05:31:03 #startmeeting i18n 05:31:03 Meeting started Tue Sep 25 05:31:03 2018 UTC. 05:31:03 This meeting is logged and archived in a public location. 05:31:04 The chair is paragan. Information about MeetBot at http://wiki.debian.org/MeetBot. 05:31:04 Useful Commands: #action #agreed #halp #info #idea #link #topic. 05:31:04 The meeting name has been set to 'i18n' 05:31:04 #meetingname i18n 05:31:04 The meeting name has been set to 'i18n' 05:31:04 #topic agenda and roll call 05:31:04 #link https://fedoraproject.org/wiki/I18N/Meetings/2018-09-25 05:31:04 # 05:31:15 ☺ 05:31:27 hi 05:31:35 hi 05:33:00 #chair mfabian tagoh pwu 05:33:00 Current chairs: mfabian paragan pwu tagoh 05:33:23 hi 05:34:15 #chair suanand 05:34:15 Current chairs: mfabian paragan pwu suanand tagoh 05:34:23 let's start the meeting 05:35:08 #topic Upcoming schedule 05:35:09 #info 2018-09-25 Beta Release (Target #1) 05:35:09 #info 2018-10-09 Final Freeze (*) 05:35:40 Fedora 29 Beta release got slipped to its target#1 date now 05:35:55 its new release date is on coming Tuesday 05:36:35 anyone got any issues to discuss for Fedora 29? 05:36:37 hi 05:37:05 * paragan checks for bugs status 05:37:12 The forward key event bug is very bad. 05:37:22 https://bugzilla.redhat.com/show_bug.cgi?id=1625142 05:37:33 Yesterday it was again rejected as a blocker. 05:37:48 hi 05:38:00 mfabian, for Final release blocker? 05:38:21 I checked again that ibus-hangul is default for Korean and one runs into that bug immediately after an installation in Korean. 05:38:24 okay I can see its proposed for F29 final release blocker 05:39:02 https://meetbot-raw.fedoraproject.org/fedora-blocker-review/2018-09-24/f29-blocker-review.2018-09-24-16.05.txt 05:39:38 okay 05:39:46 * (1625142) In Gnome Wayland, forward-key-event does not work well, 05:39:46 breaks space key completely (adamw, 17:00:29) 05:39:46 * LINK: https://bugzilla.redhat.com/show_bug.cgi?id=1625142 (adamw, 05:39:46 17:00:29) 05:39:49 * Proposed Blocker, gnome-shell, NEW (adamw, 17:00:29) 05:39:52 * AGREED: 1625142 - punt (delay decision) - we'd like some more 05:39:56 information on exactly when this happens, and particularly if 05:39:56 I think several patches are needed to fix it and I think another idea would be to revert the original patch of forward-key-event. 05:39:59 ibus-typing-booster is used by default for any language, before 05:40:02 making a decision (adamw, 17:14:23) 05:40:04 so that bug now need more information to be provided 05:40:12 it wasn't rejected 05:40:13 Therefore, I asked whether ibus-typing-booster is default for anything. 05:40:18 punt is a 'we didn't make a decision yet' 05:40:31 But even if ibus-typing-booster isn’t, ibus-hangul is certainly default. 05:40:36 adamw++ 05:40:36 juhp: Karma for adamwill changed to 7 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 05:40:50 Thank you Adam! 05:40:54 adamw++ 05:40:54 mfabian: Karma for adamwill changed to 8 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 05:41:00 mfabian: it is not 05:41:21 Any progress upstream on the review? 05:41:24 it'll automatically get reviewed again next monday, so please just add as much useful context as you can to the bug, it'll help us decide 05:41:36 I think if it is not merged, we will not see it in Fedora... 05:42:02 No progress, I think. 05:43:13 fujiwarat: If the new patches cannot get into Fedora, reverting to how it was in Fedora 28 is better than the current situation. 05:43:14 fujiwarat: you may be right that is it better to revert if that is acceptable 05:43:39 Does that have any other consequences? 05:44:43 mfabian: btw you said it also affects X Gnome? 