05:00:06 #startmeeting i18n 05:00:06 Meeting started Tue Sep 1 05:00:06 2009 UTC. The chair is juhp. Information about MeetBot at http://wiki.debian.org/MeetBot. 05:00:06 Useful Commands: #action #agreed #halp #info #idea #link #topic. 05:00:31 hi juhp 05:00:33 hi 05:00:35 hi juhp 05:00:37 hi 05:01:21 hi guys 05:01:34 welcome to our biweekly i18n meeting 05:02:30 hi 05:03:01 #topic https://fedoraproject.org/wiki/I18N/Meetings/2009-09-01 05:04:17 oops 05:04:23 #topic F12 05:04:48 beta freeze is at the end of this month - so about 4 hours of devel left 05:04:57 ugh 4 months, lol 05:05:08 this getting silly - 4 WEEKS 05:05:48 any concerns about f12 schedule? 05:06:39 juhp: What's our features in F-12? 05:08:07 I guess there are still various work/features we might hope to land this month - various some ibus improvements, yum plugin, xkb maps for indic inscript? 05:08:36 any others I am forgetting? 05:09:28 dychen: hmm might be good to list then for f13 on the wiki beyond official fedora features 05:09:57 if no comments then let's move on 05:10:13 #topic rawhide bugs 05:10:25 https://bugzilla.redhat.com/buglist.cgi?product=Fedora&version=rawhide&bug_status=__open__&emailassigned_to2=1&emailcc2=1&emailtype2=exact&email2=fedora-i18n-bugs@redhat.com 05:10:45 76 open rawhide bugs we own or are tracking 05:11:30 anyone have any particular issues they want to discuss? 05:12:30 bug 518395 05:12:31 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=518395 medium, low, ---, anaconda-maint-list, NEW, add @input-methods to Package Selection screen 05:12:43 looking increasingly unlikely 05:14:27 juhp: So we still install every im by default? :-P 05:14:46 that is what we are still doing for now 05:15:15 I don't really want to make any change until we have a reliable easy way of doing post-install 05:15:36 perhaps we should just add a uninstall script to ibus ;-P 05:16:30 heh 05:17:10 I noticed that ubuntu actually does something like that on their live spins: removing unwanted packages at the end 05:17:17 of installation 05:17:27 but anyway 05:17:52 might make sense to uninstall IMEs for other langs at that time but that is another issue 05:18:52 tagoh3: do you think it is worth the pain to turn off @input-methods for f12? 05:19:02 juhp: But users have to specify want/unwant at some point anyway. 05:19:09 indeed 05:19:38 dychen: they can still do going down into the package customization menu 05:20:05 juhp: hmm, dunno. it may depends on how many people uses IM on non-CIJK languages. 05:20:21 of course appearing on the top-level would be much easier - though it doesn't solve the I only want one IME problem 05:21:03 tagoh3: enough to make their questions common anyway iirc 05:22:16 juhp: Not sure it is appropriate to specify the IM on regional setting. 05:22:41 Kaio: What does Window 7/Vista do about this issue? 05:22:48 dychen, it is a solution 05:23:03 dychen: yeah perhaps better not to install and auto-install on login? though it is a bit annoying 05:23:16 dychen: region setting? 05:23:44 phuang_: Yeah , I think Win XP (and 2000 maybe) do in that way. 05:23:49 anyway I am proposing to postpone this to f13... 05:23:50 dychen, I think Window has different localization versions. And each version will install default IME for the location 05:24:04 right 05:24:07 juhp: yeah.. and that requires restarting desktop.. 05:24:18 hmm 05:24:45 phuang_: They do, and for english version, you can also set IM in region setting. 05:24:53 tagoh3: can't we starting ibus after that? 05:25:17 and imsettings 05:25:44 juhp: we can if we don't care of xim 05:25:52 ah yeah 05:26:03 well maybe for first session it is ok 05:26:42 dychen, yeah. But it does't install IMEs in install-time. Or maybe installed some, but does not enable any ime 05:26:45 juhp: or we can't - guess imsettings won't be installed without IM deps, so far at least. 05:28:01 phuang_: Yes they do. You can tick the Asian language support during the installation. 