06:03:15 <tagoh_> #startmeeting i18n
06:03:15 <zodbot> Meeting started Thu Feb  7 06:03:15 2013 UTC.  The chair is tagoh_. Information about MeetBot at http://wiki.debian.org/MeetBot.
06:03:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
06:03:16 <tagoh_> #meetingname i18n
06:03:16 <zodbot> The meeting name has been set to 'i18n'
06:03:16 <tagoh_> #topic agenda and roll call
06:03:16 <tagoh_> #link https://fedoraproject.org/wiki/I18N/Meetings/2013-02-07
06:03:29 <tagoh_> okay, let's have i18n meeting..
06:03:36 <epico> hi
06:03:51 <anish_> hi
06:03:53 <tagoh_> I forgot to send a reminder again. doh! sorry for that btw...
06:05:37 <juhp> hi
06:06:01 <fujiwarat> hi
06:08:37 <tagoh_> okay, let's get started
06:08:47 <tagoh_> #topic Upcoming schedule
06:08:48 <tagoh_> #info 2013-02-12        f16 EOL
06:08:48 <tagoh_> #info the end of February (NOT DECIDED YET, INITIAL TARGET!!!)  Branch Fedora 19 from Rawhide
06:08:49 <tagoh_> #info second half of May (NOT DECIDED YET, INITIAL TARGET!!!)   Fedora 19 Final Release
06:08:56 <tagoh_> no updates in schedule yet
06:09:44 <tagoh_> still needs to focus into f16 bugs for re-triaging
06:10:12 <tagoh_> #topic Outstanding topics
06:10:13 <tagoh_> #info #12: language support installation command (pnemade)
06:10:13 <tagoh_> #link https://fedorahosted.org/i18n/ticket/12
06:10:19 <tagoh_> paragan: you there?
06:11:28 <anish_> paragan, is not here.
06:11:58 <anish_> he may join after some time not sure.
06:12:17 <tagoh_> aha. okay. better move on then.
06:12:32 <tagoh_> #info #14: Fedora 16 bugs cleanup (tagoh)
06:12:33 <tagoh_> #link https://fedorahosted.org/i18n/ticket/14
06:13:07 <tagoh_> thanks for updates and activities on f16 bugs
06:13:28 <tagoh_> though we still have 20 bugs open and need to have a look at them
06:14:47 <tagoh_> for our own bugs, 2 bugs for ibus, 3 bugs for ibus-chewing, one bug for ibus-indic-table and 3 bugs for stardict.
06:15:15 <juhp> anish_, dingyichen, fujiwarat ?
06:15:27 <tagoh_> good to triage them as early as possible and help to triage other bugs too.
06:16:08 <anish_> anish_, sure I will close it in this week
06:16:10 <dingyichen> juhp, I just pushed libchewing-0.3.4
06:16:23 <dingyichen> juhp, will do ibus-chewing afterward.
06:16:43 <tagoh_> dingyichen: cool
06:17:01 <juhp> anish_, please
06:17:39 <anish_> juhp, okay thanks
06:17:46 <tagoh_> anish_: close as FIXED? :)
06:18:49 <juhp> dingyichen, thanks
06:19:26 <tagoh_> anything else about f16 bugs you may want to share?
06:20:04 <tagoh_> or want us to be focused particularly?
06:22:24 <tagoh_> okay, moving on...
06:22:57 <tagoh_> #info #16: Improving testcases/matrix for i18n test day (tagoh)
06:22:57 <tagoh_> #link https://fedorahosted.org/i18n/ticket/16
06:23:48 <tagoh_> well, updated the ticket this morning though, I created a draft of new i18n test day matrix here: https://fedoraproject.org/wiki/User:Tagoh/Draft/I18n_Test_Day
06:24:32 <tagoh_> I focused how it can be improved to see what languages the test cases are passed or not.
