=== API is now known as Guest66753 | ||
=== Guest66753 is now known as apinheiro | ||
AlanBell | ev is going to sort out the ubiquity breakage, which will mean the text of the installer will all change | 11:24 |
---|---|---|
AlanBell | Cheri703: maco: is the revised text all in place? | 11:24 |
AlanBell | the patch is in trunk to un-break it already | 11:26 |
AlanBell | so the live CD installer should be different tomorrow | 11:27 |
AlanBell | TheMuso: did you find out where to patch the orca pronunciation dictionary? | 11:27 |
maco | Not yet. I have cheri's changes so I'll apply them | 14:37 |
maco | Are we past translation freeze already though? | 14:38 |
maco | Alanbell ^ | 14:38 |
AlanBell | will check | 14:40 |
maco | Because the accessible strings will need to be translated too | 14:44 |
maco | Probably should check the strings with themuso too | 14:45 |
=== API is now known as Guest69001 | ||
=== Guest69001 is now known as apinheiro | ||
AlanBell | maco: some of them are so bad they are bugs rather than translations | 15:03 |
AlanBell | like referencing 11.04 in the partitioning options | 15:03 |
AlanBell | or the option to replace windows | 15:04 |
AlanBell | bug 817416 | 15:54 |
ubot2 | Launchpad bug 817416 in unity-2d "[launcher] contextual menus are not spoken by the Orca screen reader" [High,Fix released] https://launchpad.net/bugs/817416 | 15:54 |
AlanBell | o/ charlie-tca | 18:22 |
charlie-tca | yeah, here | 18:22 |
AlanBell | there has been some stuff happening that looks suspiciously like progress :) | 18:22 |
AlanBell | bug 781385 | 18:23 |
ubot2 | Launchpad bug 781385 in ubiquity "Ubiquity GTK should have useful accessible names set in the Glade .ui files instead of using the variable names" [Medium,Fix committed] https://launchpad.net/bugs/781385 | 18:23 |
AlanBell | fix committed :) | 18:23 |
charlie-tca | Great! | 18:23 |
charlie-tca | Guess pushing helped then | 18:23 |
charlie-tca | Once in a while, I get one right? | 18:23 |
AlanBell | yeah, I had a word with ev | 18:24 |
charlie-tca | Thank you | 18:24 |
AlanBell | basically a reminder | 18:24 |
charlie-tca | I never knew how hard it can be to determine which bugs are going to be the most important. | 18:24 |
AlanBell | yeah, that one is going to uncover some more ones | 18:24 |
AlanBell | and right now is a good time to get a11y bugs worked on | 18:25 |
AlanBell | which it shouldn't be | 18:25 |
AlanBell | but at least things are getting testable now | 18:25 |
charlie-tca | Yeah, We really should not have to wait until the final week to get things fixed | 18:26 |
charlie-tca | which reminds me. I am going to send a response to the uds-p discussion about a11y | 18:27 |
Pendulum | I wonder if that's a suggestion for something to get discussed at UDS "making a11y part of the process"? | 18:27 |
Pendulum | haha | 18:27 |
charlie-tca | Maybe we can get more people involved | 18:27 |
Pendulum | I was just thinking about that | 18:27 |
charlie-tca | yeah, somehting along those lines | 18:27 |
AlanBell | there have been quite a few discussions now about onboard and unity | 18:28 |
AlanBell | that is a significant architectural and standards fail | 18:28 |
charlie-tca | Yes, but about a few discussions about actually being allowed to make accessibility work for P ? | 18:28 |
AlanBell | the windowing system just doesn't have a signal for a window to say it really needs to stay on top, because nobody spotted it as an issue at the right time | 18:29 |
AlanBell | but that one I think should be referred to as a touchscreen issue, or a soft keyboard issue, not an a11y specific problem | 18:30 |
maco | Pendulum: will you be at uds? | 18:31 |
charlie-tca | If it is not a11y problem, it will get pushed aside, almost to wishlist | 18:31 |
maco | charlie-tca: i think AlanBell's suspicion is that they care more about tablets than about a11y | 18:32 |
charlie-tca | Maybe right, too | 18:32 |
Pendulum | maco: nope. wasn't sure if I'd be healthy enough when sponsorship was open and now can't afford it | 18:32 |
charlie-tca | Okay, I will go along with that idea, then | 18:32 |
Pendulum | OTOH, didn't Mark say on that thread specifically that Unity wasn't aimed at tablets? | 18:33 |
* charlie-tca will be there, trying to push a11y again... :) | 18:33 | |
* Pendulum will be pushing a11y on remote participation | 18:33 | |
