/irc-logs / mozilla / #accessibility / 2013-07-18 / end
Options:
- # Session Start: Thu Jul 18 00:00:00 2013
- # Session Ident: #accessibility
- # [00:03] <ialagenchev> The bug is about the tab separators being too faint and that might be an issue with people with impaired vision. They wouldn't be able to navigate their tabs
- # [00:12] * Joins: maxli (maxli@B5559308.9C54E60.2FA4EA88.IP)
- # [00:14] * Quits: maxli (maxli@B5559308.9C54E60.2FA4EA88.IP) (Ping timeout)
- # [00:17] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [00:18] <@firebot> eitan@monotonous.org granted review for attachment 777342 on bug 890940.
- # [00:18] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=890940 nor, --, mozilla25, maxli, REOP, [AccessFu] Visual bounds box is very wrong
- # [00:39] * Joins: maxli (maxli@B5559308.9C54E60.2FA4EA88.IP)
- # [00:44] * Quits: Mic (Mic@moz-92BAEE73.superkabel.de) (Quit: Instantbird -- http://www.instantbird.com)
- # [01:03] * Joins: hub (hub@moz-E12780DE.cable.teksavvy.com)
- # [01:03] * ChanServ sets mode: +o hub
- # [01:07] * Quits: gloob (gloob@moz-42625A2.litio.org) (Ping timeout)
- # [01:08] * Quits: MattN (MattN@A8588C62.B1B6FA80.709ED47B.IP) (Ping timeout)
- # [01:18] * Joins: MattN (MattN@A8588C62.B1B6FA80.709ED47B.IP)
- # [01:27] * Joins: gloob (gloob@moz-42625A2.litio.org)
- # [01:32] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Ping timeout)
- # [01:32] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [01:45] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Quit: lizzard)
- # [01:51] * Quits: ialagenchev (Adium@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
- # [01:56] <@firebot> trev.saunders@gmail.com granted review for attachment 774867 on bug 893166.
- # [01:56] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=893166 nor, --, ---, surkov.alexander, NEW, getTextBeforeOffset line end fails on wrapped lines
- # [02:05] * Quits: @hub (hub@moz-E12780DE.cable.teksavvy.com) (Ping timeout)
- # [02:19] * Joins: ialagenchev (Adium@moz-AD233F1B.dia.static.qwest.net)
- # [02:20] * Joins: hub (hub@moz-E12780DE.cable.teksavvy.com)
- # [02:20] * ChanServ sets mode: +o hub
- # [02:28] * Quits: rednaks (rednaks@C8C1A17E.AFD91B9D.55FFA9B4.IP) (Quit: Téléportation !)
- # [02:29] * Joins: fxa90id_ (fxa90id@moz-3DACD1A4.nvidia.com)
- # [02:30] <@firebot> trev.saunders@gmail.com denied review for attachment 775660 on bug 893812.
- # [02:30] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=893812 nor, --, ---, nobody, NEW, tweak document shutdown order
- # [02:50] * Joins: lizzard (ehenry@moz-CDB0DEA0.hfc.comcastbusiness.net)
- # [02:51] * Quits: ialagenchev (Adium@moz-AD233F1B.dia.static.qwest.net) (Quit: Leaving.)
- # [02:56] * Quits: lizzard (ehenry@moz-CDB0DEA0.hfc.comcastbusiness.net) (Quit: lizzard)
- # [03:00] * Joins: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net)
- # [03:09] * Quits: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net) (Quit: scott_gonzalez)
- # [03:19] * Joins: ialagenchev (Adium@moz-AD233F1B.dia.static.qwest.net)
- # [03:32] * Joins: lizzard (ehenry@moz-58147956.tmodns.net)
- # [03:39] * Quits: maxli (maxli@B5559308.9C54E60.2FA4EA88.IP) (Quit: Leaving.)
- # [03:40] * Quits: lizzard (ehenry@moz-58147956.tmodns.net) (Quit: lizzard)
- # [03:57] * khuey is now known as khuey|away
- # [03:59] * Joins: Jamie (Instantbir@moz-CA26021.jantrid.net)
- # [03:59] * Joins: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net)
- # [04:16] * Quits: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net) (Quit: scott_gonzalez)
- # [04:29] * Parts: ialagenchev (Adium@moz-AD233F1B.dia.static.qwest.net)
- # [04:49] <@firebot> jamie@nvaccess.org cancelled needinfo?(jamie@nvaccess.or g) on bug 795957.