06:24:54 <tagoh_> rather than who tested
06:26:10 <juhp> aha
06:27:16 <juhp> I think it looks good
06:27:27 <juhp> maybe good to have QA people look over it?
06:28:02 <tagoh_> test cases may not covers all so the result doesn't necessarily ensures the input, rendering and printing etc works on that language but it can be improved by adding more testcases later that we are missing
06:28:08 <tagoh_> sure. yeah, good idea
06:29:19 <tagoh_> and FLTG too perhaps
06:30:02 <tagoh_> let me post this to the list later.
06:30:32 <tagoh_> any other comments?
06:33:04 <tagoh_> okay, let's move on.
06:33:09 <tagoh_> #topic Outstanding task
06:33:10 <tagoh_> #info #15: F19 Feature: libkkc (ueno)
06:33:10 <tagoh_> #link https://fedorahosted.org/i18n/ticket/15
06:33:24 <tagoh_> dueno: can you update the status of this feature?
06:34:13 <dueno> sure, currently libkkc and ibus-kkc is on review
06:34:23 <tagoh_> cool
06:35:01 <dueno> but it turned out there are some arch-dependent data and repackaging them now
06:35:43 <tagoh_> aha
06:37:41 <tagoh_> dueno: how about feature implementation? what do you need to do for 40% of works?
06:38:52 <dueno> probably it needs UI to allow users to have custom keybindings
06:39:46 <juhp> and testing? :)
06:40:22 <dueno> right - both usability testing and unit testing :)
06:41:50 <tagoh_> improving accuracy of conversion and making it stable may be able to be done outside of this feature? sort of it may be hard to estimate the goal?
06:43:00 <tagoh_> or do you have any milestones for that? like you blogged about benchmark?
06:43:48 <dueno> yeah, at least it should keep over 90% accuracy (in my feeling though)
06:43:58 <tagoh_> aha
06:45:45 <juhp> dueno, will you clear up the license disparity between the source and the copying file?
06:46:18 <tagoh_> well, one more question about migration from anthy. do you have any detailed plans for that?
06:46:22 <dueno> juhp, sorry, which package?
06:46:37 <juhp> dueno, I meant https://bugzilla.redhat.com/show_bug.cgi?id=906255#c7
06:46:45 <juhp> libkkc
06:47:22 <juhp> erm ibus-kkc
06:47:34 <dueno> I see
06:47:45 <juhp> anyway it is lesser thing - just nice to be consistent i think
06:48:44 <juhp> tagoh_, what do you have in mind - the physical change of IME or config/data?
06:49:43 <tagoh_> juhp: both if it's the case. are we going to retire packaging of ibus-anthy and anthy?
06:49:49 <juhp> but maybe ibus-kkc should not obsolete ibus-anthy at this point?
06:49:57 <juhp> I guess not yet?
06:50:17 <juhp> but maybe in the future?
06:50:22 <tagoh_> juhp: maybe. that's why I'm asking now :)
06:50:27 <juhp> ok
06:50:49 <tagoh_> for fresh install, it would be easy. we can just modify comps. how about upgrading?
06:51:12 <juhp> epico, what did you do for ibus-pinyin to ibus-libpinyin - though the history there is different
06:51:26 <juhp> tagoh_, maybe do nothing?  hmm
06:51:52 <juhp> dueno, any thoughts?
06:51:57 <tagoh_> one idea from me is to obsolete anthy only this time if libkkc provides compatible interface?
06:52:14 <juhp> aha
06:52:17 <juhp> hmm
06:52:24 <juhp> I see
06:52:29 <epico> juhp, the users can install either one.
06:52:37 <juhp> right
06:52:44 <epico> but comps will recommend ibus-libpinyin.
06:53:00 <juhp> epico, so you didn't provide any "forced" migration right?
06:53:15 <epico> no forced migration.
06:53:22 <epico> yes
06:53:40 <juhp> but wasn't fedora ibus-pinyin already using libpinyin?
