/irc-logs / mozilla / #accessibility / 2012-02-27 / end
Options:
- # Session Start: Mon Feb 27 00:00:01 2012
- # Session Ident: #accessibility
- # [00:05] * Quits: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP) (Input/output error)
- # [00:54] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [01:19] <@firebot> philringnalda@gmail.com changed the Status on bug 728905 from ASSIGNED to RESOLVED.
- # [01:19] <@firebot> philringnalda@gmail.com set the Resolution field on bug 728905 to FIXED.
- # [01:19] <@firebot> philringnalda@gmail.com changed the Target Milestone on bug 728905 from --- to mozilla13.
- # [01:19] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=728905 nor, --, mozilla13, markcapella, RESO FIXED, telemetry ZoomText, YouDao Dictionary and Kazaguru
- # [01:40] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [01:44] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [01:45] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Ping timeout)
- # [01:57] * Joins: surkov (surkov@D5570440.C1645F5A.34044A7F.IP)
- # [01:57] * ChanServ sets mode: +o surkov
- # [01:58] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [01:59] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:06] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [02:10] <@firebot> surkov.alexander@gmail.com granted feedback for attachment 600808 on bug 696975.
- # [02:10] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=696975 nor, --, ---, markcapella, ASSI, extend the list of legitimate data table structures for layout-guess object attribute
- # [02:11] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [02:11] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Ping timeout)
- # [02:11] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:13] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [02:14] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:16] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [02:24] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [02:24] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:28] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [02:29] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:31] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [02:31] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:34] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 12.0a2/20120225042011])
- # [04:18] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [04:48] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [04:59] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [06:20] * Joins: MarcoZ (marco.zehe@moz-1907351C.dip.t-dialin.net)
- # [06:20] * ChanServ sets mode: +o MarcoZ
- # [06:27] * Quits: @MarcoZ (marco.zehe@moz-1907351C.dip.t-dialin.net) (Input/output error)
- # [06:28] * Joins: MarcoZ (marco.zehe@moz-1907351C.dip.t-dialin.net)
- # [06:28] * ChanServ sets mode: +o MarcoZ
- # [06:49] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [06:52] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Client exited)
- # [06:57] * Joins: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP)
- # [07:03] <@tbsaunde> surkov: hi
- # [07:04] <@surkov> hi, tbsaunde
- # [07:05] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [07:06] <@tbsaunde> surkov: thoughts on if we want to export the TableAccessible interface
- # [07:06] <@tbsaunde> and if we want to put it in a namespace?
- # [07:06] <@surkov> 1) let's do if we need 2) makes sense
- # [07:06] <@surkov> idea to put whole a11y under a11y namespace, right?
- # [07:07] <@surkov> about 1) if -> when
- # [07:08] <@tbsaunde> surkov: I'm not sure it really makes sense to namespace everything
- # [07:08] <@tbsaunde> a11y::Accessible seems redundant and I'm not sure who else would want to have a class Accessible or FooAccessible
- # [07:09] <@surkov> compare to dom::Element
- # [07:09] <@tbsaunde> true
- # [07:09] <@surkov> since we aren't libxul anymore then anybody can include our files without export feature
- # [07:09] <@surkov> and maybe it makes sense to avoid possible collisions
- # [07:10] <@surkov> I used to think we don't need a11y namespace for stuffs we don't export but it seems we add it to everything now
- # [07:10] <@tbsaunde> yeah, on the other hand exporting a header can be seen to imply that your ok with other people using that interface
- # [07:11] <@tbsaunde> well, afaik you've been the one adding it :)
- # [07:11] <@surkov> that makes sense but the exported or not exported state doesn't deny them to use it
- # [07:12] <@surkov> hm, I thought that something comes from you :)
- # [07:12] <@tbsaunde> true
- # [07:16] <@tbsaunde> surkov: when did I add namespacing to stuff we didn't export?
- # [07:17] <@surkov> not sure, I just expressed my feeling :)
- # [07:17] <@tbsaunde> surkov: does adding another generic directory really make sense to you? I can't think of many generic interfaces we're going to have but I don't really know all oyur plans
- # [07:19] <@surkov> tbsaunde: we 17 files for accessible objects agains 67 in directory
- # [07:20] <@surkov> some objects are closely related with accessible objects to be considered as part of accessible object, for example, NotificationController
- # [07:21] <@tbsaunde> yeah, ok I gues I'll add the generic directory then
- # [07:22] <@surkov> ok
- # [07:56] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [07:58] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [08:14] * khuey is now known as khuey|away
- # [08:29] <@firebot> trev.saunders@gmail.com requested review from surkov.alexander@gm ail.com for attachment 600848 on bug 648265.
