, it is a lot of nuance :P
  358. # [19:13] <@eeejay> tbsaunde, https://developer.mozilla.org/en/New_in_JavaScript_1.7#Block_scope_with_let_%28Merge_into_let_Statement%29
  359. # [19:14] <@eeejay> basically you are guaranteed not to clobber any same variable names outside the block
  360. # [19:24] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
  361. # [19:24] * ChanServ sets mode: +o hub
  362. # [19:24] <@tbsaunde> eeejay: ugh, now I remember why I avoid js
  363. # [19:33] * Quits: @MarcoZ (marco.zehe@moz-703E4635.dip.t-dialin.net) (Quit: l8er)
  364. # [19:40] * @davidb is now powered by shitake udon!
  365. # [19:41] <@davidb> accessibility API could be really interesting in a js-only world
  366. # [19:45] <@tbsaunde> davidb: ?
  367. # [19:46] <@davidb> tbsaunde: basically taking advantage of js feature detection pattern.
  368. # [19:46] <@davidb> you can extend roles and states with wonton abandon… which is good and bad.
  369. # [19:48] <@tbsaunde> davidb: I think I'd call that a mess waiting to happen
  370. # [19:48] <@davidb> yeah, there are ways to avoid mess though, as you can imagine… this is a problem that has needed solving.
  371. # [19:48] <@davidb> one is heavy automated tests
  372. # [19:49] <@tbsaunde> I'm not sure I've drunk a lot of there coolade, but I'm not a big fan of dynamic typing
  373. # [19:49] <@davidb> I like it.
  374. # [19:49] <@davidb> I also like the compiler to catch errors though.
  375. # [19:50] <@tbsaunde> iono I tend to find code with explicit types easier to understand
  376. # [19:50] <@davidb> you probably learned static typing first right?
  377. # [19:51] * Joins: tty234 (telex@moz-F9058B8A.net)
  378. # [19:51] <@tbsaunde> davidb: true
  379. # [19:51] <@davidb> the problem with drinking the dynamic typing koolaid is that the flavour keeps changing
  380. # [19:52] * khuey|away is now known as khuey
  381. # [19:55] <@davidb> ok, yeah that was weak.
  382. # [20:04] <@eeejay> aaronlev, is it true that there is discussion about cursor features in aria 2?
  383. # [20:04] <@aaronlev> it was brought up
  384. # [20:04] <@aaronlev> did you see the google doc ?
  385. # [20:04] <@eeejay> aaronlev, any thread somewhere?
  386. # [20:04] <@eeejay> nope
  387. # [20:04] <@eeejay> please share!
  388. # [20:04] <@aaronlev> i innocently started a doc and got more contributions than i expected
  389. # [20:05] * Quits: tty234 (telex@moz-F9058B8A.net) (Ping timeout)
  390. # [20:05] <@aaronlev> dominic mazzoni from google was suggesting it
  391. # [20:05] <@aaronlev> and raman from google countered :)
  392. # [20:05] <@eeejay> smart guy
  393. # [20:05] <@eeejay> gal
  394. # [20:05] <@eeejay> that means it must be a killer idea
  395. # [20:05] <@aaronlev> i'm shopping for a laptop for a killer developer machine
  396. # [20:06] <@aaronlev> here: https://docs.google.com/document/d/1xMMl7WazqYauAueKnSX1uGWsSiLugBj9ML8oPdEg5h4/edit?hl=en_US
  397. # [20:07] <@eeejay> i "requested access", hope you are on the receiving end of that.
  398. # [20:07] <@eeejay> aaronlev, does it need to be mac?