- # [04:50] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=795957 nor, --, ---, yura.zenevich, NEW, [AccessFu] Support live regions
- # [04:54] * Joins: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net)
- # [04:58] * Quits: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net) (Quit: scott_gonzalez)
- # [05:06] * Quits: fxa90id_ (fxa90id@moz-3DACD1A4.nvidia.com) (Quit: Leaving)
- # [05:28] * Quits: Jesse (jruderman@moz-BBE3ABD.mv.mozilla.com) (Quit: Jesse)
- # [05:33] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Quit: yzen)
- # [05:35] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [05:37] <yzen> hi Jamie, thanks for the comment
- # [05:40] * Joins: surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com)
- # [05:40] * ChanServ sets mode: +o surkov
- # [05:51] * Joins: lizzard (ehenry@AE94D82.5FA88BA2.67E53AB7.IP)
- # [06:05] * Quits: lizzard (ehenry@AE94D82.5FA88BA2.67E53AB7.IP) (Quit: lizzard)
- # [06:14] * Quits: peteb-away (ptbrunet@moz-B51E1692.austin.res.rr.com) (Client exited)
- # [06:30] * Quits: @surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com) (Quit: surkov)
- # [07:11] * Joins: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net)
- # [07:24] * Quits: @firebot (firebot@moz-F8C1AF63.carolina.res.rr.com) (Ping timeout)
- # [07:38] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [07:43] * Joins: Jesse (jruderman@moz-537BCF9.hsd1.ca.comcast.net)
- # [07:53] * Joins: firebot (firebot@moz-F8C1AF63.carolina.res.rr.com)
- # [07:53] * ChanServ sets mode: +o firebot
- # [07:56] <@firebot> yura.zenevich@gmail.com requested feedback from eitan@monotonous. org for attachment 777602 on bug 795957.
- # [07:56] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=795957 nor, --, ---, yura.zenevich, NEW, [AccessFu] Support live regions
- # [07:59] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Quit: lizzard)
- # [08:01] * Joins: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net)
- # [08:36] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Quit: yzen)
- # [09:17] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Quit: lizzard)
- # [09:29] * Joins: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de)
- # [09:30] * ChanServ sets mode: +o marcoz
- # [09:35] * Joins: Gijs (gijs@moz-C11B0461.dsl.alice.nl)
- # [10:03] * Joins: nhirata (anonymous@514D186.F3BBB17D.144F44FA.IP)
- # [10:29] * Joins: rednaks (rednaks@2BFFE0C9.74C6FD40.360EF119.IP)
- # [11:03] <@firebot> New Core - Disability Access APIs bug 895323 filed by jruderman@gmail.com.
- # [11:03] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=895323 nor, --, ---, nobody, NEW, "ASSERTION: Bullet and accessible are in sync already!"
- # [11:10] * Joins: victorporof (victorporo@3692B70F.2484615.4A6B528C.IP)
- # [11:34] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [12:20] * Quits: Jamie (Instantbir@moz-CA26021.jantrid.net) (Quit: Instantbird 1.5a1pre -- http://www.instantbird.com)
- # [12:28] <Gijs> Does anyone know offhand where the code to display accesskey underlines on Windows lives?
- # [12:28] <Gijs> mxr-ing for 'accesskey' doesn't really get me anywhere. :(
- # [12:32] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [12:36] * Quits: @marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de) (Quit: Leaving.)
- # [12:39] * Quits: victorporof (victorporo@3692B70F.2484615.4A6B528C.IP) (Connection reset by peer)
- # [13:02] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [13:06] * Joins: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de)
- # [13:08] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Ping timeout)
- # [13:08] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [13:12] * Quits: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de) (Ping timeout)
- # [13:28] * Joins: scott_gonzalez (scott_gonz@moz-91C81A39.hrbgpa.fios.verizon.net)
- # [13:41] * Joins: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de)
- # [13:43] * Quits: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de) (Ping timeout)
- # [13:49] * Joins: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de)
- # [13:50] <marcoz> Gijs: I
- # [13:50] <marcoz> d suspect somewhere in the toolkit/widgets/windows code.
- # [13:51] <marcoz> Damn keyboard with the apostrophe in the wrong place
- # [13:51] <Gijs> marcoz: thanks!