- # [08:29] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=648265 nor, --, ---, trev.saunders, NEW, provide dexcomed table interface version
- # [09:03] <@firebot> surkov.alexander@gmail.com granted review for attachment 600848 on bug 648265.
- # [09:03] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=648265 nor, --, ---, trev.saunders, NEW, provide dexcomed table interface version
- # [09:11] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 600850 on bug 729831.
- # [09:11] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=729831 cri, --, ---, surkov.alexander, NEW, Crash @ mozilla::a11y::StyleInfo::Margin
- # [09:22] * Quits: @hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
- # [09:28] <@firebot> trev.saunders@gmail.com granted review for attachment 600850 on bug 729831.
- # [09:28] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=729831 cri, --, ---, surkov.alexander, NEW, Crash @ mozilla::a11y::StyleInfo::Margin
- # [09:48] <@tbsaunde> surkov: fwiw implementing both TableAccessible and nsIAccessibleTable on the same object seems to currently create a bunch of over loaded virtual warnings, its not particularly nice but I think the best option may just be to ignore them until we setup xpcom wrappers
- # [09:48] <@tbsaunde> and try to do that soon
- # [09:49] <@surkov> tbsaunde: ok
- # [09:58] <@firebot> jamie@nvaccess.org changed the Status on bug 677154 from RESOLVED to VERIFIED.
- # [09:59] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=677154 maj, P1, mozilla13, surkov.alexander, VERI FIXED, Detached document accessibility tree
- # [10:04] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [10:38] * Quits: @surkov (surkov@D5570440.C1645F5A.34044A7F.IP) (Quit: surkov)
- # [11:20] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [11:27] * Joins: sgimeno (santiago.g@A922400D.48362F18.BCC03455.IP)
- # [12:00] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [12:33] * Joins: surkov (surkov@D5570440.C1645F5A.34044A7F.IP)
- # [12:33] * ChanServ sets mode: +o surkov
- # [12:39] * Quits: @surkov (surkov@D5570440.C1645F5A.34044A7F.IP) (Quit: surkov)
- # [12:48] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [12:52] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [13:22] * Quits: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP) (Input/output error)
- # [13:25] * Joins: surkov (surkov@D5570440.C1645F5A.34044A7F.IP)
- # [13:25] * ChanServ sets mode: +o surkov
- # [13:31] <@tbsaunde> surkov: namepsacing stuff does have the sort of unfortunate effect that you have mozilla::a11y::foo all over in headers
- # [13:32] <@surkov> that's right until the header is in namespace
- # [13:33] <@tbsaunde> true
- # [13:35] <@tbsaunde> so, i had been planning to remove the impls of methods on TableAccessible when they were dexpcomed and weren't needed to make things build, but it occurs to me sometimes we could just leave the default impl and only sometimes implement it for example for caption() we could leave the return null definition on TableAccessible and only over ride it for html tables
- # [13:38] * Quits: @jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
- # [13:51] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [13:53] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [13:56] <@tbsaunde> surkov: ?
- # [13:56] <@surkov> tbsaunde: 2nd is reasonable
- # [13:57] <@tbsaunde> ok
- # [13:59] <@tbsaunde> surkov: so I don't really understand what you want with casting stuff, the downcasting function on nsAccessible is with the other downcasting functions, but the classes that over ride it don't have any other down casting functions so I'm not sure what your going to group it with though I suppose we could add a new section to them instead of throwing it in with general nsAccessible things
- # [14:01] <@surkov> tbsaunde: ok, up to you
- # [14:01] <@surkov> that's not down casting, that' true
- # [14:12] * Quits: nhirata (nhirata.bu@moz-2A9C9106.hsd1.ca.comcast.net) (Quit: nhirata)
- # [14:17] * Joins: jprmc (jprmc@69F4FD58.5BCEC6DB.DA78B690.IP)
- # [14:17] * ChanServ sets mode: +o jprmc
- # [14:21] * Joins: nhirata (nhirata.bu@moz-2A9C9106.hsd1.ca.comcast.net)
- # [14:31] <@MarcoZ> surkov: tbsaunde: I'm off to CSUN tomorrow, so reviews or other stuff may be slow as is usual for this kind of event. :)
- # [14:31] <@surkov> MarcoZ: ok, happy jorney
- # [14:31] <@MarcoZ> surkov: Thanks!
- # [14:44] * Joins: jpr_ (jprmc@A21DCCBC.5BCEC6DB.DA78B690.IP)
- # [14:45] * Quits: @jprmc (jprmc@69F4FD58.5BCEC6DB.DA78B690.IP) (Ping timeout)
- # [14:45] * nhirata is now known as nhirata|brb
- # [15:06] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [15:06] * ChanServ sets mode: +qo davidb davidb
- # [15:09] * Quits: jpr_ (jprmc@A21DCCBC.5BCEC6DB.DA78B690.IP) (Quit: Leaving)
- # [15:11] <@davidb> hi all!
