05:01:37 <tagoh> #startmeeting i18n
05:01:37 <zodbot> Meeting started Thu Sep 15 05:01:37 2011 UTC.  The chair is tagoh. Information about MeetBot at http://wiki.debian.org/MeetBot.
05:01:37 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
05:01:43 <tagoh> #meetingname i18n
05:01:43 <zodbot> The meeting name has been set to 'i18n'
05:01:50 <tagoh> #topic agenda and roll call
05:02:00 <tagoh> https://fedoraproject.org/wiki/I18N/Meetings/2011-09-15
05:02:16 <tagoh> hi guys, shall we have weekly i18n meeting here.
05:02:44 <dueno> hi
05:02:49 <epico_laptop> hi
05:02:57 <juhp> hi
05:05:17 <jni> hi
05:06:15 <tagoh> okay, let's get started
05:06:19 <tagoh> #topic F16
05:07:35 <tagoh> started counting down to f16 beta now and in current schedule, beta will be available 27th September.
05:08:43 <tagoh> we still have a chance to fix bugs for GA if it's critical or for 0-day errata.
05:08:51 <juhp> hmm not sure how that C-w went to my irc client...
05:09:03 <tagoh> keep up to make f16 better.
05:09:44 <tagoh> also we are planning to have another test day for ibus new feature next Thursday
05:10:07 <juhp> I started editing the relnotes a bit but still not finished - I need to verify some of the ibus features to understand if I am writing correctly
05:10:31 <juhp> fujiwarat, may ask you later for some clarification on some points
05:10:53 <juhp> sorry perhaps I am jumping ahead?
05:11:07 <tagoh> are we going to postpone the ibus feature for f16 right? apparently no updates of the percentage nor category on the feature page though
05:11:30 <juhp> well I am not sure - looking at it again it seems mostly done?
05:11:45 <juhp> just hasn't been updated yet...
05:11:51 <pravin_s> hi
05:12:23 <tagoh> hmm
05:12:30 <juhp> there were some question marks about gnome-shell  3.2 extensions earlier in the week but I gather it is still possible
05:12:33 <juhp> fujiwarat, ?
05:13:46 <tagoh> as you all know, the due date of the feature complete was over. so basically all of the features' status has to be complete.
05:14:32 <tagoh> I don't see any mails reminding that anywhere unless I'm missing
05:14:39 <juhp> tagoh: right
05:14:45 <juhp> I want to update it today
05:14:50 <tagoh> cool
05:15:08 <juhp> fujiwarat, okay let's talk about that too then
05:15:37 <tagoh> so back to the docbeats stuff
05:16:24 <tagoh> juhp: thanks for updating. not sure when is the final deadline to compose, so would be nice to finish it as early as possible
05:17:25 <juhp> tagoh: ok yep
05:17:36 <tagoh> ah, and good to review the testcases for ibus features too..
05:17:58 <juhp> I think I have done about 1/3 now - will finish off this week
05:18:07 <juhp> tagoh: yes
05:18:36 <tagoh> since there was some outdated information for the past test day and updated a bit. hope it will be improved before starting
05:19:09 <tagoh> okay, anything else for f16?
05:19:19 <dingyichen> hi
05:20:26 <tagoh> juhp: ah, forgot to test langpacks issue. will have a time later..
05:21:04 <juhp> cool
05:21:13 <juhp> hope it is better now
05:22:03 <tagoh> okay, better move on
05:22:10 <tagoh> #topic Input Methods
05:22:27 <tagoh> any updates to share for IM?
05:23:36 <tagoh> 41 bugs keep open for IM now
05:23:58 <juhp> hm
05:24:24 <tagoh> that seems including some RFEs. maybe we should review that if it's still necessary or useful.
05:26:15 <juhp> mm
05:27:06 <tagoh> would be nice to have a look at bugs if one owns such bugs anyway
05:27:17 <pravin_s> .bug 727019
05:27:20 <zodbot> pravin_s: Bug 727019 ibus compose file needs a symbol tag for gnome-shell - https://bugzilla.redhat.com/show_bug.cgi?id=727019
05:28:12 <juhp> how about other IMEs?