- # [13:58] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [14:02] * Joins: maxli (maxli@F2D29657.F60B0462.67AC9B1.IP)
- # [14:07] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [14:16] * Joins: Justin_o (Justin_o@67828CC7.C1A51174.9D42CF23.IP)
- # [14:17] * Joins: surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com)
- # [14:17] * ChanServ sets mode: +o surkov
- # [14:20] * Joins: peteb-away (ptbrunet@moz-B51E1692.austin.res.rr.com)
- # [14:21] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [14:31] * Gijs grumbles
- # [14:31] <Gijs> Why does googling for "tabindex" not bring up *any* good reference page?
- # [14:31] <Gijs> Even webaim's page is unhelpful.
- # [14:35] * Quits: rednaks (rednaks@2BFFE0C9.74C6FD40.360EF119.IP) (Ping timeout)
- # [14:39] <marcoz> Gijs: What question do you have about tabindex? 0 is "integrate into natural tab order", −1 is "don't land on it at all", anything above 0 means "first the ones with 1, then the ones with 2" etc.
- # [14:39] <Gijs> marcoz: I can never remember which way around -1 and 0 are.
- # [14:39] <Gijs> and none of the first few pages explained (or even mentioned either!)
- # [14:39] <@firebot> yura.zenevich@gmail.com cancelled feedback?(eitan@monotonous. org) for attachment 777602 on bug 795957.
- # [14:39] <Gijs> we actually have https://developer.mozilla.org/en-US/docs/Accessibility/Keyboard-navigable_JavaScript_widgets
- # [14:39] <@firebot> yura.zenevich@gmail.com requested feedback from eitan@monotonous. org for attachment 777743 on bug 795957.
- # [14:39] <Gijs> which is very nice
- # [14:39] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=795957 nor, --, ---, yura.zenevich, NEW, [AccessFu] Support live regions
- # [14:40] <Gijs> But MDN's article on the accesskey property (rather than attribute? :S ) is like the 9th result on google, and then I had to click the "see also" link to get that info.
- # [14:40] <Gijs> webaim just talks about 1-N, and say using it is "not recommended"
- # [14:41] * Gijs is trying to make the Australis menupanel keyboard-navigable
- # [14:42] <marcoz> Gijs: Hehe, when I looked for that exact info a few months ago for my Easy ARIA Tip #6 blog post, I looked on MDN first for the tabindex attribute. I had the same problem as you, didn't remember which values did what. I later did a search on Google and found the same bogus results you did.
- # [14:42] <yzen> marcoz: hi, thanks for feedback, i updated the patch a little bit.
- # [14:42] <marcoz> yzen: Hi! Cool, just saw it!
- # [14:42] <yzen> marcoz: at the moment the hide tests are not working though, will take a look at them soon
- # [14:43] <yzen> marcoz: did you have any comments about the logic by any chance ?
- # [14:43] * Gijs can now tab around in the menupanel - woohoo! :)
- # [14:44] <marcoz> Gijs: Very cool!
- # [14:44] <marcoz> yzen: Not off-hand, no. I was looking through it and then looking at what the tests say currently. The Hide stuff may be tricky because the moment the event is fired, the text and so is already hidden/removed.
- # [14:45] <yzen> marcoz: ya, however i tried logging the content of the event's accessible (children and their names) and it was still there so i ll investigate a little further
- # [14:46] <yzen> marcoz: i guess the traversal will not include defunct nodes
- # [14:46] <marcoz> yzen: Remember that screen readers on Windows usually deal with a cached version of the document to which they apply the event results. So if they receive a hide or textRemoved event, they can still use the cached data. AccessFu, as well as Orca on Linux and VoiceOver on the Mac, operate with our representation directly without caching it. So some of the spec about hidden text may be tailored to the Windows use case very specifica
- # [14:46] <marcoz> This spec is a few years old, nobody was really talking about mobile then, and Windows was the dominant desktop operating system.
- # [14:46] <marcoz> We're charting a bit of new ground here.
- # [14:47] <yzen> marcoz: i think i might be able to get hold of it if for the case of hidden defunct stuff is actually included ?
- # [14:47] <marcoz> yzen: We are purposely not trying to traverse defunct accessibles, because they can be cycle-collected at any moment.