- # [15:12] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [15:13] <@MarcoZ> Hi davidb!
- # [15:13] <@davidb> heyo
- # [15:14] <@davidb> surkov: does having a root frame mean the frame tree is complete?
- # [15:15] <@surkov> davidb: what do you mean by complete?
- # [15:16] <@davidb> surkov: oh I see you mention the crash only happens on doc accessible. ok.
- # [15:16] * @surkov shrugs
- # [15:16] <@davidb> heh
- # [15:16] <@davidb> surkov: nice debugging.
- # [15:16] <@surkov> thx
- # [15:17] * nhirata|brb is now known as nhirata
- # [15:18] <@MarcoZ> surkov: So was this bug there all along, or was it indeed introduced by the other bug fix?
- # [15:19] <@surkov> MarcoZ: I assume you talk about bug that marked as regression? then no
- # [15:20] <@MarcoZ> surkov: Exactly.
- # [15:20] <@surkov> if you mean the patch approach
- # [15:20] <@surkov> technically we didn't have a problems until crash, we just did things that doesn't make sense to do
- # [15:20] <@MarcoZ> surkov: Basically, what I'm asking is, what was it that started exposing this crash? Was it something recent, and the actual crash cause had been lurking for longer? Or was this mechanism introduced recently and this possibility of a crash just overlooked?
- # [15:22] <@surkov> potentially crash lived in our code, internally we work under assumption everything is alright. That crash shown us that this assumption may be broken under certain circumstances so did a fix to let us working under that assumption
- # [15:23] <@surkov> potentially crash -> potential crashes
- # [15:24] * khuey|away is now known as khuey
- # [15:27] <@MarcoZ> surkov: OK!
- # [15:28] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 600890 on bug 728907.
- # [15:29] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=728907 nor, --, ---, surkov.alexander, ASSI, don't use GetComputedStyle for text attributes
- # [15:29] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [15:29] * ChanServ sets mode: +o jprmc
- # [15:31] <@MarcoZ> davidb: Got mail from Jason Kiss asking whether I could help with the role mappings for Mad a11y for ARIA and HTML5. And I just thought: "Man, we're just trying to figure out ourselves what's appropriate." ;)
- # [15:32] <@MarcoZ> s/mad/Mac/
- # [15:32] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [15:33] <@davidb> heheh
- # [15:33] <@davidb> I was cc'ed on that email.
- # [15:33] <@davidb> Jason is really good. I hope he can keep his energy/productivity up in the standards work.
- # [15:33] <@davidb> I fear it is the 'honeymoon' period :)
- # [15:34] <@MarcoZ> davidb: Yes, he is, and I hope so, too! I'll let him and Steve know that Hub and the rest of our team are still picking up the pieces and trying to get things into shape, and that we're figuring out things as we go.
- # [15:35] <@davidb> OK. Maybe reach out to James Craig for collab?
- # [15:37] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [15:38] <@MarcoZ> I will do that!
- # [15:38] <@MarcoZ> But before I do anything concrete, I want to get some 1:1 time with Hub this week, possibly Wednesday where people set up their booths, to sit down and hack and figure out things that are just plain wrong in our role mappings and other stuff.
- # [15:42] <@MarcoZ> ยด/msg DavidB To be honest, I think Hub needs to show more what he is cooking up, put more patches into bugs even when they're WIPs. I am never really sure what he's working on.
- # [15:43] <@MarcoZ> Oops
- # [15:43] <@MarcoZ> :) But that's OK.
- # [15:46] <@davidb> heehee
- # [15:46] <@davidb> we're all a bit like that
- # [15:46] <@davidb> that reminds me i need to do my update thing
- # [15:47] <@davidb> http://benjamin.smedbergs.us/weekly-updates.fcgi/
- # [15:48] <@firebot> bolterbugz@gmail.com requested review from roc@ocallahan.org for attachment 598488 on bug 495912.
- # [15:48] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=495912 nor, --, ---, bolterbugz, NEW, Expose alternative content in Canvas element to ATs
- # [15:51] * Joins: shorlander-away (shorlander@moz-853043D6.dhcp.insightbb.com)
- # [15:52] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [15:52] <@MarcoZ> davidb: Can you kick off a try build of that Canvas work that I can download and try, possibly even show, in San Diego?
- # [15:53] * shorlander-away is now known as shorlander
- # [15:53] <@davidb> yes
- # [15:55] <@davidb> MarcoZ: I think this is it: http://ftp.mozilla.org/pub/mozilla.org/firefox/try-builds/dbolter@mozilla.com-58ad9e6f9dd9/
- # [15:56] <@davidb> surkov: I'm unsure about childAtPoint testing, since that's phase 2 really.