05:28:43 tagoh3: perhaps we could install imsettings by default? 05:29:08 anyway 05:29:21 juhp: yep 05:29:30 phuang_: http://pcsupport.about.com/od/operatingsystems/ss/installxpnew2_3.htm 05:29:57 dychen, right 05:30:28 how is rawhide fontconfig? 05:30:53 we got new fontconfig-2.7.2 in rawhide now 05:30:56 bug 518161 05:30:57 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=518161 medium, low, ---, besfahbo, NEW, "contains" expression seems not working on the fontconfig rule 05:31:07 will be testing it with new lohit-fonts-2.4.2 05:31:16 * juhp notes it is being pushed to f11 too 05:31:24 paragan: cool 05:31:29 yes 05:31:40 (should be interesting;) 05:32:41 we also plan to split lohit-fonts in per font tarball release 05:32:43 hmm guess devel might have been enough for now ;) 05:32:51 paragan: aha 05:33:05 yes devel is enough for now 05:33:19 (I meant fontconfig:) 05:33:47 anyway hope there is time to test fontconfig in rawhide before f11 gets pushed... 05:34:07 yes 05:34:32 wonder if that is the only change 05:34:52 splitting lohit-fonts will be really good for maintenance as well 05:35:09 pravins: ok - sounds good then :) 05:35:37 juhp, I see changelog which shows 10 commits since 2.7.1 05:35:46 aha 05:36:17 3 of them for Windows 05:36:26 hm 05:40:44 next topic ^^ 05:40:46 phuang_: the new ibus icon looks different! 05:41:13 juhp, You saw the new icon? 05:41:19 yeah :) 05:41:27 juhp, I have built the new ibus include the new icons 05:41:30 different to the old one I mean :) 05:41:35 thanks 05:41:43 just getting used to it :) 05:41:44 No 05:41:50 ? 05:42:15 Just installed a set of icons in different size 05:42:24 aha 05:42:50 The last ibus uses 48x48, so the it looks not good 05:42:56 ok 05:43:08 any other topics to discuss today? 05:43:28 harfbuzz indic test suit 05:44:01 there are some discussion goind on harfbuzz list for making harfbuzz rendering engine compatible to new OT SPec 05:44:07 recent rawhide seems ok to me mostly - though I had to install from a old-ish image to avoid anaconda backtraces 05:44:17 same as Uniscribe 05:44:21 pravins: I see - interesting 05:44:37 does pango look ok? 05:44:37 so behdad requested test cases for indic 05:44:42 aha 05:44:55 he want o/p as >OUTPUT: <1,0,0>,<5,10,30> 05:44:56 >>Where the output tuples are glyph id, X and Y. 05:45:23 hm 05:45:38 anyone knows whats this glyph id and x, y?? 05:45:53 i was trying to prepare first draft for same 05:46:09 maybe better to ask? 05:46:13 hmm 05:46:54 but new OT specs 1.6 has lots of changes 05:47:06 many new feature tags are introduces in it 05:47:17 for indic script 05:47:51 might we will need to update lohit-fonts as well with this new feature may be 05:48:28 http://www.microsoft.com/typography/SpecificationsOverview.mspx 05:48:59 tagoh3: will we rename knm_new-fonts? 05:49:30 pravins: hmm dunno - guess it might be a lot of work? 05:50:11 juhp: hmm, sure - I'm not still fully understanding though. 05:50:28 yeah, but firstly rendering engine need update w.r.t OT spec 05:50:35 ah yeah 05:50:53 pravins: not sure if/when that might happen? 05:51:32 this work is going on for harfbuzz 05:52:03 tagoh3: sounds like underscore breaks fontspackages, so might be easiest or as you mentioned do a workaround in package? 05:52:08 aha 05:52:31 juhp: right 05:52:45 hope mailhot can fix droid issue before beta... 05:53:06 juhp: well, that makes sense if we do get rid of any exceptions from the policy though. 05:53:24 tagoh3: aha - so there are already other exceptions? 05:53:59 there are some - not font specific but in the package naming policy. 05:54:08 ok sure 05:54:23 tagoh3: is knm_new installed by default for ja now? 05:54:43 juhp: I don't think so 05:54:46 ok 05:55:52 paragan: do FTBFS have to be priority/severity high?? 05:56:17 I guess no 05:56:43 cause we are having source but just its upstream verification can not be done as upstream url is down 05:56:58 paragan: for fonts-ISO8859-2 ? 