- # [14:48] <marcoz> Gijs: So, is the tabbing efficient? ;)
- # [14:48] <yzen> marcoz: good point, alright bbib
- # [14:49] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Quit: yzen)
- # [14:49] <Gijs> marcoz: not sure what you mean... :)
- # [14:51] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [14:51] * ChanServ sets mode: +qo davidb davidb
- # [14:52] <@davidb> heyo
- # [14:53] * Gijs waves
- # [14:53] <Gijs> marcoz: out of interest, is there an easy way to find the first focusable element to focus when showing a panel?
- # [14:55] <Gijs> By default, it seems showing a panel doesn't move focus into it, and tab/shift-tab will continue to navigate in the main window.
- # [14:55] * Gijs would like to avoid having to "manually" figure out what the first focusable item in the panel is
- # [15:04] <marcoz> Gijs: Well, if it is a doorhanger, focus is not supposed to go to the panel automatically.
- # [15:05] <marcoz> Gijs: If it should be focusable, setting focus to it should automatically put focus on the first focusable item inside it.
- # [15:05] <marcoz> Gijs: There is a property for a panel I don't recall, but I remember that this was discussed at length in bug 616136 and dependencies.
- # [15:05] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=616136 nor, --, Firefox 6, margaret.leibovic, VERI FIXED, Give popup notification panels role="alert" and make their close buttons tabbable
- # [15:05] <marcoz> Heyo davidb!
- # [15:06] <Gijs> it's not a doorhanger, so I'm not sure how it works...
- # [15:06] <Gijs> on the phone now, will get back to this later
- # [15:06] * Gijs is now known as Gijs_away
- # [15:06] <@firebot> surkov.alexander@gmail.com cancelled review?(trev.saunders@gmail .com) for attachment 776521 on bug 891338.
- # [15:06] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 777757 on bug 891338.
- # [15:07] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=891338 maj, --, ---, surkov.alexander, NEW, Popup accessibility broken
- # [15:07] <marcoz> Gijs: And by efficient I mean this: If tab moves into a toolbar, the best practice suggests to not jump to each individual tool bar item via tab, but to use the arrow keys instead. Tab should move to the next tool bar or any focusable item following that tool bar. That is to avoid too many tab stops. Arrow keys are a by-design means to navigate items within toolbars and menus.
- # [15:15] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [15:24] * Joins: yzen (yzen@C8A3A5EA.A6295926.9D42CF23.IP)
- # [15:24] * Joins: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net)
- # [15:25] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [15:25] * Quits: yzen (yzen@C8A3A5EA.A6295926.9D42CF23.IP) (Ping timeout)
- # [15:26] <@firebot> marco.zehe@googlemail.com granted feedback for attachment 777743 on bug 795957.
- # [15:26] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=795957 nor, --, ---, yura.zenevich, NEW, [AccessFu] Support live regions
- # [15:34] * Quits: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com) (Ping timeout)
- # [15:41] * Joins: rednaks (rednaks@C8C1A17E.AFD91B9D.55FFA9B4.IP)
- # [15:43] * Joins: Mic (Mic@moz-6A3BC257.superkabel.de)
- # [15:43] * Joins: yzen (yzen@67828CC7.C1A51174.9D42CF23.IP)
- # [15:46] * Gijs_away is now known as Gijs
- # [15:46] * Quits: Justin_o (Justin_o@67828CC7.C1A51174.9D42CF23.IP) (Quit: Justin_o)
- # [15:47] <Gijs> marcoz: hrm. That sounds difficult to do. Do we implement this anywhere else?
- # [15:48] <Gijs> marcoz: I don't think our toolbars are keyboard-accessible at all at the minute, apart from their input elements
- # [15:49] <Gijs> marcoz: so the menupanel essentially has a big area with buttons in it which do all kinds of different things. Could be other things than buttons, too. Then at the bottom it has a button called "Customize", and a "Help" button.
- # [15:49] <Gijs> marcoz: (those are static; the buttons in the big area can be changed by the user)
- # [15:50] * Quits: rednaks (rednaks@C8C1A17E.AFD91B9D.55FFA9B4.IP) (Quit: Téléportation !)
- # [15:50] <Gijs> marcoz: I'm guessing we would want one tabstop in the top area, and then tabstops for the two buttons in the footer, and then tab out of the panel?
- # [15:52] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [15:52] <Gijs> marcoz: I guess it's also relevant to note, I would only move focus to it if you make the panel appear using a keyboard shortcut. It will never "pop up" automatically.
- # [15:54] <marcoz> Gijs: Well, difficult to say without having actually tried it. We do have the F6 key to move out of a panel, so having a tab stop for each button may be OK. I'm just worried that users will lose a lot of efficiency, like Alt+t, O for Tools/Options is very much deeply entrenched in people's muscle memory.