- # [15:56] <@davidb> phase 2 is likely about binding elements to draw paths
- # [15:56] <@davidb> it is under-specced
- # [15:57] <@surkov> davidb: but you expose accessibles for subtree, right? then they must have boundaries and etc
- # [15:57] <@davidb> they have incorrect ones really
- # [15:57] <@surkov> why?
- # [15:57] <@surkov> don't you "just" render them?
- # [15:57] <@davidb> right
- # [15:58] <@davidb> the user clicks on canvas renderings, which are to cause events on the bound element
- # [15:58] <@davidb> all we can do know is do hit testing without this important piece
- # [15:58] <@firebot> surkov.alexander@gmail.com cancelled review?(trev.saunders@gmail .com) for attachment 599959 on bug 473576.
- # [15:58] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 600898 on bug 473576.
- # [15:58] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=473576 nor, --, ---, surkov.alexander, ASSI, font-family text attribute should expose actual font used
- # [15:59] <@davidb> surkov: do you think we need to test events for canvas?
- # [15:59] <@surkov> davidb: ok, that makes sense
- # [15:59] <@surkov> maybe :)
- # [15:59] <@davidb> heh
- # [15:59] <@surkov> it's better to ask Jamie
- # [15:59] <@davidb> it is pre-emptive test coverage. i think i'll do a little.
- # [16:00] <@surkov> I feel that hit testing is quite different from click
- # [16:00] <@davidb> Jamie says no to everything :)
- # [16:00] <@surkov> and he's right :)
- # [16:00] <@surkov> usually
- # [16:00] <@davidb> heheh
- # [16:01] <@davidb> surkov: so you think i can just remove the test_recreation changes?
- # [16:01] <@davidb> and not bother with test_gencontent?
- # [16:01] <@surkov> iirc they don't make any specifics for canvas content
- # [16:02] <@davidb> surkov: well test_recreation failed for one iteration of my canvas work.
- # [16:02] <@davidb> but that doesn't mean i need to test it now i guess
- # [16:02] <@davidb> now that we have reflow impl.
- # [16:03] <@surkov> hm, it would be interesting to know which and why
- # [16:03] <@surkov> (put that into bug please)
- # [16:03] <@davidb> it was because i hadn't implemented reflow
- # [16:03] <@surkov> I wonder if tree update testing covers that
- # [16:03] <@davidb> it should
- # [16:03] <@surkov> or tree
- # [16:03] <@surkov> ok
- # [16:04] <@davidb> surkov: would you prefer test_canvasupdate.html or just add to existing files?
- # [16:04] <@davidb> i don't really care :)
- # [16:05] <@davidb> i mean either seems good
- # [16:06] <@surkov> davidb: it's ok to keep it as separate file but you don't need world update if you are in tree update I guess
- # [16:06] <@davidb> surkov: world/word? ok.
- # [16:06] <@surkov> word
- # [16:06] <@davidb> alrighty.
- # [16:16] <@davidb> surkov: i guess for now (phase 1) i could just add the visibility tests and nothing else - sound ok?
- # [16:18] * @davidb looks at the various tree update tests
- # [16:20] <@surkov> davidb: I'm not aware about phases, I refer to accessible tree. How phase # after 1 affects on boundaries and hit testing?
- # [16:20] <@davidb> we'll want to report the boundaries of a canvas draw path for the corresponding element.
- # [16:21] <@davidb> i don't mind adding a test that will break for phase 2.
- # [16:22] <@surkov> davidb: what if draw path is not used, if I get right that's a11y extension?
- # [16:22] <@davidb> right
- # [16:23] <@surkov> should we expose them with 0 rectangle or provide some from frame tree?
- # [16:23] <@davidb> we shouldn't expose 0 rect.
- # [16:24] <@davidb> surkov: test_gencontent
- # [16:24] <@davidb> oops
- # [16:24] <@davidb> surkov: www.w3.org/wiki/Canvas_hit_testing
- # [16:25] <@davidb> this is just a proposal ^
- # [16:26] <@surkov> ok, then skip testing but that proposal is crazy
- # [16:29] <@davidb> heheh
- # [16:29] <@davidb> it is from Frank (Microsoft)
- # [16:30] <@MarcoZ> That doesn't mean it can't be crazy. ;-)
- # [16:32] <@MarcoZ> Oh man, I've never had so many problems with a flight before. Nobody seems to be able, or responsible for, entering my dog into the booking for the last two flights from Houston to Frankfurt to Hamburg. I suspect it's the United/Continental merger chaos paired with general Star Alliance inflexibilities.
- # [16:33] <@surkov> that's so crazy
- # [16:34] <@davidb> MarcoZ: what would happen if Hexe wasn't pre-checked-in?