charlie-tca | Pendulum: I am pretty sure they are getting tired of hearing from me now | 18:33 |
AlanBell | maco: not quite, I know it is going to be a PITA to fix, I would rather they thought they were doing something that inconvenient for tablets than a11y | 18:34 |
maco | i might sneak into uds for a11y sessions | 18:39 |
maco | turns out the train is within my budget, so i figure couchsurfing is doable | 18:39 |
Pendulum | :) | 18:40 |
maco | AlanBell: i'm going to commit cheri's a11y updates now. it involves removing the label-setting code from the partitioner page but the stuff in her patch *should* obsolete that anyway. i guess we'll find out with tomorrow's build how that works | 20:46 |
AlanBell | cool | 20:47 |
AlanBell | is that in lp:unity or lp:ubuntu/unity? | 20:47 |
maco | ubiquity | 20:48 |
maco | lp:ubiquity | 20:48 |
maco | oh hm right yes that may not do anything to the build | 20:48 |
maco | unless....hm no its too late to bug cjwatson to package the branch :-/ | 20:48 |
maco | ok fine i guess we'll find out the day after tomorrow | 20:48 |
maco | (by pestering cjwatson tomorrow) | 20:51 |
maco | Cheri703: your patch is committed | 20:51 |
Cheri703 | cool! :) I hope I did it right >.> | 20:51 |
maco | it looked reasonable to me | 20:51 |
maco | and you used glade, so syntax bugs are not a worry. i need to update the other .ui files by hand since glade falls over on them. in that case, syntax bugs are a possible eek | 20:52 |
maco | AlanBell: what is it that says 11.04 incorrectly? | 20:56 |
maco | AlanBell: nevermind. i see your bug comment | 21:00 |
maco | AlanBell: ok that's fixed now too | 21:02 |
AlanBell | yay, thanks | 21:02 |
AlanBell | what do you think about the replace windows one? | 21:03 |
AlanBell | http://bazaar.launchpad.net/~ubuntu-installer/ubiquity/trunk/view/head:/gui/gtk/stepPartAsk.ui#L327 | 21:06 |
AlanBell | it asked me if I would like to replace windows with ubuntu and I didn't have windows at all | 21:07 |
maco | doh | 21:07 |
maco | i'd file a bug on that | 21:07 |
AlanBell | Replace the current operating system with Ubuntu would be more accurate | 21:07 |
maco | because that sounds like some OS-prober stuff should make tha label change | 21:07 |
maco | hmm that would also work | 21:07 |
maco | i was just going "wait does it ask Mac users that too??" | 21:07 |
AlanBell | it would do I expect | 21:08 |
AlanBell | bug 860051 | 21:11 |
ubot2 | Launchpad bug 860051 in ubiquity "radio buttons in partitioning page of installer mentions windows" [Undecided,New] https://launchpad.net/bugs/860051 | 21:11 |
Cheri703 | for the record, if anyone else has easy "type things into boxes" things that need to be done for patches and whatnot, I may be able to help :) (other types of stuff not as much) | 21:19 |
maco | AlanBell: the .desktop for "Universal Access".... can you run dpkg -S on it? | 21:35 |
AlanBell | if I could boot my oneiric VM I would | 21:36 |
AlanBell | ah, if firefox gives me back a couple of gigs of memory maybe a VM will start | 21:37 |
maco | it shouldnt have changed since natty...i think | 21:38 |
maco | its just that im on kde | 21:38 |
maco | so i dont have it installed | 21:38 |
AlanBell | gnome-universal-access-panel.desktop is in gnome-control-center | 21:44 |
maco | thank ye kindly | 21:44 |
AlanBell | and that isn't in natty | 21:44 |
maco | oh boo | 21:44 |
maco | oh right. gnome did a lot of kersplat to gnome-control-center | 21:44 |
AlanBell | gersplat | 21:45 |
AlanBell | you need to get out of that K way of thinking :) | 21:45 |
maco | haha | 21:45 |
AlanBell | so unity isn't searching the keywords is it | 21:47 |
maco | i dont think there are keywords | 21:48 |
AlanBell | there are squillions of them | 21:49 |
maco | but a keywords extension to the desktop spec would be nice given gnome kde unity all have searchy launchers | 21:49 |
maco | hmm? | 21:49 |
maco | h/o im looking at a bug for skaet right now | 21:49 |
AlanBell | X-GNOME-Keywords=Keyboard;Mouse;a11y;Accessibility;Contrast;Zoom;Screen Reader;text;font;size;AccessX;Sticky Keys;Slow Keys;Bounce Keys;Mouse Keys | 21:51 |
AlanBell | http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/oneiric/gnome-control-center/oneiric/view/head:/panels/universal-access/gnome-universal-access-panel.desktop.in.in#L17 | 21:51 |
maco | go figure | 21:52 |