- # [15:54] <Gijs> marcoz: oh, we're not getting rid of the normal menu.
- # [15:55] <marcoz> Gijs: So what does this panel do, then?
- # [15:55] <Gijs> marcoz: so we're not breaking anything (except the Firefox app menu in the top-left corner, but to my knowledge that was never keyboard-accessible)
- # [15:55] <Gijs> marcoz: it has a bunch of buttons which provide more visibility to 'popular' options.
- # [15:56] <Gijs> marcoz: eg. it has buttons for opening the add-ons manager, going fullscreen, opening the find bar, etc.
- # [15:56] <Gijs> marcoz: a surprising number of users have no idea those features exist, so this gives them more visibility.
- # [15:57] * Gijs isn't entirely sure why it needs to be keyboard-accessible per-se, as everything has keyboard-accessible equivalents
- # [15:59] <marcoz> Gijs: Well, it certainly helps to make this keyboard accessible, improves communication between blind and sighted people, like saying "Go to the new menu pael and select…". Many sighted folks don't even know how to open a menu in Windows any more. Mouse users, that is.
- # [15:59] * marcoz thinks Firefox should just get Ribbons like Office. And smirks.
- # [16:00] * Gijs shudders
- # [16:00] <Gijs> I think ribbons rank close to the top of the most horrible UI I know.
- # [16:00] * Gijs can never find anything
- # [16:00] <Gijs> Then again, maybe if I was confronted with them more I would understand whatever the logic is and get used to them.
- # [16:00] <Gijs> Anyhow! :)
- # [16:01] <marcoz> Gijs: That's why I was smirking. I wasn't at all serious. The only person who really seems to like them is my father. God knows why. But I switched him over to a Mac 2 years ago, and he doesn't use the MS cruft any more.
- # [16:01] <Gijs> marcoz: so I added a shortcut to open the menu, and now I need to make the contents accessible.
- # [16:01] <Gijs> this is essentially where I'm at.
- # [16:01] <Gijs> there are some subtleties with subviews (essentially, the panel's contents shift visually and show a "submenu" kind of thing), but I'll sort that out separately.
- # [16:02] <Gijs> I'll try to see if focusing the panel works.
- # [16:02] <marcoz> Gijs: I'd say just go with the tab stop for each focusable item.
- # [16:02] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [16:02] * clown is now known as clown_mtg
- # [16:02] <marcoz> Gijs: For screen reader users, this has much more characteristics of a dialog or giant two dimensional tool bar than a menu anyway.
- # [16:02] <marcoz> And as long as the buttons have access keys, there is a way to get to them quicker than by tabbing once one memorises them.
- # [16:03] <marcoz> Gijs: Ah huh that sounds a bit more tricky. Will be interesting to see this in action once Australis lands!
- # [16:04] * Gijs whistles at the mention of access keys
- # [16:05] <Gijs> (no, they don't... :( )
- # [16:05] <Gijs> (and because of visual issues we're having with the labels, I'm not sure they will because we can't show the access keys because of silly implementation details of the toolbarbutton)
- # [16:06] <Gijs> marcoz: anyway, thanks a lot for sparring about this, very helpful!
- # [16:07] <marcoz> Gijs: You're very welcome!
- # [16:09] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [16:11] <marcoz> maxli: Regarding bug 895000, is Reader mode now a certain page action?
- # [16:11] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=895000 nor, --, ---, maxli, NEW, New unlabelled buttons can be accessibility focused on awesomebar
- # [16:11] <maxli> marcoz: yes
- # [16:12] <marcoz> maxli: Cool!
- # [16:13] <marcoz> maxli: Also, thanks for catching this so quickly and acting on it!
- # [16:13] <maxli> marcoz: np
- # [16:14] <yzen> marcoz: so have you had any thoughts about hide where the subtree is defunct? should we announce it somehow ? i do have access to the event's accessible itself at least
- # [16:17] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [16:22] * Quits: yzen (yzen@67828CC7.C1A51174.9D42CF23.IP) (Ping timeout)
- # [16:22] <marcoz> yzen: I am very ambivalent about announcing hides. I think we should go by the spec, that is, only announce them if specifically requested by the author, but not in other cases. I think things might get too talkative otherwise.