- # [16:34] <@davidb> err booke
- # [16:34] <@davidb> booked
- # [16:34] * Quits: nhirata (nhirata.bu@moz-2A9C9106.hsd1.ca.comcast.net) (Quit: nhirata)
- # [16:36] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [16:36] <@MarcoZ> davidb: I honestly don't know. It happened once with Air Canada that somehow the information wasn't entered even though I had called them, but we sorted it out just fine. This was my first Toronto visit in 2010.
- # [16:37] <@MarcoZ> So I think it's generally not a problem, but they can make one out of it if they so choose.
- # [16:37] <@MarcoZ> I guess.
- # [16:37] <@davidb> they would be fools
- # [16:38] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [16:38] <@MarcoZ> davidb: Yeah, I can make a lot of noise if I have to.
- # [16:38] <@davidb> heheh
- # [16:41] <@davidb> surkov: one thing I would like to see, is some structure introduced to how people use canvas, and for that structure to be tied to what is draw on canvas. if we can make decent hit testing, then maybe game developers can benefit too, and end up building accessible games. i dream.
- # [16:43] <@surkov> DOM games must be something new
- # [16:44] <@davidb> right, anything non-trivial anyways
- # [16:45] * @davidb is seeing complaints about mixed line endings in some test files
- # [16:50] <@MarcoZ> davidb: surkov: If you think I talk too much about what our thoughts are re b2g, just tell me to stop. But I think that's the way to go really, use the good stuff we have and make it talk in b2g ourselves, not rely on a 3rd party SR, as we already thought of in the past.
- # [16:50] <@davidb> MarcoZ: bug #?
- # [16:50] <@MarcoZ> davidb: Twitter.
- # [16:50] <@davidb> ok
- # [16:51] <@surkov> MarcoZ: nowadays everybody talk much about b2g :)
- # [16:51] <@davidb> heheh
- # [16:52] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
- # [16:52] <@surkov> MarcoZ: which APIs do you mean?
- # [16:53] <@MarcoZ> surkov: Well, APIs may not be the right term here, but in a way yes: Write correct HTML, and it'll be accessible. In a way, that's an API, too.
- # [16:53] <@surkov> MarcoZ: on b2g? who's gonna to make it "accessible"?
- # [16:53] <@MarcoZ> surkov: Us.
- # [16:53] <@MarcoZ> surkov: Mozilla.
- # [16:54] <@surkov> MarcoZ: do we want to be os x?
- # [16:54] <@MarcoZ> surkov: I think we have to be sort of, since we also control the devices to some extent.
- # [16:54] <@MarcoZ> "we" being Mozilla.
- # [16:54] <@surkov> so no 3d party screen readers?
- # [16:54] <@davidb> we are the os
- # [16:55] <@MarcoZ> surkov: No. I don't think so.
- # [16:55] <@MarcoZ> davidb: Right!
- # [16:55] <@surkov> why?
- # [16:55] <@MarcoZ> So in a way, yes it is like iOS, but open-sourced.
- # [16:55] * @surkov thought it's open not just open source
- # [16:56] <@surkov> so everybody can create apps he wants
- # [16:56] <@MarcoZ> Yes, it's open, anyone is welcome to hack on it and provide patches like to the rest of Gecko.
- # [16:56] <@davidb> b2g is the web
- # [16:56] <@davidb> as an os
- # [16:56] <@surkov> b2g is not the web if you don't allow 3d party apps
- # [16:57] <@davidb> web apps
- # [16:57] <@davidb> will run
- # [16:57] <@MarcoZ> surkov: The only languages "apps" are written in in b2g are HTML and JS.
- # [16:57] <@surkov> MarcoZ: contribution to open project is quite different from openess of os
- # [16:57] <@MarcoZ> surkov: To give the users a good accessible experience right from the start, we *have* to create something ourselves and deliver it with b2g. otherwise, we're as bad as Android is. And I don't want that.
- # [16:57] <@surkov> MarcoZ: fine with that but the questions is why 3d party apps like screen readers are denied?
- # [16:58] <@surkov> MarcoZ: sure we have, but question is different
- # [16:58] <@surkov> we have -> we have to
- # [16:58] * khuey is now known as khuey|away
- # [16:59] <@tbsaunde> yeah, I'm going to agree with surkov here, providing a screen reader we should probably do, but we shouldn't prevent other people writing there own if they like
- # [16:59] <@MarcoZ> surkov: I'm not sure it's going to be possible to allow a web app such dep integration to deliver screen reading services. Davidb, what do you think?
- # [16:59] <@davidb> interesting discussion.
- # [16:59] <@surkov> MarcoZ: web apps will be allowed to dial phone numbers, they are allowed to do almost everything
- # [16:59] <@MarcoZ> Indeed!
- # [16:59] <@surkov> except being a screen reader
- # [16:59] <@surkov> that's quite strange
- # [17:00] <@MarcoZ> surkov: That's not set in stone. Those were just my thoughts.