  399. # [20:07] <@aaronlev> bo
  400. # [20:07] <@aaronlev> no
  401. # [20:08] <@aaronlev> i was going to get a pc and do mac development on a separate machine
  402. # [20:08] <@eeejay> aaronlev, i enjoy the thinkpad x220 muchly for development
  403. # [20:08] <@aaronlev> didn't see a powerful enough mac
  404. # [20:08] <@eeejay> fast and portable
  405. # [20:08] <@eeejay> oh
  406. # [20:08] <@eeejay> nm then
  407. # [20:08] <@aaronlev> i'm looking for quad core 16 gb ram with ssd
  408. # [20:08] <@eeejay> yeah, good luck with that
  409. # [20:09] * @eeejay blasts air mozilla in full volume
  410. # [20:09] <@hub> we have it here in the conference room
  411. # [20:10] <@davidb> aaronlev: possible alienware?
  412. # [20:10] <@aaronlev> davidb: between alienware and asus g74
  413. # [20:10] <@davidb> aaronlev: I have a windows beast under my desk and just VPN + RDP from my macbook air.
  414. # [20:10] <@aaronlev> davidb: alienware seems to often get bad online reviews
  415. # [20:11] * @davidb shrugs
  416. # [20:11] <@davidb> they've aligned with Dell a bit.
  417. # [20:11] * @davidb is also in the meeting
  418. # [20:11] <@davidb> my left arm is in the video
  419. # [20:11] <@davidb> exciting!
  420. # [20:12] <@hub> you don't see us because we use Air
  421. # [20:12] <@davidb> for now
  422. # [20:12] <@davidb> just wait until the new digs are done!
  423. # [20:12] <@hub> :-D
  424. # [20:30] <@aaronlev> Ok, I ordered the "laptop"
  425. # [20:31] <@eeejay> aaronlev, which "laptop"
  426. # [20:31] <@eeejay> ?
  427. # [20:32] <@aaronlev> http://www.amazon.com/gp/product/B005PAJICG/ref=ox_ya_os_product
  428. # [20:32] <@aaronlev> 2.2 GHz Intel
  429. # [20:32] <@aaronlev> 16 GB DIMM
  430. # [20:32] <@aaronlev> 750GB 7200RPM HDD + 160GB SSD Hard Drive
  431. # [20:32] <@aaronlev> 17.3-Inch Screen, Nvidia GTX 560M 3GB GDDR5
  432. # [20:32] <@aaronlev> Windows 7 Home Premium 64-bit
  433. # [20:32] <@hub> that's not a laptop. that's a luggageable
  434. # [20:32] <@davidb> sweet
  435. # [20:32] <@aaronlev> quad core
  436. # [20:32] <@aaronlev> yeah
  437. # [20:33] <@eeejay> holy mackerel
  438. # [20:33] <@aaronlev> should be able to do the job
  439. # [20:34] <@eeejay> yep
  440. # [20:34] <@aaronlev> comes next day air for $4.00 with amazon prime trial
  441. # [20:34] * Parts: @aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com)
  442. # [20:34] * Joins: aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com)
  443. # [20:34] * ChanServ sets mode: +o aaronlev
  444. # [20:34] <@davidb> wish i could use amazon prime
  445. # [20:35] <@eeejay> they should throw in a toy wagon so you could tow it around
  446. # [20:35] <@davidb> doesn't it have a jet on it?
  447. # [20:35] <@davidb> you can fly it
  448. # [20:35] <@hub> davidb: yeah but given that the Govt barred Amazon from opening shop in Canada....
  449. # [20:35] <@davidb> hub: exactly
  450. # [20:35] <@hub> as retarded as the telco regulation on foreign investments
  451. # [20:37] <@aaronlev> wait, this is core two duo
  452. # [20:37] <@aaronlev> doh
  453. # [20:37] <khuey> you could always join us
  454. # [20:37] <khuey> america could use a few more states
  455. # [20:37] <@aaronlev> hunh
  456. # [20:37] <@aaronlev> Processor: 2.2 GHz Intel Core Duo
  457. # [20:37] <@aaronlev> Number of Processors: 4
  458. # [20:37] <khuey> and we'll give you a completely different set of political problems to deal with!
  459. # [20:37] <@davidb> khuey: watch it… you don't want us to invade you ;)
  460. # [20:38] <khuey> I read somewhere that until after world war two pretty much the only military planning canada did was on how to resist an american invasion long enough for the british to show up
  461. # [20:38] <@tbsaunde> khuey: lol
  462. # [20:39] <@hub> khuey: I'd return to Sarkostan if that's the case.