- # [16:23] * Joins: yzen (yzen@C8A3A5EA.A6295926.9D42CF23.IP)
- # [16:23] <marcoz> yzen: I am very ambivalent about announcing hides. I think we should go by the spec, that is, only announce them if specifically requested by the author, but not in other cases. I think things might get too talkative otherwise.
- # [16:25] <yzen> marcoz: yes, that's right, it will only announce hide if relevant, but when we do, perhaps you have suggestion on what should be announced since the subtree would be defunct as we discussed earlier
- # [16:28] <marcoz> yzen: If you have access to the text that got removed, and can do that safely, we should announce it followed by the word "hidden".
- # [16:29] * marcoz believes nobody has actually implemented removals for the exact reason we're discussing right now. ;-)
- # [16:30] <yzen> marcoz: good point :)
- # [16:35] <@firebot> gijskruitbosch+bugs@gmail.com requested review from mconley@mozilla.com for attachment 777794 on bug 881937.
- # [16:35] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=881937 nor, --, ---, gijskruitbosch+bugs, ASSI, The Australis panel menu should be keyboard accessible
- # [16:43] <@firebot> gijskruitbosch+bugs@gmail.com requested review from mconley@mozilla.com for attachment 777802 on bug 881937.
- # [16:58] <@firebot> gijskruitbosch+bugs@gmail.com cancelled review?(mconley@mozilla.com ) for attachment 777794 on bug 881937.
- # [16:58] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=881937 nor, --, ---, gijskruitbosch+bugs, ASSI, The Australis panel menu should be keyboard accessible
- # [17:00] * Quits: Gijs (gijs@moz-C11B0461.dsl.alice.nl) (Quit: poof)
- # [17:16] * Quits: @surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com) (Ping timeout)
- # [17:19] * Joins: surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com)
- # [17:19] * ChanServ sets mode: +o surkov
- # [17:19] <@firebot> surkov.alexander@gmail.com granted in-testsuite on bug 882767.
- # [17:19] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=882767 nor, --, ---, surkov.alexander, ASSI, don't expose whitespace accessibles in context of grids
- # [17:21] * Quits: Mic (Mic@moz-6A3BC257.superkabel.de) (Ping timeout)
- # [17:23] <@firebot> dao@mozilla.com changed the Component on bug 881937 from Keyboard Navigation to Toolbars and Customization.
- # [17:23] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=881937 nor, --, ---, gijskruitbosch+bugs, ASSI, The Australis panel menu should be keyboard accessible
- # [17:42] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [17:44] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [17:44] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [17:44] * ChanServ sets mode: +qo davidb davidb
- # [18:10] * marcoz filed a number of GAIA bugs. Poor eeejay :)
- # [18:12] <@davidb> bring it on
- # [18:14] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [18:14] <marcoz> davidb: All dependencies of bug 793789. :)
- # [18:14] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=793789 nor, --, ---, nobody, NEW, [RTL] The expand icon is on the left in article edit
- # [18:14] <marcoz> Or maybe not.
- # [18:15] <@davidb> lol
- # [18:16] <marcoz> davidb: Bug 893789.
- # [18:16] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=893789 nor, --, ---, nobody, NEW, [Meta] GAIA accessibility bugs
- # [18:16] <@davidb> nice
- # [18:17] <marcoz> OK, but now going off-line for the evening. See you tomorrow!
- # [18:17] * Quits: marcoz (marco.zehe@moz-6B3B4E43.dip0.t-ipconnect.de) (Quit: Leaving.)
- # [18:20] * Quits: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com) (Ping timeout)
- # [18:24] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [18:25] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Quit: victorporof)
- # [18:26] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [18:32] * khuey|away is now known as khuey
- # [18:42] * clown_mtg is now known as clown_afk
- # [18:53] <@eeejay> surkov, what about adding a "force activate" to the action list of all accessibles?
- # [18:53] <@surkov> eeejay: I could live with it too :)
- # [18:54] <@surkov> eeejay: but I think I'd go with negative index
- # [18:54] <@eeejay> surkov, then it would remain semantically correct, no?
- # [18:55] <@surkov> not sure, some accessible having "force activate" might be not really activated on this action
- # [18:55] <@eeejay> surkov, with the current action implementation (where it almost always performs a click anyway), i could see ATs never even looking at the available actions, and always simply doing doAction(-4)
- # [18:55] * @eeejay puts down his thoughts in bug..