- # [17:00] <@davidb> the parts we know: we have to provide the ingredients for tts, etc, in b2g and expose to web content.
- # [17:00] <@surkov> davidb: I think we should work on a11y api for the web
- # [17:00] <@davidb> surkov: i was thinking about that
- # [17:00] <@surkov> because nobody should invent the wheel
- # [17:00] <@tbsaunde> if we can provide one what would be the technical problem with allowing others? other than security model?
- # [17:01] <@davidb> surkov: critical to that working is people building web apps based on libraries that implement js a11y server
- # [17:01] <@MarcoZ> surkov: The wheel has to be invented for this, since nobody has ever written a screen reader for a fully web-based phone interface before. :)
- # [17:01] <@surkov> tbsaunde: we shouldn't have any problems because ti shouldn't be different from other web apps apis
- # [17:01] <@davidb> if anyone can do it, we can.
- # [17:01] <@MarcoZ> 5
- # [17:01] <@surkov> MarcoZ: the question how many of wheels
- # [17:02] <@tbsaunde> surkov: yeah, it was meant as a retorical question
- # [17:02] <@surkov> ok
- # [17:02] <@davidb> tbsaunde: hi!
- # [17:02] <@tbsaunde> MarcoZ: it could probably be a port of nvda for large values of "port"
- # [17:02] <@tbsaunde> davidb: hi
- # [17:03] <@tbsaunde> large values such as say rewriting in the assembly of the web
- # [17:03] <@davidb> eeejay has to be part of this discussion.
- # [17:03] <@MarcoZ> tbsaunde: Oh yeah, it would not only be a port in APIs, but a port in languages, too, since NVDA, like Orca, is written in python. :)
- # [17:04] <@davidb> python to js should be trivial
- # [17:04] <@davidb> i also think google should be part of this discussion
- # [17:04] <@MarcoZ> surkov: Again, I wasn't saying that we shouldn't do it, or cannot do it, I was just being cautious. And I am strongly for providing one ourselves since that's the only way we can guarantee a fully accessible experience right from the point where the user switch on their b2g phone for the very first time.
- # [17:05] <@davidb> yeah
- # [17:05] <@MarcoZ> davidb: Why Google?
- # [17:06] <@davidb> MarcoZ: because they might have thought of dead ends
- # [17:07] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [17:07] <@davidb> we don't need them in the discussion yet
- # [17:08] <@MarcoZ> OK guys, I've got to close down and really start packing. I'll be sporadically reachable over e-mail in the next 36 hours, and back fully online when I'm in San Diego.
- # [17:09] <@davidb> ok!
- # [17:09] <@davidb> safe travels
- # [17:09] <@MarcoZ> Thanks!
- # [17:09] * Quits: @MarcoZ (marco.zehe@moz-1907351C.dip.t-dialin.net) (Quit: San Diego, here I come!)
- # [17:29] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [17:32] * khuey|away is now known as khuey
- # [17:42] * Joins: hub (hub@83874EA1.EB7C1AF9.6F478678.IP)
- # [17:42] * ChanServ sets mode: +o hub
- # [17:44] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 600926 on bug 728907.
- # [17:44] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=728907 nor, --, ---, surkov.alexander, ASSI, don't use GetComputedStyle for text attributes
- # [18:04] <@tbsaunde> surkov: btw the guy on the orca list complaining about thunderbird starting slowely is I believe the same issue of us calling atk_object_set_parent() and causing an unneeded event
- # [18:05] <@surkov> tbsaunde: sure, we need to figure out solution
- # [18:05] <@surkov> I don't know much about atk unfortunately
- # [18:05] <@surkov> I guess we need some help
- # [18:05] <@surkov> otherwise it's hard to say for me whether the proposed patch is correct
- # [18:07] <@tbsaunde> ok, do you have objections to a patch just to RefChildAt() and not trying to fix anything else?
- # [18:09] <@tbsaunde> surkov: the only thing I can think of that patch could possibly hurt is adding an extra branch
- # [18:10] <@surkov> tbsaunde: extra branch?
- # [18:10] <@surkov> I don't think anything bad if you add a check of parent before setting it
- # [18:11] <@surkov> if that helps to the problem then let's take it, but it doesn't seem to go with bug description
- # [18:11] <@tbsaunde> surkov: if we have if (atkObj->parent != parent) atk_set_parent(parent); instead of just atk_object_set_parent(parent) the code needs to test if the pointers are equal and branch on that no/
- # [18:11] <@tbsaunde> yeah, I should probably just do that in its own bug
- # [18:11] <@surkov> ah, what's bad in this, I assume atk_object_set_parent does some extra work anyway
- # [18:12] <@surkov> tbsaunde: then please
- # [18:13] <@tbsaunde> its a tiny deal, this code is not particularly optimized so worrying about a single branch is probably obserd
- # [18:15] <@firebot> New Core - Disability Access APIs bug 730841 filed by trev.saunders@gmail.com.