05:57:17 oh! is it set to high? 05:57:41 all (3) of them, yes 05:58:02 dychen: any idea about UnihanDb? 05:58:11 then reporter considered it as High priority 05:58:26 paragan: they are filed by ftbfs :) 05:58:51 bug 511680 05:58:52 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=511680 high, high, ---, pnemade, ASSIGNED, FTBFS fonts-ISO8859-2-1.0-20.fc11 05:59:01 bug 511482 05:59:02 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=511482 high, high, ---, dchen, ASSIGNED, FTBFS UnihanDb-5.1.0-7.fc11.1 05:59:07 and that is nirik I guess 05:59:08 juhp: That's because it tooks too long to build. 05:59:09 please have a look anyway 05:59:17 dychen: haha 05:59:19 hmm 05:59:32 So I was profiling it. 05:59:36 ok 05:59:41 I see 05:59:53 please add a comment anyway 06:00:34 anyway let's stop the meeting here if noone has anything else needing attention? 06:01:40 phuang_: tagoh3: Do you know any background of ibus-anthy icon? 06:01:53 fujiwarat: scim-anthy? 06:02:09 * paragan closed bug 511680 06:02:11 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=511680 high, high, ---, pnemade, CLOSED RAWHIDE, FTBFS fonts-ISO8859-2-1.0-20.fc11 06:02:11 yeah 06:02:13 * juhp wishes we could change it to something more meaningful 06:02:34 It's crown icon. 06:02:38 ifrom scim-anthy 06:02:43 from scim-anthy 06:02:43 fujiwarat: or you mean the upstream story? 06:02:47 I don't know why crown 06:03:07 I checked anthy home page but there is no icon. 06:03:15 fujiwarat: translating, ashie-san wrote "don't ask... ever" iirc 06:03:23 in his blog long ago 06:03:37 or maybe on the mailing list? 06:03:44 Personally I don't like the icon :) 06:03:51 fujiwarat: me neither 06:04:06 I encouraged tagoh3 to change it for fedora for scim-anthy ;-P 06:04:08 fujiwarat: You can "borrow" gcin's jp icon. :-P 06:04:09 it should be a user contribution - just a hope to becomes a king of japanese IM IIRC 06:04:15 tagoh3: thx 06:04:44 fujiwarat: how about stealing an icon from uim say? 06:05:04 juhp: what's the icon? 06:05:04 like あ 06:05:16 It's ok if it was suitable from scim-anthy. 06:05:52 If it can be changed, I'd like to consider the design. 06:05:55 juhp: it's not for IM icon but to indicate the input mode. they use the same crown for IME icon though. 06:05:57 anyway my "secret" disruptive plan is to try to get rid of the ibus IME icons completely if possible ;o) 06:06:15 do they? hmm 06:06:55 maybe I hid it or misremember perhaps 06:07:08 haven't played with uim in a while anyway 06:07:22 fujiwarat: anyway I support changing the ibus-anthy icon 06:07:22 need to create new one from scratch if we want to replace 06:07:38 tagoh3: why? 06:07:52 fujiwarat: any ideas for a replacement? 06:08:12 juhp: no better icons for _anthy_ I meant :) 06:08:28 I think あ icon would be an improvement 06:08:48 juhp: aha - thought you don't like a crown anyway 06:08:52 I don't see a strong need for something anthy specific 06:08:57 tagoh3: I don't 06:09:21 Yes, right. I just thought alphabet 'A' with red pen ;) 06:09:22 it is completely meaningless IMHO 06:09:47 fujiwarat: A for anthy? hmm I dunno 06:09:59 (the crown I meant) 06:10:26 maybe we just need a different UI anyway 06:10:40 at least the new ibus icon made me think so :-/ 06:11:09 juhp: just imagine from "improvement" you are about to work on it which is based on a crown icon, but anyway. 06:11:31 Probably we may discuss this later. I just wanted to know the background. 06:12:32 tagoh3: better crown? :) 06:12:54 juhp: heh. you said you don't like a crown :) 06:12:54 fujiwarat: I agree if anthy adopted some meaningful icon that would be better 06:13:11 tagoh3: I just trying to parse :) 06:13:46 sure 06:14:46 dunno I only suggested using something like uim's ja_hiragana.svg 06:15:17 fujiwarat: sure - hope we can ship something better that makes sense 06:15:53 OK, I see. 06:16:47 thanks for bringing it up :) 06:17:08 #endmeeting