05:01:01 #startmeeting i18n 05:01:01 Meeting started Thu Jun 30 05:01:01 2011 UTC. The chair is tagoh. Information about MeetBot at http://wiki.debian.org/MeetBot. 05:01:01 Useful Commands: #action #agreed #halp #info #idea #link #topic. 05:01:06 #meetingname i18n 05:01:06 The meeting name has been set to 'i18n' 05:01:15 #topic agenda and roll call 05:01:28 https://fedoraproject.org/wiki/I18N/Meetings/2011-06-30 05:01:47 hi, shall we have weekly i18n meeting 05:01:56 who else here? 05:02:10 hi 05:03:07 hi 05:03:17 hi 05:04:11 hi 05:04:54 hi 05:05:13 hi 05:05:29 okay, let's get started then 05:05:37 #topic F16 05:07:02 well, the feature submission deadline is coming within about 2 weeks. so guess that may be good to go? 05:07:24 is there anything else we'd propose for f16? 05:07:33 hm 05:09:05 hi 05:09:31 should we discuss with Input Integration feature more? 05:09:33 juhp: btw I'm still thinking to get rid of imsettings integration to gnomecc or siply add it as we planned in f15. what do you think? 05:09:38 s/with/the 05:10:01 tagoh: for f16 or forever? :) 05:10:16 at least for f16. 05:10:38 okay 05:10:46 any chance of an update for f15? 05:11:11 tagoh: can you update the page for that then? 05:11:29 it sounds a bit too late to start negotiating upstream for integration for f16 05:11:33 juhp: sure 05:11:35 will do 05:11:43 true 05:12:06 but still maybe worth trying 05:12:22 anything else concerns to push this feature proposal? 05:12:30 even if it doesn't land now 05:12:47 juhp: yep. good to start now for future release 05:12:53 ok 05:14:01 fujiwarat, dueno: do you have anything for this feature? 05:14:17 Yesterday I updated ibus packages for f15 and it's pretty good. Probably I can start the discussion with upstream soon. 05:15:14 I separated xkb group layouts by engines and now input method and keyboard layouts are shown in ibus menu. 05:16:03 Now ibus menu also has the item of 'g-c-c- region'. 05:17:04 okay 05:18:48 aha - sounds good 05:18:54 going to try it 05:19:14 juhp: Now I have one problem about the new Control+Space. 05:19:31 so we've discussed enough for feature and is it ready to go? anything else? 05:20:01 The new Control+Space switches engines so the properties are lost if they are not saved in gconf. 05:20:38 Previously the properties are lost when ibus-daemon restarts. 05:21:50 aha 05:22:09 fujiwarat, so do we need a property for last used engine? 05:22:22 * tagoh updated the feature page 05:22:29 tagoh: yeah, should we propose the feature? hmm 05:22:55 juhp: any concerns? 05:23:01 I guess we should show it to the desktop team first perhaps 05:23:16 aha 05:23:42 fujiwarat, do you think we can propose the Feature for f16? 05:23:45 juhp: I'm not sure if we think it's a bug or specification. 05:23:47 post to the desktop list or ? 05:24:00 fujiwarat, yeah maybe it is not that serious 05:24:11 but would be nice to fix 05:24:33 e.g. When we change Hiragana to Katakana, Ctrl+Space resets the setting. 05:24:36 tagoh: yeah I guess that would work 05:24:51 ah 05:25:13 ah "previously" 05:25:49 sounds like a bug to me 05:25:57 fujiwarat, so fixable? 05:26:09 juhp: Hmm.., probably I would need the first feedback from upstream about f16. 05:27:15 fujiwarat, okay - so do you want to post something fedora desktop list to start the conversation, or ? 05:27:19 to 05:28:30 Ah, I'm thinking to contact irc with their time zone at first with updated the gnome bugzilla report. 05:29:57 fujiwarat, which bug? 05:30:33 https://bugzilla.gnome.org/show_bug.cgi?id=641531 05:30:36 fujiwarat, not sure if we are talking about the same :) 05:30:49 ok but what about our Feature page? 05:31:22 heh the bug is still unconfirmed ;) 05:31:45 fujiwarat, so it is not all upstream yet? 05:32:20 I thought, if the upstream agreed with the f16 time frame, probably we can update the feature page. 05:32:50 fujiwarat, okay sure 05:33:01 that would be good input indeed 05:33:58 guess we could get anything until next meeting? otherwise we might miss the deadline for feature submission 05:35:13 fyi, the feature submission deadline is July 12 05:35:29 fujiwarat, is that possible? 05:35:51 hmm. The submission is important for us? 05:36:13 fujiwarat, it is important in the sense that it gives attention to ibus 05:36:41 fujiwarat, when do you think you might get an indication from upstream? 05:37:41 anyway let's see next week or so - if we feel confident we can proceed with the feature submission otherwise leave it i guess 05:37:44 Hmm.., one problem is the ibus discussion depends on the new gjs package which is not released yet. 05:38:01 fujiwarat, will it be in f16? 