- # [18:15] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=730841 nor, --, ---, trev.saunders, NEW, call atk_object_set_parent() less
- # [18:17] <@tbsaunde> surkov: inted part of the method to two spaces or leve it all at 4?
- # [18:18] <@surkov> 2 please
- # [18:18] <@surkov> ugly but what we can do :)
- # [18:19] <@tbsaunde> ok, coming up
- # [18:19] <@tbsaunde> surkov: well, I could reindent all of it
- # [18:20] <@surkov> if it's not too big, up to you
- # [18:24] <@firebot> trev.saunders@gmail.com requested review from surkov.alexander@gm ail.com for attachment 600933 on bug 730841.
- # [18:24] <@tbsaunde> surkov: ok, there you go, I've go to run to class
- # [18:24] <@surkov> see ya!
- # [18:28] <@eeejay> morning
- # [18:29] <@eeejay> could someone help me understand rpbl.m.o?
- # [18:30] <@eeejay> Bug 729237 was backaed out and i am trying to figure out the failure
- # [18:30] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=729237 nor, --, ---, nobody, NEW, Re-bridge emitGeckoAccessibilityEvent
- # [18:34] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [18:35] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [18:36] <@firebot> surkov.alexander@gmail.com granted review for attachment 600933 on bug 730841.
- # [18:36] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=730841 nor, --, ---, trev.saunders, NEW, call atk_object_set_parent() less
- # [18:38] <@surkov> eeejay: maybe you should ask mobile team, I don't know android java code
- # [18:40] <@eeejay> davidb, hub, are you available for a call?
- # [18:40] <@davidb> eeejay: yes
- # [18:41] <@eeejay> surkov, i don't really understand the procedure, does it mean that is was certainly my patch or could have been any one of a bunch of patches?
- # [18:42] <@surkov> eeejay: it could be either option, they just suspected your patch
- # [18:42] <@eeejay> hm
- # [18:42] <@surkov> did you try the try server builld?
- # [18:43] <@surkov> anyway try another try server build for your patch and see if there are failures
- # [18:43] <@surkov> if not then land it again
- # [18:46] <@davidb> eeejay: what about?
- # [18:46] <@eeejay> davidb, rainer pinged us about video
- # [18:46] <@eeejay> wants to get on a call
- # [18:47] <@davidb> yep
- # [18:47] <@eeejay> didn't get that?
- # [18:47] <@davidb> yes, just wanted to be sure what you're talking about :)
- # [18:52] * @davidb hangs in vidyo
- # [18:53] <@eeejay> i'll be there in a sec
- # [18:54] * Joins: drexler (chatzilla@moz-BD26ECCD.hsd1.vt.comcast.net)
- # [19:15] <@hub> eeejay: I don't have my headset, I was about to go to the office for that
- # [19:15] * Joins: victorporof (victorporo@986FC73E.7E2AA079.79933D60.IP)
- # [19:15] <@hub> c-ya,
- # [19:16] <@davidb> hub: we already chatted - we'll catch you up
- # [19:16] * Quits: @hub (hub@83874EA1.EB7C1AF9.6F478678.IP) (Ping timeout)
- # [19:20] * Quits: @jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
- # [19:24] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [19:24] * ChanServ sets mode: +o jprmc
- # [19:33] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [19:33] * ChanServ sets mode: +o hub
- # [19:37] <@hub> davidb: loging into vidyo, but it ask me to upgrade.
- # [19:37] <@hub> *sigh*
- # [19:37] <@davidb> yeah
- # [19:38] <@hub> davidb: your or a11y?
- # [19:38] <@davidb> me
- # [19:44] <@tbsaunde> eeejay: re tbpl or the backout comment probably poitns to logs you could red them
- # [19:45] * Quits: @surkov (surkov@D5570440.C1645F5A.34044A7F.IP) (Quit: surkov)
- # [19:45] <@tbsaunde> or if it only links to tbpl and that's still available it should say at the top what test suite failed and you can click on it and go to the bottum and there's some information and links to full logs
- # [19:53] <@davidb> hub: if you end up going to LATAM, we'll have to roadmap out our a11y story on b2g. It'll be a forcing function.
- # [19:53] <@davidb> (http://news.cnet.com/8301-17939_109-57385616-2/telefonica-mozillaphone-is-ten-times-cheaper-than-an-iphone/)
- # [19:55] <@hub> I was reading that article
- # [19:55] <@hub> yes we would
- # [19:57] <@hub> eeejay: btw I got a mike yesterday
- # [19:58] <@davidb> lol
- # [19:58] <@davidb> sorry
- # [19:58] <@davidb> i am amused since you didn't write mic
- # [19:59] <@eeejay> hehe
- # [20:00] <@eeejay> hub, cool, what kind?