- # [18:56] <@surkov> eeejay: they can do, but action interface allows you to list possible actions
- # [18:56] <@surkov> like if it's jump then they can say jump
- # [18:56] <@eeejay> surkov, right. but who are we kidding? It is always a click :)
- # [18:57] <@surkov> on implementation level yes
- # [18:57] <@surkov> currently we have only one element that can have two actions: img
- # [18:57] <@surkov> click and longdesc
- # [18:57] <@eeejay> surkov, you mean a link image?
- # [18:59] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [19:00] <@surkov> yes
- # [19:00] <@surkov> anyway longdesc is not a click in current implementation
- # [19:00] <@surkov> eeejay: but AT can rely on action name to announce it properly
- # [19:01] <@surkov> it makes sense too
- # [19:01] <@surkov> if it's jump then say jump, if it's activate then say activate
- # [19:01] <@surkov> otherwise AT should keep this logic on their side
- # [19:02] <@surkov> like if it's tree item and if it's in collapsed state then say "expand" action
- # [19:03] <@eeejay> surkov, yeah. but i am afraid of "force activate" being misused. Lets say that in the future we introduce a control that has a default action that is not implemented with a click. The AT will try "force activate" and fail, because it is too lazy to interrogate the object for available actions.
- # [19:04] <@eeejay> surkov, right now, we at least know that doAction(0) will do whatever the default action is
- # [19:04] <@surkov> eeejay: I'm pretty sure they will fix it when they realize it :)
- # [19:06] <@surkov> btw, if img is not clickable then default action is longdesc and it's not click-activated action
- # [19:08] <@eeejay> surkov, right.. that is what i thought. so the only way for it to have two actions is to have an onclick.
- # [19:08] <@surkov> yep
- # [19:09] <@eeejay> changing the role, i assume would get rid of the longdesc action?
- # [19:09] * Joins: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de)
- # [19:09] <@tbsaunde> using a negative index sounds better to me, but the action interface is a train wreck
- # [19:10] <@tbsaunde> eeejay: I don't believe so, I believe something like <img role=checkbox longdesc="blah"> would have a longdesc action
- # [19:11] <@eeejay> how about having a "force activate" action only on items that don't have a default action already (including longdesc)
- # [19:12] <@tbsaunde> I wouldn't be suprised if that causes issues on desktop because now everything is acionable, but otherwise I guess its liveable, but negative index seems easier and nicer
- # [19:12] <@eeejay> that seems least disruptive to me, and semantically complete: "Well, it isn't a button or anything else you would normally click on, but you could poke it anyway"
- # [19:13] <@tbsaunde> I'd sort of rather such an action not be the default
- # [19:14] * Quits: nhirata (anonymous@514D186.F3BBB17D.144F44FA.IP) (Quit: nhirata)
- # [19:14] <@tbsaunde> negative indexes for here is a general thing you can do is an existing concept so adding another thing like that seems nice
- # [19:15] <@eeejay> tbsaunde, i guess we are in luck that ia2 and at-spi use signed indexes as arguments
- # [19:16] <@eeejay> tbsaunde, what else are they used for?
- # [19:17] <@eeejay> nm, found it
- # [19:18] <@tbsaunde> you could always do the same thing with uint_MAX / UINT_MAX - 1
- # [19:23] <@eeejay> tbsaunde, would we need to reserve -4 in ia2 for this?
- # [19:23] <@eeejay> or if they add -4 for something else, we need to translate it..
- # [19:23] <@eeejay> same with at-spi i guess
- # [19:34] <@firebot> surkov.alexander@gmail.com requested review from the wind for attachment 777918 on bug 893166.
- # [19:34] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=893166 nor, --, ---, surkov.alexander, NEW, getTextBeforeOffset line end fails on wrapped lines
- # [19:36] * khuey is now known as khuey|away
- # [19:36] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [19:37] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 777918 on bug 893166.
- # [19:37] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [19:41] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Ping timeout)
- # [19:41] * Quits: @surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com) (Ping timeout)
- # [19:42] * Quits: peter (peter@moz-BEB1F1E3.adsl.alicedsl.de) (Client exited)
- # [19:43] * Joins: surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com)
- # [19:43] * ChanServ sets mode: +o surkov
- # [19:44] <@tbsaunde> eeejay: possible, but shouldn't be nearly as annoying as the thing you proposed
- # [19:45] <@eeejay> tbsaunde, how would it be annoying?