05:44:59 Yes, I was surprised but apparently it does. 05:45:20 I think the reverting the original patch is no risk. 05:45:36 Okay then good to propose that as you did 05:45:41 When entering something with typing-booster or ibus-hangul into the "enter command" dialog, it happens even in Gnome-Xorg. (endless loop) 05:45:42 I think the forward-key-event bug exists in GNOME Wayland only. 05:45:52 I think we could reassign to carlosg? 05:46:02 or fmuller? 05:47:50 juhp: Probably carlosg. 05:50:10 Done 05:50:16 good to add some more information on that bug 05:50:26 That is also there now 05:50:44 juhp, thanks 05:50:56 adamw: do you think it is sufficient info now? :) 05:51:06 paragan: Wouldn’t typing-booster be a better default than ibus-m17n for Indian languages? 05:52:08 so to be clear: you all agree that Korean is affected out of the box, but no other language is? 05:52:14 mfabian, it can be but as per Government standard we have some default keymaps to use for languages 05:52:30 adamw: correct 05:52:39 that's the most important thing i guess, thanks 05:52:45 Yep 05:53:08 adamw: Yes, that seems right. I just tried Vietnamese because the Vietnamese input method ibus-bogo is also affected by this bug, but ibus-bogo seems not to be default after a Vietnamese installation. 05:53:19 paragan: well keymap and IM is really different things 05:53:20 ok 05:54:00 paragan: ibus-typing-booster uses the same keymaps as ibus-m17n, it only adds completion. 05:54:08 mfabian: maybe we should pull in bogo as a "langpack" for vi 05:54:11 juhp, yes IM can provide more functionality compared to keymap 05:54:20 paragan: The inscript2 keymaps are the goverment standard, right? 05:54:29 yes 05:55:24 mfabian, if you want to propose to make them default then I am fine 05:55:39 mfabian: might be interesting to explore that change for F30 05:55:41 paragan: ibus-typing-booster defaults to the inscript2 keymaps for the Indian locales. 05:56:04 juhp: Yes, I think it is no disadvantages but some extra features. 05:56:18 I long thought we should have Indic IMEs - well that partly led to the creation of typing-booster perhaps 05:56:31 ibus-bogo is not installed after a Vietnamese installation. Is that on purpose? 05:57:27 mfabian: I guess it would be on a netinstall 05:57:45 mfabian: just the usual story we don't have a way to do that yet 05:58:06 It is in @vietnamese-support 05:58:19 juhp: I tried an installation from live because I could do that faster (netinst takes much longer on my machine due to the download) 05:58:36 Right you won't get it for Live - live install is static 05:58:48 On top of that, something seems broken with the repos at the moment. 05:59:46 (Also it is being proposed to drop the Workstation os tree for F30, but of course netinstalls will still be possible) 06:00:47 [mfabian@Fedora-Workstation-Live-x86_64-2 ~]$ sudo dnf install langpack-vi 06:00:47 [sudo] mfabian のパスワード: 06:00:47 Fedora Modular 29 - x86_64 0.0 B/s | 0 B 00:10 06:00:47 エラー: Failed to synchronize cache for repo 'fedora-modular' 06:00:50 [mfabian@Fedora-Workstation-Live-x86_64-2 ~]$ 06:00:50 When we get langpack installation working in Gnome - we can start thinking about additional IMEs etc 06:00:53 06:01:01 I think that prevents me from doing netinstalls at the moment. 06:02:35 mfabian: you may need to update fedora-repos package perhaps? 06:04:04 tagoh: I tried a netinst using Fedora-Workstation-netinst-x86_64-29-20180923.n.0.iso and it gave an error on the page after the language selection, not being able to find the repos. 