- # [20:01] <@eeejay> that is a really exciting article
- # [20:02] * Quits: sgimeno (santiago.g@A922400D.48362F18.BCC03455.IP) (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
- # [20:05] <@davidb> yeah
- # [20:07] <@hub> azden
- # [20:07] <@hub> http://www.azdencorp.com/new/product.php?productid=93297&cat=22&page=1
- # [20:08] <@eeejay> cool, a shotgun. me too :)
- # [20:09] <@hub> with wide and narrow setting
- # [20:10] <@davidb> don't call it that at customs
- # [20:11] <@hub> nor at security
- # [20:11] <@davidb> right
- # [20:12] <@tbsaunde> you know it would be nice if I didn't have to send hg SIGINT to get it to finish
- # [20:20] <@eeejay> wow, i can't believe i only tried the try server now for the first time
- # [20:21] <@davidb> wuh?
- # [20:23] <@eeejay> yeah
- # [20:23] <@eeejay> it seems kind of scary
- # [20:23] <@eeejay> but it isn't
- # [20:27] <@hub> it is usefull
- # [20:27] <@hub> to catch breakage
- # [20:27] <@hub> but there are a lot of orange that are misleading :-/
- # [20:31] <@eeejay> i was afraid i was going to clobber the entire mozilla project forever
- # [20:33] <@davidb> eeejay: OMG you just destroyed everything!
- # [20:33] <@davidb> j/k
- # [20:33] <@eeejay> davidb, you made me check. just in case
- # [20:39] <@eeejay> Androidciser, for your android a11y testing needs: https://github.com/eeejay/Androidciser
- # [20:49] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [20:51] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [20:55] * khuey is now known as khuey|away
- # [20:58] * khuey|away is now known as khuey
- # [20:58] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [20:59] * @eeejay runs out for an errand
- # [20:59] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [21:01] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 13.0a1/20120226214756])
- # [21:13] <@davidb> what!?
- # [21:14] <@davidb> androidciser?
- # [21:14] * @davidb looks at code
- # [21:14] <@davidb> ok
- # [21:55] * Quits: @jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
- # [21:55] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [21:55] * ChanServ sets mode: +o jprmc
- # [22:26] <@firebot> bolterbugz@gmail.com requested review from surkov.alexander@gm ail.com for attachment 601034 on bug 495912.
- # [22:26] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=495912 nor, --, ---, bolterbugz, NEW, Expose alternative content in Canvas element to ATs
- # [22:26] <@firebot> bolterbugz@gmail.com cancelled review?(surkov.alexander@gm ail.com) for attachment 601034 on bug 495912.
- # [22:43] <@hub> what kind of accessible object am I supposed to have in the URL bar?
- # [22:43] <@hub> it is not an nsHyperTextAccessible, that's sure
- # [22:47] <@firebot> bolterbugz@gmail.com requested review from surkov.alexander@gm ail.com for attachment 601045 on bug 495912.
- # [22:47] <@davidb> i'm not sure
- # [22:47] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=495912 nor, --, ---, bolterbugz, NEW, Expose alternative content in Canvas element to ATs
- # [22:47] <@davidb> gotta run
- # [22:47] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [22:49] <@hub> I think it might just be enough to the a nsIAccessibleText
- # [22:49] <@hub> to get a
- # [23:01] * Quits: victorporof (victorporo@986FC73E.7E2AA079.79933D60.IP) (Quit: Linkinus - http://linkinus.com)
- # [23:02] <@tbsaunde> hub: I'm pretty sure the only things that implement nsIAccessibleText inherit from nsHyperTextAccessible
- # [23:02] <@tbsaunde> hub: try domi?
- # [23:02] <@tbsaunde> alternatively on linux I've written little python scripts to use pyatspi and you could probably get a good gues from that
- # [23:11] <@firebot> fmdevelopertim@gmail.com changed the Component on bug 730514 from Untriaged to Keyboard Navigation.
- # [23:11] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=730514 nor, --, ---, nobody, UNCO, auto-select for address bar, etc. is awfully annoying
- # [23:21] * Parts: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [23:22] <@hub> tbsaunde: we'll see if that works.
- # [23:52] <@firebot> trev.saunders@gmail.com granted review for attachment 600926 on bug 728907.
- # [23:52] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=728907 nor, --, ---, surkov.alexander, ASSI, don't use GetComputedStyle for text attributes
- # [23:58] <@firebot> trev.saunders@gmail.com granted review for attachment 600890 on bug 728907.
- # [23:59] <@tbsaunde> hub: btw what are you working on?
- # Session Close: Tue Feb 28 00:00:00 2012
The end :)