05:28:42 <pravin_s> fujiwarat, tagoh  here i am bit confuse what symbol can i add for input method, since if we add proper name of IME it will take lots of space of task bar
05:29:03 <pravin_s> multi byte
05:29:15 <tagoh> pravin_s: seems 2 characters only allowed there
05:29:22 <pravin_s> ohh
05:29:39 <pravin_s> i still think, icon is more expressive
05:29:39 <anish_> i am working on 683743,683826
05:30:18 <pravin_s> tagoh: i think this is injustice of gnome-shell on IME developer, they are using icons for network preferences, sound control. then what is problem with IME icon
05:30:20 <juhp> pravin_s, is it possible to make some mini-string?
05:31:04 <juhp> hm
05:31:17 <pravin_s> juhp: yeah, even i am thinking for same but it will be difficult for user to recognize
05:31:34 <juhp> pravin_s, I think using text will give a more unified feel but I hear what you're saying
05:32:24 <pravin_s> juhp: i am only feeling problem with text will take more space, but year need to search for some shortcut
05:32:26 <tagoh> pravin_s: well, vice versa IMHO unless the icon is well-designed nor easy to imagine what we expect.
05:32:27 <juhp> pravin_s, you mean hard to distinguish between maps?  I think current icons look like one "letter" mostly
05:33:00 <pravin_s> tagoh: yeah, agree icon must match with gnome shell theme, yeah controlling that might be difficult
05:33:23 <pravin_s> its like hi-i (hindi inscript)
05:33:32 <pravin_s> hi-p (hindi phonetic)
05:33:56 <juhp> pravin_s, it would be good to experiment anyway - I guess it needs changes to m17n-lib (and m17n-{db,contrib}) perhaps
05:34:10 <juhp> paragn, how do you think?
05:34:33 <pravin_s> juhp: actual problem happen when we will  be localize environment, since it can take more time that time ;)
05:34:37 <pravin_s> s/time/sapce
05:34:39 <juhp> dueno, I guess ibus-m17n supports the symbol tags already right?
05:34:51 <tagoh> btw do we postpone to ship ibus-gnome3 by default in f16?
05:35:05 <juhp> fujiwarat, ping?
05:35:24 <pravin_s> juhp: yes, they support
05:35:44 <juhp> tagoh: latest I heard from him was that it could still be possible with latest gnome-shell-3.1.99(?)
05:35:59 <juhp> using some gsettings, but I am not completely clear on this
05:36:22 <tagoh> juhp: but need to activate the extension manually by adding a value to gsettings?
05:36:45 <juhp> pravin_s, I don't think the symbol strings will be localized they are intended to be statuc
05:36:47 <juhp> static
05:37:05 <juhp> tagoh: perhaps - i didn't fully understand
05:37:20 <tagoh> should we add some notes to relnotes for that perhaps?
05:38:11 <juhp> tagoh: yeah probably
05:38:14 <pravin_s> juhp: actually i feel as it says symbol, symbol is different from character
05:38:34 <pravin_s> i think gnome-shell want people to use symbol provided by unicode
05:39:10 <pravin_s> like in CJK each character is like word (if i am not wrong)
05:39:23 <pravin_s> so same way need to find some symbol may be from unicode charts
05:39:26 <juhp> pravin_s, well I think currently this is coming more from ibus than gnome-shell - so there are no strings corresponding to the current icons?
05:39:55 <juhp> if there are that would be a good start IMHO
05:40:18 <pravin_s> yeah, need to experiment on it then
05:40:25 <juhp> okay
05:40:55 <pravin_s> juhp: presently i can see single character for marathi-inscript instead of icon
05:41:03 <pravin_s> "क"
05:41:05 <juhp> is paragn around?
05:41:13 <juhp> aha
05:41:25 <pravin_s> but it is not much expressive
05:41:55 * juhp has gotten notifier working yet with gnome-shell 3.2-pre
05:42:06 <juhp> pravin_s, so missing the subscript?
05:42:27 <juhp> pravin_s, what does the icon say?