- # [19:47] <@tbsaunde> it interacts with other things
- # [19:48] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [19:54] <@firebot> trev.saunders@gmail.com granted review for attachment 777918 on bug 893166.
- # [19:54] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=893166 nor, --, ---, surkov.alexander, NEW, getTextBeforeOffset line end fails on wrapped lines
- # [19:57] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [19:57] * Joins: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net)
- # [20:00] * khuey|away is now known as khuey
- # [20:06] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Ping timeout)
- # [20:08] * Joins: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net)
- # [20:15] * Quits: yzen (yzen@C8A3A5EA.A6295926.9D42CF23.IP) (Quit: yzen)
- # [20:32] * khuey is now known as khuey|away
- # [20:32] * khuey|away is now known as khuey
- # [20:34] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [20:42] * Joins: davidb (davidb@moz-C56D5C4E.dsl.bell.ca)
- # [20:42] * ChanServ sets mode: +qo davidb davidb
- # [20:44] * Quits: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com) (Ping timeout)
- # [20:56] * Joins: victorporof_ (victorporo@C5842721.834833CE.4A6B528C.IP)
- # [20:57] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [20:57] * Quits: victorporof_ (victorporo@C5842721.834833CE.4A6B528C.IP) (Client exited)
- # [20:58] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [21:01] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [21:05] * Joins: victorporof_ (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [21:06] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [21:07] * Quits: victorporof_ (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [21:21] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [21:23] * Joins: rednaks (rednaks@C8C1A17E.AFD91B9D.55FFA9B4.IP)
- # [21:27] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [21:28] * Joins: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP)
- # [21:28] * Quits: lizzard (ehenry@moz-5A4BE8E8.dsl.static.sonic.net) (Quit: lizzard)
- # [21:30] * khuey is now known as khuey|away
- # [21:30] * Quits: victorporof (victorporo@B7024D9D.5270B130.78DC9724.IP) (Ping timeout)
- # [21:31] * Quits: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com) (Ping timeout)
- # [21:58] * Quits: maxli (maxli@F2D29657.F60B0462.67AC9B1.IP) (Quit: Leaving.)
- # [21:59] * Joins: maxli (maxli@F2D29657.F60B0462.67AC9B1.IP)
- # [22:00] * Joins: nhirata (anonymous@moz-7E1F115E.range86-169.btcentralplus.com)
- # [22:00] <yzen> eeejay: hi
- # [22:07] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [22:07] * Joins: fxa90id_ (fxa90id@moz-3DACD1A4.nvidia.com)
- # [22:15] * Quits: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com) (Ping timeout)
- # [22:31] * Quits: maxli (maxli@F2D29657.F60B0462.67AC9B1.IP) (Quit: Leaving.)
- # [22:35] * Quits: @surkov (surkov@moz-6EE3EE5C.cpe.net.cable.rogers.com) (Quit: surkov)
- # [22:47] * Joins: SteveF (chatzilla@moz-3F778890.cable.virginmedia.com)
- # [22:48] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Quit: yzen)
- # [22:49] * Joins: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP)
- # [22:50] * Joins: yzen1 (yzen@moz-2C4D9C95.eng.wind.ca)
- # [22:51] * Quits: yzen (yzen@AA2C085E.10C8AEF8.9A8C35B4.IP) (Ping timeout)
- # [22:56] * khuey|away is now known as khuey
- # [22:57] * Quits: yzen1 (yzen@moz-2C4D9C95.eng.wind.ca) (Ping timeout)
- # [23:06] * Quits: @davidb (davidb@moz-C56D5C4E.dsl.bell.ca) (Quit: davidb)
- # [23:08] * Joins: yzen (yzen@moz-2C4D9C95.eng.wind.ca)
- # [23:13] * Quits: yzen (yzen@moz-2C4D9C95.eng.wind.ca) (Ping timeout)
- # [23:13] * Quits: clown_afk (clown@67828CC7.C1A51174.9D42CF23.IP) (Quit: Leaving.)
- # [23:51] * Joins: yzen (yzen@moz-2C4D9C95.eng.wind.ca)
- # [23:57] <@firebot> mounir@lamouri.fr cancelled needinfo?(mounir@lamouri.fr ) on bug 291082.
- # [23:57] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=291082 nor, --, ---, almasry.mina, NEW, preventDefault doesn't block keyboard navigation in select-one drop-down lists
- # Session Close: Fri Jul 19 00:00:00 2013
The end :)