  463. # [20:39] <@davidb> jolly good plan that
  464. # [20:40] <@tbsaunde> sarkostan?
  465. # [20:40] <@davidb> is that like sarcastistan?
  466. # [20:41] <@hub> Sarkozy's republic
  467. # [20:42] <@hub> also known as France
  468. # [20:42] <khuey> sarcastistan sounds like my kind of country
  469. # [20:42] <@hub> khuey: before Obama I was calling is Bushistan
  470. # [20:42] <@hub> s/is/it/
  471. # [20:43] <khuey> heh
  472. # [20:44] <@hub> I don't know how to call it today given the little power Obama has
  473. # [20:45] * khuey shrugs
  474. # [20:45] <khuey> the USA is pretty much ungovernable at this point
  475. # [20:45] <khuey> if the Democrats didn't just roll over when the Republicans are in power it would be completely ungovernable
  476. # [20:53] * khuey is now known as khuey|away
  477. # [21:06] * Joins: tty234 (telex@moz-F9058B8A.net)
  478. # [21:13] * khuey|away is now known as khuey
  479. # [21:22] * Quits: victorporof (victorporo@95A92D4E.9F23B8E6.79933D60.IP) (Quit: victorporof)
  480. # [21:39] * eeejay is now known as eeejay_is_afk
  481. # [22:00] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Quit: nhirata)
  482. # [22:11] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
  483. # [22:18] <@firebot> clegnitto@mozilla.com granted approval-mozilla-aurora for attachment 575834 on bug 701669.
  484. # [22:18] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=701669 cri, --, mozilla11, surkov.alexander, RESO FIXED, crash when changing profile in profile manager
  485. # [22:18] <@firebot> clegnitto@mozilla.com set status-firefox10 to affected on bug 701669.
  486. # [22:23] <@firebot> bolterbugz@gmail.com cancelled review?(surkov.alexander@gm ail.com) for attachment 576985 on bug 591363.
  487. # [22:23] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=591363 nor, --, ---, bolterbugz, NEW, (in)visible state is not always correct?
  488. # [22:28] <@firebot> clegnitto@mozilla.com granted approval-mozilla-aurora for attachment 576870 on bug 703198.
  489. # [22:28] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=703198 nor, --, mozilla11, surkov.alexander, RESO FIXED, JAWS doesn't announce combobox navigation in collapsed combobox
  490. # [22:32] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
  491. # [22:33] * Quits: @aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com) (Quit: aaronlev)
  492. # [22:34] * bwinton is now known as bwinton_away
  493. # [22:41] * Joins: aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com)
  494. # [22:41] * ChanServ sets mode: +o aaronlev
  495. # [22:43] <@firebot> hub@mozilla.com cancelled review?(hwaara@gmail.com) for attachment 338789 on bug 455443.
  496. # [22:43] <@firebot> hub@mozilla.com requested review from the wind for attachment 577354 on bug 455443.
  497. # [22:43] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=455443 nor, --, ---, surkov.alexander, ASSI, cache the parent for the accessibilityAttributeValue(NSAccessibilityParentAttribute)
  498. # [22:56] * Quits: @aaronlev (aaronlev@moz-CDA191A6.c3-0.arl-ubr1.sbo-arl.ma.cable.rcn.com) (Quit: aaronlev)
  499. # [23:04] * Parts: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
  500. # [23:24] * Quits: Flyingmana_ (Mana@moz-821D4244.adsl.alicedsl.de) (Quit: Verlassend)
  501. # [23:27] * eeejay_is_afk is now known as eeejay
  502. # [23:29] * bwinton_away is now known as bwinton
  503. # [23:29] * Joins: Jamie (jamie@moz-CA26021.jantrid.net)
  504. # [23:57] * bwinton is now known as bwinton_away
  505. # Session Close: Tue Nov 29 00:00:01 2011

The end :)