maco | i wonder if caps counts... | 21:52 |
AlanBell | which bug? | 21:52 |
maco | bu 852583 | 21:53 |
maco | bug 852583 | 21:53 |
ubot2 | Launchpad bug 852583 in gnome-orca "Orca does not pronounce "Oneiric Ocelot" very well" [High,Confirmed] https://launchpad.net/bugs/852583 | 21:53 |
AlanBell | :) | 21:53 |
maco | so for the keywords, Name and Comment are being searched but not X-GNOME-Keywords, it seems | 21:59 |
maco | https://bugs.launchpad.net/unity/+bug/828356 | 22:00 |
ubot2 | Ubuntu bug 828356 in unity "Support X-GNOME-Keywords" [Undecided,New] | 22:00 |
AlanBell | that should be pretty easy to fix | 22:01 |
* maco fetches a unity branch | 22:03 | |
AlanBell | it would be in the apps lens | 22:06 |
AlanBell | at the point the names all get flung into the search engine thing the keywords need to go in too | 22:09 |
maco | AlanBell: skaet marked that one as high | 22:13 |
AlanBell | yay | 22:16 |
AlanBell | and it is a "fix the a11y reported bug and *lots* of other bits get better" type of bug | 22:17 |
AlanBell | maco: did you mean to remove the rls-mgr-o-tracking tag? | 22:18 |
maco | no? | 22:19 |
maco | there wasnt one there when i clicked the edit | 22:19 |
AlanBell | I think you and skaet argued over the edit :) | 22:19 |
* skaet goes and puts tag back ;) | 22:19 | |
* maco does a test build on espeak | 22:23 | |
AlanBell | maco: are you fixing it in espeak? | 22:23 |
maco | AlanBell: yes | 22:23 |
maco | is that not the right way? | 22:23 |
AlanBell | great, better place to do it than in orca | 22:23 |
maco | oh ok | 22:23 |
maco | yeah espeak has dictionaries for each language | 22:24 |
AlanBell | perfect, you can override stuff in orca, but yes, espeak is the right place if you can do that :) | 22:24 |
maco | if you look in espeak's source package, there is a dictsources/ | 22:24 |
maco | in there theres a **_rules and **_list for each language | 22:25 |
maco | if you dont want to make for evil merges, you add things by making a **_extra | 22:25 |
AlanBell | spd-say kubuntu | 22:25 |
AlanBell | spd-say coobuntu | 22:25 |
maco | Ubuntu's in the _list file | 22:25 |
maco | kubuntu sounds like queuebuntu | 22:26 |
AlanBell | it does | 22:26 |
AlanBell | spd-say cuh-buntu | 22:26 |
maco | the syntax you're using in spd-say isnt anything like what goes in the files unfortunately | 22:26 |
maco | here's what im attempting for oneiric ocelot | 22:27 |
maco | oneiric oUn'i@3@k | 22:27 |
maco | ocelot 0sEl0t | 22:27 |
maco | doh | 22:27 |
AlanBell | gosh | 22:27 |
maco | that shouldve been oUn'i:r@k | 22:27 |
maco | @ is for a schwa sound | 22:28 |
meetingology | maco: Error: "is" is not a valid command. | 22:28 |
maco | haha | 22:28 |
TheMuso | AlanBell: Yeah I started digging into the code yesterday to do that, and I found it, however I found another bug that I needed to work with upstrea to fix. Now that is addressed, I will be doing the work today. | 22:30 |
AlanBell | TheMuso: maco is doing it now in espeak | 22:30 |
TheMuso | Actually espeak would be a better location, yes. | 22:31 |
maco | my new version of espeak is building now | 22:31 |
TheMuso | maco: Ok cool., | 22:31 |
TheMuso | Oh and plesae don't screw with the ubuntu words as they are pronounced correctly IMO. | 22:31 |
maco | it sounds odd to me for the glide to be there. like kyu instead of ku | 22:32 |
maco | but i wont touch unless im told to | 22:33 |
AlanBell | the ubuntu part of kubuntu sounds very different to the hinted plain ubuntu | 22:35 |
TheMuso | fair enough | 22:36 |
AlanBell | maco: are you using espeakedit? | 22:39 |
maco | no? | 22:39 |
AlanBell | !info espeakedit | 22:40 |
maco | !find espeakedit | 22:40 |
* maco glares at the bots | 22:40 | |
AlanBell | it is a phoneme editor thing | 22:40 |
AlanBell | and it doesn't work | 22:41 |
AlanBell | Wrong version of espeak-data at: | 22:42 |
AlanBell | /home/alan/espeak-data | 22:42 |
AlanBell | Version 0x14404 (expects 0x14300) | 22:42 |
AlanBell | didn't mean to paste that here, but anyhow, that is the error message I am filing a bug about | 22:43 |
TheMuso | espeakedit is an older version | 22:47 |
AlanBell | yeah, doesn't match the data file | 22:47 |
TheMuso | Right | 22:50 |
AlanBell | http://espeak.sourceforge.net/test/latest.html there is a new upstream version | 22:55 |
AlanBell | bug 860118 | 23:03 |