05:38:40 juhp: Yes, I think so. The bug is recently fixed. 05:39:49 OK, I will contact the upstream in this week. 05:39:55 fujiwarat, ok - we only need to commit to the feature at this point, delivery is a bit later 05:39:57 okay thanks 05:40:44 maybe also good to talk to the desktop team as well. they could help us if they think it's needed? 05:40:47 perhaps at the same time I could bring to the attention of the fedora desktop list for any comments 05:40:50 yes 05:41:31 okay, anything else on f16 topic? 05:41:46 if not, let's move on 05:42:38 #topic Input Methods 05:42:50 any updates on IM? 05:43:45 or does anyone have anything to continue the discussion about multilib issue? 05:45:54 nope? 05:45:56 hmm I don't do anything on the multilib - feel like I had a action item 05:46:14 actually wanted to bring up about firefox langpacks... 05:46:23 for f16 05:48:05 juhp: yep - it was to raise the issue to upstream or the desktop team iirc 05:48:15 heh :) 05:48:26 okay - let me note it down.... 05:48:53 juhp: any chance to get langpack on firefox? 05:49:05 though... hmm might need to re-read last week's log first.... 05:49:17 don't remember the context/discussion well 05:49:29 well, actually the meeting two weeks ago 05:49:33 tagoh: I don't know - I think think it is a good idea 05:49:37 ah 05:49:47 still think 05:49:52 we didn't have a time to discuss this last week 05:49:53 (subpackaging) 05:49:57 ok 05:50:11 that helps to explain the haziness ;) 05:50:36 okay 05:50:45 anything else? 05:51:43 okay, better move on then 05:51:53 #topic Fonts and Rendering 05:52:02 any updates on fonts and rendering? 05:53:26 https://fedoraproject.org/wiki/Features/FontConfigurationTool 05:53:32 oh 05:54:13 juhp: btw for PANGO_LANGUAGE config tool, I'd like to bring this up prior to start detailed discussion - is there any actual cases/issues that one needs to set to satisfy their requrement? 05:55:04 tagoh: requirement? 05:55:09 epico_laptop: maybe better not limit CIJK? 05:55:20 tagoh:ok 05:55:39 juhp: doesn't one use PANGO_LANGUAGE they have any issues? 05:55:48 ^because 05:56:48 tagoh: sorry not completely sure on your question - are you asking what problem we are trying to solve, or ? 05:56:53 s/sure/clear 05:57:11 yes 05:57:14 For some Chinese users, some one are using English locale, but will read Chinese document. The fonts are mixed up with SC fonts and TC fonts. 05:57:42 set PANGO_LANGUAGE=zh_CN:en to solve this. 05:57:43 right 05:58:03 so in that sense it is particularly not for CIJK users :) 05:58:05 well, just want to clarify the tool we are going to develop is just for the corner-case or 05:58:33 tagoh: but epico_laptop thinks also need font priority config 05:58:43 which can be used by all - so not that corner 05:58:46 because we don't have it now by default and apparently not seeing any issues on bz? 05:58:57 heh 05:59:17 I think it is kind of a FAQ? 05:59:19 . bug 708921 05:59:21 epico_laptop: Bug 708921 Chinese small font in Fedora 15's Firefox not as good as that of Ubuntu's - https://bugzilla.redhat.com/show_bug.cgi?id=708921 05:59:45 I prefer to let the user to config the fonts himself. 06:00:00 dangerous business ;0) :) 06:00:28 instead of changing the defaults in Fedora. 06:00:35 tagoh: of course PANGO_LANGUAGE config is useless for users of ja desktop, etc 06:01:43 in the ideal world it would be part of the desktop font/lang config I guess 06:01:59 advanced I suppose 06:02:02 right 06:02:10 * epico_laptop think the user can specify wqy-microhei-fonts for sans/serif in his desktop. 06:02:38 epico_laptop, your desktop? 06:02:55 yes, that idea sounds good but how to determine the language from fontset? 06:03:06 I use the default wqy-zenhei-fonts with bitmap-font enabled. 06:03:57 tagoh: it only changes ~/.fonts.conf with . 06:04:18 hm, we are talking about different thing then :) 06:04:42 tagoh: well PANGO_LANGUAGE and font priority would be orthogonal 06:04:52 tagoh: which tools? pango language or font priority? 06:04:54 PANGO_LANGUAGE stuff may be one of desktop integration task on gnomecc region perhaps? 06:05:44 tagoh: the pango tool is selected by the users, user tells pango, which languages he use. 06:08:02 what's wrong to be done in region settings then. there should be a lot of combinations though, there should be some patterns for useful combinations IMHO 06:08:39 the pango tool allows a list of language instead one. 06:09:06 so currently on font config tool in gnome3? 