06:04:56 paragan: shall we move on? 06:05:23 okay 06:05:49 mfabian, try again and see if problem goes, maybe some mirror repo issue 06:06:08 #topic Outstanding issue 06:06:08 #info #102: Docs Beats for Fedora 29 (pnemade) 06:06:08 #link https://pagure.io/i18n/issue/102 06:06:21 we have worked on 3 Changes this release 06:06:25 we should document them 06:06:37 mfabian: seems okay for me here 06:07:10 I see IBus and CJK not fonts Change already have their own tickets in fedora-docs repo 06:07:28 We need to do a PR? 06:07:43 A reminder would be helpful 06:07:51 of the process I mean 06:07:58 I think bcotton missed fedora-docs ticket for https://fedoraproject.org/wiki/Changes/Liberation_Fonts_2 06:08:16 Should we ask? 06:08:48 We can also create ticket there 06:09:26 Okay the process is in the tickets 06:09:37 True... 06:09:43 vishal_vvr, please report release notes ticket for Liberation fonts Change to https://pagure.io/fedora-docs/release-notes/issues 06:10:30 vishal_vvr, see https://pagure.io/i18n/issue/102 06:11:21 vishal_vvr: maybe ask bcotton first? 06:11:58 juhp, sure will ask him first 06:12:01 Perhaps because it was quite late submission 06:12:13 probably 06:13:21 so either individual Change owners if want can submit PR to https://pagure.io/fedora-docs/release-notes for their Change or if you want me to create PR then do tell me so 06:14:16 any questions on this? if not we can move to next ticket 06:15:51 #topic No assignee issues 06:15:51 #info #97: Replace comps language support groups with weak dependencies (petersen) 06:15:51 #link https://pagure.io/i18n/issue/97 06:15:56 tagoh: do you want to add some details to https://pagure.io/fedora-docs/release-notes/issue/195 ? 06:16:26 juhp: sure. will do 06:16:33 fujiwarat: details for https://pagure.io/fedora-docs/release-notes/issue/190 ? 06:16:35 great thanks 06:17:14 I have a plan to push relnotes. 06:17:30 good thanks :+1: 06:17:38 Please 06:18:32 When is the right time to address the language groups? 06:18:40 pwu: Maybe better to note hangul default mode. 06:18:58 It might be a good feature for F30? 06:19:02 fujiwarat: good point 06:19:28 fujiwarat: the patch is not merged. 06:19:42 I think that is fujiwarat's point :) 06:20:00 or did we patch fedora? 06:20:04 Probably I can describe it together with ibus core. 06:20:17 sure 06:20:37 not patched Fedora yet. 06:20:49 mfabian: ibus-bogo could be a good testcase for IME langpacks 06:20:54 I guess it's an intention. 06:21:03 juhp: Yes. 06:21:04 Right 06:21:16 juhp: You said the netinst works for you?? 06:21:24 shall I put notes in https://pagure.io/fedora-docs/release-notes/issue/190 for ibus-hangul? 06:21:28 mfabian: I said it should work :) 06:21:38 mfabian: I meant dnf update works for me 06:22:24 juhp: I get this for about a week now: https://imgur.com/a/4uUMn9r 06:22:37 "Error setting up base repository" 06:23:50 pwu: Probably you can update git but not sure. 06:23:55 pwu: sure you can comment of course - though I guess strictly it is not part of ibus-1.5.19 - maybe too much overhead to a have a separate ticket for that 06:24:22 mfabian: maybe you have a bad mirror? as tagoh suggested, dunno 06:25:01 juhp: let me comment in the ticket for ibus-hangul. 06:25:05 mfabian: I meant netinstall should pull in ibus-bogo today 06:25:23 .whoowns ibus-bogo 06:25:24 juhp: I’ll try that as soon as netinst works again for me. 06:25:25 juhp: owner: tuanta 06:25:38 juhp: Maybe I have some network problems ... 06:26:10 Most people do Live install anyway I suppose 06:26:47 juhp: And after a live install, dnf update doesn't work for me, probably same reason why the netinst doesn't work. 06:27:05 how about ibus-cangjie? dunno if any other IME should be pulled in for any langs? 06:27:34 mfabian: yeah sounds bit strange - which mirror? 06:27:52 mfabian: you could override the mirror in anaconda 06:28:27 juhp: Maybe I should try without vpn, sometimes this makes a difference. 06:29:34 we have reached one hour time. 06:30:01 Let's close the meeting and you can continue discussion in #fedora-i18n 06:30:26 before closing anyone got any other issue to present here? 06:32:12 Thank you all for coming to this meeting 06:32:15 #endmeeting