ubot2 | Launchpad bug 860118 in espeakedit "Wrong version of espeak-data at: ~/espeak-data Version 0x14404 (expects 0x14300)" [Undecided,New] https://launchpad.net/bugs/860118 | 23:03 |
maco | AlanBell, TheMuso: http://people.ubuntu.com/~maco.m/oneiric_ocelot.mp3 | 23:04 |
TheMuso | maco: eeeeeeeeviiiil | 23:05 |
maco | :(( | 23:05 |
maco | why evil? | 23:05 |
AlanBell | pretty good I think | 23:05 |
TheMuso | Oslot does not sound correct to me.... | 23:05 |
AlanBell | bit oscilot rather than oscelot | 23:05 |
maco | mmk lemme mess with that one some more | 23:06 |
AlanBell | *masses* better than one irick oh slot as it was before :) | 23:08 |
maco | so do you mean it sounds like theres a i instead of a e? | 23:09 |
AlanBell | yes, a bit I think | 23:09 |
maco | hmrph i told it to use the e as in tech | 23:09 |
AlanBell | hmm, listening a few more times it sounds better | 23:09 |
AlanBell | I can't decide quite how it should sound | 23:12 |
maco | heh | 23:13 |
maco | for my accent, rhyme with lancelot | 23:13 |
maco | like an "uh" sound | 23:13 |
AlanBell | something like that | 23:13 |
maco | it doesnt sound much different to me when i give it the character for a schwa either | 23:13 |
maco | lets see whats available in the u section.... | 23:14 |
AlanBell | could be more of a sort "o" or possibly "er" sound | 23:15 |
AlanBell | ocolot or ocerlot | 23:15 |
AlanBell | or ocalot | 23:15 |
maco | the er would be distinctly british, to my ear | 23:15 |
maco | i notice r sounds in a lot of places in brits' words where there's not an r written :P | 23:16 |
* AlanBell speaks the Queen's English | 23:19 | |
AlanBell | anyhow I am off to bed now, thanks maco | 23:20 |
maco | good night | 23:20 |
maco | should i make a patch of this version or keep poking? | 23:20 |
TheMuso | I'll let you guys decide, because I am happy to live with whatever espeak says now, but I can understand the thought that it shoudl be different. | 23:27 |
* AlanBell says patch it maco (and goes to bed for real this time) | 23:29 | |
maco | i think i got it to do something a bit closer to correct now | 23:30 |
maco | but my phone's voice recorder doesnt seem able to pick up the difference | 23:30 |
maco | -_- the debian/rules file doesnt include the command to rebuild the voices. | 23:36 |
maco | TheMuso: should i be manually compiling the voice then committing the changed voice? | 23:36 |
TheMuso | maco: If you need to rebuild the voices, rebuild them. | 23:37 |
maco | ok | 23:37 |
maco | just surprised that the build didnt do that | 23:37 |
TheMuso | Right. | 23:38 |
TheMuso | maco: And thanks for your work on ubiquity as well. I'm about to build packages from the bzr branch to test. | 23:38 |
maco | oh thanks | 23:39 |
maco | i still dont know how to do that myself | 23:39 |
maco | there are still more accessible names to fill in in the ui files, but at least now they wont be stomped over | 23:39 |
TheMuso | Right. | 23:40 |
maco | hm this is odd | 23:42 |
maco | if i put the recompiled voice into espeak-data/ (because it insists on always recompiling in /usr/share....) bzr doesnt pick up that the file has changed at all | 23:42 |
TheMuso | You can tell espeak that the datadir is elsewhere, check the manpage/ | 23:44 |
maco | ok | 23:45 |
TheMuso | maco: You might also want to back up the files that get rebuilt, and restore them on package clean. | 23:45 |
TheMuso | The clean target shoudl always get the package back to the state it is in when its unpacked. | 23:45 |
maco | the bzr branch contains the **_dict compiled voices, so i guess they're there when the package is unpacked too | 23:46 |
TheMuso | I don't work with the bzr branch. | 23:46 |
TheMuso | I maintain this with the pkg-a11y team in debian using git. | 23:46 |
TheMuso | But whatever works. | 23:46 |
maco | i just did an apt-get source and they're in there when i unpack too. so yeah, those binary-ish dictionaries are shipped precompiled | 23:47 |
maco | which seems a little odd to me | 23:47 |
maco | i can just do it as a debdiff if you prefer | 23:48 |
maco | either way itll be funny lookin with the binaryish thing being modified | 23:48 |
TheMuso | debdiff works. | 23:48 |
TheMuso | those pre-compiled files came in the upstrea zip file. | 23:48 |
maco | i need to learn not to take ibuprofen on an empty stomach. heartburn is not pleasant | 23:55 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!