06:09:16 s/on/no/ ! 06:09:22 s/allows/will allow/ 06:09:54 hmm again it may be hard to sell upstream though 06:10:24 this will be simpler than fontmanager. 06:10:30 s/this/this tool/ 06:10:30 right 06:10:45 we could do that at the startup time. and kind of this needs should be one time. 06:11:03 well, talking aboug PANGO_LANGUAGE 06:11:07 about 06:11:21 yeah that would be enough I guess - ideally it should be dynamic of course 06:11:46 guess we could merge this into lang.sh perhaps? 06:12:09 tagoh: how? 06:12:28 just set? 06:12:44 to what? :) 06:13:05 that may be guessable from lang no? 06:13:20 probably not I guess 06:13:21 as we have LANGUAGE for Chinese? 06:13:33 when it is guessable we don't need it :) 06:13:38 I think 06:14:08 the rfe is to allow overriding PANGO_LANGUAGE if you like 06:14:29 of course advanced users who know can already do that today 06:14:42 speaking of Chinese case, does everyone who use on Chinese desktop necessarily not need PANGO_LANGUAGE=zh_CN:en say? 06:15:22 yes, but some ones accustomed to English locale. 06:15:30 one may not like it but one may like it? 06:15:53 locale is probably en_US.UTF-8 06:16:01 For Chinese locale, it is not needed. 06:16:12 juhp: hm 06:16:48 would be interesting to know what percentage of non-US users use en_US 06:17:23 a lot 06:17:31 well, reminds me that using non-english language on en_US locale is really unpredictable. I don't think we can support all of requirements on it... 06:17:31 yeah also my feeling 06:17:55 but we can try to make it better 06:18:18 unless you mean non-utf-8? 06:18:48 no 06:19:21 For SC users, PANGO_LANGUAGE=zh_CN:en is enough. 06:19:21 I try to use my native language but then I get confused, so en_US is better ( at least for me ) 06:19:42 hicham, I know the feeling 06:20:05 hicham, curious though - what is your native lang? 06:20:30 juhp: arabic 06:20:35 ah right 06:21:09 hicham: dunno what kind of difficulty you saw, is it impossible to get it fixed instead of changing the locale to en? 06:21:22 translations? 06:21:39 translations for SC also I think. 06:22:07 I think venture relunctantly guess to say that is the largest problem generally :-/ 06:22:09 juhp: fwiw explicitly LC_MESSAGES is set since f15 06:22:21 aha 06:22:33 s/think// 06:22:41 tagoh : first, no support in terminals (most of them), only one supporting bidi is mlterm 06:23:11 also in the past, fonts in SC locale is not so good. 06:23:24 hicham: aha 06:23:38 epico_laptop: font selection tool looks good, since presently we are as a developer decide, which font to use for language. 06:23:49 will be good if user can also get some control for this 06:23:58 tagoh : second, not everything can be localized, example : your login name, computer name ... 06:24:08 looks even better now that gnome doesn't seem to have one? :o) 06:24:23 pravins:thanks:) 06:24:29 juhp: any code to test for fontconfig tool ? 06:24:38 hicham: heh. it's not arabic specific issue right? 06:25:03 I think terminal rendering may also be poor in some other locales 06:25:13 dunno if Korean any better now 06:25:16 juhp: right 06:25:17 and yeah old SC 06:25:19 tagoh : yeah, most issues are not arabic specific 06:25:48 tagoh : one that is specific to arabic is ligatures 06:26:35 hm, well, we should try to fix as long as any issues can be fixed technically 06:27:07 hicham: fyi: bicon is available for fedora: 06:27:08 http://desktopi18n.wordpress.com/2011/01/31/bicon/ 06:28:43 hicham: is that issue filed in bz btw? or known in upstream? 06:29:05 feel I saw that anywhere 06:29:35 tagoh : I believe somebody filed a bug, let me look 06:29:42 ok 06:31:06 anyway, that's good to see we could clarify concerns for future tasks 06:31:43 tagoh : it is fonts specific, kacst to be precise 06:31:46 ok 06:32:12 hicham: aha 06:32:23 tagoh : ojuba ( an arabic fedora derivative ) is using some patches for them 06:32:41 do you have a bug#? 06:33:04 tagoh : no, but ojuba maintainer told me he tried to get the patches into fedora and failed 06:33:23 aha 06:34:14 may be nice if you can get a bug#. interesting why it's failed 06:35:05 anyway, we are spending too much time on the meeting so better stop here today unless there are anything else urgent topics? 06:35:17 tagoh : here is a screenshot of the ligatures bug http://imageshack.us/photo/my-images/4/screenshotzgw.png/ 06:35:34 fujiwarat: thanks :) ! 06:36:15 hicham: ah, I think I saw that in bz.. 06:36:22 or similar 06:37:40 okay, thanks everyone for the meeting! 06:38:30 #endmeeting