05:42:49 <pravin_s> its is just single character from script
05:42:55 <juhp> pravin_s, or is it better to use "mr" for Marathi?
05:43:09 <pravin_s> yeah, i think it might help
05:43:19 <juhp> pravin_s, so what is the difference? :)
05:43:21 <pravin_s> like for english we are using en
05:43:41 <tagoh> hm, but it may be conflict to the keyboard layout?
05:43:42 <alam> 'mr' can be good, but which keymap for 'mr'?
05:43:46 <juhp> pravin_s, I guess symbol string should be up to the native users
05:44:04 <juhp> alam, we can use mr1, mr2, etc for example
05:44:09 <juhp> or mr_1
05:44:10 <tagoh> that may be hard to see which one is IM or keyboard.
05:44:31 <juhp> do people switch maps a lot?
05:44:36 <juhp> for the same language?
05:45:23 <pravin_s> juhp: considering normal user, he will use only 2, english and marathi
05:45:27 <pravin_s> :)
05:45:41 <pravin_s> and only one layout of marathi
05:45:52 <juhp> then "mr" may be ok :)
05:46:10 <pravin_s> we developer use these in bulks ;)
05:46:11 <juhp> if "mr" is better than "क" say
05:46:40 <juhp> pravin_s, even for en with as using en1, en2 etc to distinguish layouts
05:46:44 <juhp> and gnome too
05:46:54 <juhp> well en_1, en_2 etc I think
05:47:03 <tagoh> can't say off hand because I didn't try to add a keyboard layout except en and ja. but IIRC ibus displays the keyboard layout anyway even if it has only one.
05:47:04 <pravin_s> juhp: yes, en1
05:47:17 <pravin_s> i am using english with rupee sign, it shows me en1
05:47:21 <juhp> anyway perhaps we should move on dunno :)
05:47:35 <dueno> hmm, maybe mr is better than क, since currently क is also used for ks, mai, sa, etc.
05:47:35 <juhp> tagoh: yeah I think so
05:47:44 <juhp> dueno, right
05:47:47 <juhp> okay
05:47:54 <pravin_s> tagoh: juhp yeah, need to find better symbol name that is conclusion :)
05:47:58 <tagoh> better trying. we could discuss against it later.
05:48:13 <juhp> pravin_s, so how are the icons distinguishable currently - by subscript?
05:48:24 <alam> why not langauge name's first chracter instead of script
05:48:36 <pravin_s> icon contains lang name i.e hi
05:48:36 <juhp> alam, sure
05:48:44 <juhp> ah
05:48:47 <pravin_s> and also layout name initial
05:49:16 <pravin_s> alam: yeah, that is more logical
05:49:56 <juhp> pravin_s, perhaps you can discuss with Parag since he is involved with the m17n maps
05:50:16 <pravin_s> juhp: i am thinking for booster maps as well ;)
05:50:30 <pravin_s> did lots of efforts for developing icon :)
05:50:31 <tagoh> guess that may be even good with the icon if it's based on the same theme, maybe. dunno
05:50:58 <pravin_s> tagoh: yeah, it will be best solution.
05:51:21 <pravin_s> but i guess gnome cant control themes of individual icons
05:51:55 <tagoh> anyway, try to use some symbols and see how it looks like. and try to find out the best.
05:52:03 <juhp> agreed
05:52:14 <juhp> we should try at least
05:52:27 <pravin_s> juhp: tagoh i will take a look at symbols provided by unicode http://unicode.org/charts/#symbols
05:52:40 <pravin_s> agree
05:52:48 <tagoh> cool
05:52:55 <tagoh> okay, anything else?
05:53:09 <alam> don't we should ask ISO code to provide a lokalize symbol for each locale with script?:)
05:53:19 <alam> tagoh: a bug
05:53:34 <alam> https://bugzilla.redhat.com/show_bug.cgi?id=727806
05:53:43 <tagoh> alam: sure
05:53:57 <alam> it is marked as 'FutureFeature', but kde it is actually problem
05:54:08 <juhp> pravin_s, I don't think symbols needs to be symbols :)
05:55:09 <pravin_s> :) yeah, i think symbol is more expressive than character, hmm but lots of limitation now since unicode involved
05:55:43 <tagoh> alam: dunno how many people use gtk+ apps on KDE but if you think it's a bug and important for f16, feel free to update.