06:53:55 <tagoh_> but it's a bit different case compared to kkc. we have already used libpinyin in ibus-pinyin right?
06:54:00 <juhp> nod
06:54:28 <tagoh_> so if we use libkkc instead of anthy in ibus-anthy, it should be same :)
06:54:41 <epico> that version of ibus-pinyin has been obsoleted, currently ibus-pinyin is the same as upstream, libpinyin integration patch has been removed.
06:54:50 <juhp> so in a sense it was a bit strange if people were using ibus-pinyin with libpinyin and then reverted to upstream ibus-pinyin
06:54:58 <juhp> epico, indeed
06:55:21 <juhp> though I can't remember all the timing
06:55:55 <juhp> epico, Obsoletes: ibus-pinyin < 1.4.0-17
06:56:02 <epico> right
06:56:13 <juhp> so actually that is kind of migration
06:56:18 <juhp> anyway
06:56:19 <tagoh_> that said if libkkc is better than anthy, it's unfortunate if one can't use it when upgrading
06:56:23 <epico> but the ime name in ibus menu is not changed.
06:56:57 <juhp> tagoh_, true - well there is yum install but...
06:57:31 <juhp> in the ideal world we probably would obsolete ibus-anthy... but it seems slightly aggressive at this time?
06:57:54 <juhp> anyway still time to think about this I guess
06:58:01 <tagoh_> yep
06:58:43 <juhp> if it could be done in a way that still allows reinstalling ibus-anthy it might be ok
06:59:35 <juhp> but it needs care to keep old releases ver-rel under some threshold
06:59:42 <tagoh_> hmm, guess it may be difficult? that would basically means one can't be upgraded to libkkc when it can be done?
07:00:32 <juhp> if it is really to be done probably easier to do it across all releases at once say
07:00:47 <juhp> like for ibus-pinyin?
07:01:07 <juhp> with the short devel cycle seems hard to do that
07:01:16 <tagoh_> right
07:01:42 <juhp> so I probably I would such to punt that to a later release
07:02:00 <tagoh_> I guess some people may still use ibus-pinyin without knowing ibus-libpinyin perhaps?
07:02:00 <juhp> erm suggest
07:02:42 <juhp> at least the obsoletes should have switched most users I think
07:03:07 <tagoh_> if they do yum update before releasing new version of ibus-pinyin right
07:03:31 <juhp> or maybe the older IMEs should have some message to recommend switching
07:03:47 <juhp> anyway
07:04:17 <juhp> if gnome showed preferred IMEs that would also help here
07:04:44 <tagoh_> let's think about what's better later and update here if someone has any idea... or at least I should remind you to have any decision as needed.
07:04:45 <juhp> fujiwarat, can you talk to rtcm and aday about IME labelling in g-c-c?
07:05:41 <tagoh_> #topic Open Floor
07:06:17 <tagoh_> anything else we want to discuss in the meeting?
07:08:02 <tagoh_> juhp: do they still have hardcoding list for default IME?
07:08:52 <juhp> hm
07:09:20 <juhp> tagoh_, in ibus?
07:09:33 <tagoh_> in GNOME (g-s-d?)
07:09:35 <juhp> ah gnome I guess
07:09:57 <juhp> I think there should be - some probably needs to be updated for 3.8
07:10:07 <juhp> if we switch to ibus-kkc
07:10:22 <tagoh_> hm
07:10:37 <juhp> s/some/so/
07:10:47 <juhp> well I am not 100% sure
07:10:51 <tagoh_> okay
07:11:06 <juhp> but assume that is how it would setup default IME based on locale
07:11:28 <juhp> I guess fujiwarat would know
07:12:26 <juhp> mfabian, perhaps that should also be part of the region kbd data?
07:14:40 <tagoh_> okay, anything else? if not, let's stop here then
07:17:45 <tagoh_> okay... thanks everyone for the meeting!
07:17:49 <tagoh_> #endmeeting