05:56:24 <alam> tagoh:  it is about missing icon : https://bugzilla.redhat.com/show_bug.cgi?id=737435 (No ibus icon on KDE desktop)
05:56:36 <tagoh> alam: we should bring this up on the meeting for KDE perhaps. I've filed another patch for xsettings-kde though, no response from the maintainer yet.
05:56:55 <tagoh> so may need some attention from KDE SIG
05:56:56 <alam> Icon issue, which is depends upon xsettings-kde
05:57:53 <alam> tagoh: is it possible to get build to test one?
05:57:59 <alam> s/one/once
05:57:59 <tagoh> alam: I know. well, there are workaround though.
05:58:52 <tagoh> alam: yes if I'm ready to have a patch for that
05:59:45 <tagoh> will update bz
05:59:46 <alam> can you please scratch build for me, if possible?
05:59:58 <tagoh> okay
06:00:02 <alam> thanks
06:00:52 <tagoh> okay, let's move on then
06:00:57 <tagoh> #topic Fonts and Rendering
06:01:05 <tagoh> any updates?
06:01:21 <tagoh> or concerns to share?
06:02:01 <pravin_s> tagoh: as you know, soon we are going to change lohit to OFL
06:02:34 <tagoh> yep
06:02:50 <juhp> very good news
06:03:38 <pravin_s> that is the only update
06:03:46 <pravin_s> .bug 733106
06:03:48 <alam> ah, available with andriod then:)
06:03:48 <zodbot> pravin_s: Bug 733106 fonts.alias refer to encodings not listed in fonts.dir - https://bugzilla.redhat.com/show_bug.cgi?id=733106
06:04:30 <pravin_s> tagoh: for this bug i am thinking it on Fedora 12 and see was it working fine on f12
06:04:58 <tagoh> pravin_s: right. something went wrong these days
06:05:13 <juhp> is that a font tools regression/change?
06:05:21 <juhp> utils
06:05:50 <tagoh> juhp: might be a regression in ttmkfdir perhaps
06:05:57 <juhp> hm
06:06:46 <pravin_s> juhp: last patch was yours
06:06:54 <juhp> lol
06:06:55 <pravin_s> - simplify ttmkfdir-3.0.9-encoding-dir.patch to drop X11R6/ check (#173705)
06:06:58 <pravin_s> :)
06:07:09 <juhp> so it is my fault? :)
06:07:27 <pravin_s> juhp: nope, actually want to say there is not changes in it from long time
06:07:33 <juhp> nod
06:07:50 <juhp> pravin_s, is your package using ttmkfdir?
06:08:17 <pravin_s> juhp: problem is with sazanami-fonts
06:08:44 <tagoh> it's just no one or less people uses X core fonts right now and didn't report it anyway
06:10:25 <juhp> ah
06:10:54 <tagoh> maybe good to test on older release perhaps if it's hard to track it down. and if you suspect any patch may affects, try to fix it.
06:11:27 <juhp> wonder if xorg-x11-font-utils helps at all?
06:12:05 <tagoh> IIRC mkfontscale doesn't perfectly work for CJK fonts.
06:12:40 <juhp> I see might be worth just testing again - I guess it is more actively maintained anyway
06:12:41 <pravin_s> tagoh: yeah agree, even i tried it and it was giving only 2 encoding
06:12:52 <juhp> hm
06:13:07 <pravin_s> where as ttmkfdir gives 5 encoding in scale file
06:15:33 <pravin_s> tagoh: there are now bulk for bugs for liberation fonts, need to check that in f16. most of these for hinting issues
06:15:35 <juhp> ok
06:15:59 <tagoh> okay
06:16:01 <pravin_s> will check it this week
06:16:06 <tagoh> cool
06:17:08 <pravin_s> thats it from me
06:17:45 <tagoh> okay, let's stop here today unless anything else.
06:18:55 <tagoh> thanks everyone for the meeting!
06:19:00 <tagoh> #endmeeting