=== asac_ is now known as asac [07:16] fta: yeah ... thats good news [07:16] fta: i convinced them ;) [07:17] now i have to hunt down those kernel folks that broken my iwl setup [08:36] last ffox/xulrunner test build [08:36] for last upload to hard [08:36] y [08:36] fta: ^^ [08:36] carlos: hi! ... [08:36] a question ;) [08:36] asac: hi [08:36] sure, go ahead :-P [08:37] i think jtv said that the reason for all those untranslated strings is that we import the same LANGCODE.xpi to both packages [08:37] carlos: i have now a script to filter that [08:37] so we could split the upstream .xpi up before uploading translations [08:37] would that help? [08:37] s/the reason/one reason/ [08:38] i assume there will still be ambiguities left, but maybe we can reduce the amount of not properly mapped strings from 250 to 20 per package ;) [08:38] asac: I don't think it will make a huge difference, our problem is that we don't read manifest files yet [08:39] and thus, we find files from manifest *and* files that are not referred in the manifest and that have duplicated IDs [08:39] so we handle that as a conflict, which is not such conflict in en-US [08:39] and thus, the messages are tagged as different [08:39] carlos: yes, but i can filter the de.xpi for instance to only include files that are in manifest [08:40] asac: hmm, that would help, yes [08:40] carlos: https://code.edge.launchpad.net/~mozillateam/rosetta/po2xpi [08:40] tere is a runxpi2xpi script ;) [08:50] carlos: did you manage to kick es translators to fix their accesskeys? [08:50] id love to include es from launchpad [08:51] but i need to be sure that the accesskeys are fixed as good as possible ;) ... otherwise mozilla corp will sue me ;) [08:51] no luck yet, sorry. I will use email to be sure I get an answer :-) [08:56] carlos: do you know a nick of a translator? [08:56] asac: btw, is that just for firefox or xulrunner too? [08:56] asac: yeah, but he only connects when he needs something from me :-P [08:56] carlos: xulrunner is even more important because everybody appears to forget that its half of the text [08:56] that's why I didn't have luck looking for him [08:57] asac: talking about the acesskeys [08:57] carlos: finnish translation accesskeys were fixed in 10 minutes ... should be really easy for him ;) [08:57] carlos: i added a note to the firefox 3 template but apparently nobody reads that [08:57] carlos: there are a few _new_ languages started ... they of course only started on firefox project [08:58] carlos: any idea how we can help translators to notice that there is still work to do? [08:58] yeah, I saw your note about xulrunner [08:58] carlos: i bumped priority to 1000 ... maybe i should go for a million? [08:58] well, I could give higher priority to xulrunner [08:58] that will make it appear near the top of lists to translate [08:58] i can do that too? at least i thought i can [08:59] hmm, not sure... let me check... [08:59] i can defeinitly set priority for my templates [08:59] no idea if they help though [08:59] what would be a reasonable value? [09:00] indeed, you are able to change it [09:00] well... the problem is that the template is huge [09:00] and thus, I guess people get scared :-) [09:00] carlos: why scared? lots of karma to gain i'd say :) [09:01] imo karma should be multiplied by priority :) ... so i could reward folks with a higher karma by bumping prio ;) [09:02] :-P [09:02] carlos: lets say we sort out the import problems later ... could we overwrite those that are currently manually confirmed? [09:02] asac: https://translations.edge.launchpad.net/ubuntu/hardy/+lang/es [09:02] you can see there that firefox got higher priority [09:02] however, if you want xulrunner, it should have higher priority than firefox ;-) [09:02] cool [09:03] carlos: so 1000 was enough to make it show up? [09:03] on top? [09:03] * asac looks [09:04] ok dumped firefox to 900 again ;) [09:04] for last upload to hardy <= ? when ? b5 or pre ? [09:05] b5 [09:05] now [09:05] fta: mozilla corp didn't grant us exception to use official branding for a glorified nightly [09:05] so we have to eat this [09:05] ok [09:05] its all on .dev branch [09:05] i will merge it to .head if you don't come first [09:06] i don't have much time at the moment, go ahead. [09:06] fta: sure [09:06] thanks [09:19] asac: xulrunner is now the topmost translation of all, and firefox second. they were the top two two/three days ago already, so they should definitely be noticed. people might be scared away from translating the access keys since the actual context is shown only at the bottom of the string box in small text and has to be "parsed" to human language ("en-US.xpi/en-US.jar!/locale/browser/browser.dtd(printPreviewCmd.accesskey)" -> "ah, p [09:20] basically a little bit of information "yes, you can/should translate those, and look at the string to parse the context" should be enough. should I send an e-mail to ubuntu-translators mailing list? [09:20] (or carlos) [09:22] Mirv: yeah. i think we are fine now [09:22] Mirv: we will have three languages from rosetta in the first hardy langpacks [09:22] that should be enough to claim this a victory :) [09:22] after all all those disambiguitites that currently need manual action should go away in future imports when rosetta has improved [09:22] asac: alright :) [09:23] yep. [09:23] personally i just want to ensure that all this works, so _new_ translations can be done in launchpad [09:23] there is not much sense in redoing the upstream work [09:23] unless we find a procedure to review translations and give back improvements to upstream [09:23] at the end of each dev cycle [09:24] (and if upstream disagrees, remove our diff again) [09:24] Mirv: isn't an info message appearing explaining what's that key for? [09:25] carlos: i think its just a "warning" [09:25] if it doesn't appear, is a bug [09:25] asac: the text we agreed should appear in each message [09:25] asac: well, it's an info message, yeah [09:25] carlos: yeah ... but thats just a warning "don't change translation if you don't know what to do" [09:25] which is correct [09:26] hmm, didn't we add an explanation for it? [09:26] point is atm, it should read "please don't change existing translation, but if ther eis no translation check the menu entry this refers to and choose a letter that's in the word" [09:26] hmm, too many things in my brain... [09:26] carlos: two problems: 1. it says "Select the shortcut key you want to use" both when it's commandkey and accesskey, and it takes some knowledge to understand that not all of those are similar. 2. the actual context is only in the "Located in" text. [09:26] I see [09:27] asac, Mirv: That's a really easy change in our side, please, could you file a bug against Launchpad Translations [09:27] yeah ... its hard to spot which menu entry it actually refers to [09:27] with a suggestion for each type of key? [09:27] carlos: well ;) ... i have no idea how to best explain it in a short paragraph [09:27] carlos: I can try [09:27] above was just an example that sprang to my mind [09:27] asac: that's why GNOME and KDE note it as part of the menu translation ;-) [09:28] Mirv: it could be different for each kind of ID so commandkey and accesskey would have different explanations [09:28] btw, I found also a .key [09:29] so as far as I know, there are commandkey, acceskey and key [09:29] ? [09:29] carlos: yeah ... they can be anything [09:29] hmm, I wonder what that .key was [09:29] i found several cmdKey [09:29] as well [09:29] and others ... i think we cannot really find all [09:29] ah, same as commandkey/cmdkey [09:31] carlos: once import is perfect, can we overwrite anything yet translated/confirmed? [09:32] or at least detect differences so we can manually review them? [09:32] asac: we can do some cleanup to use whatever upstream has vs. what we have in Launchpad, yes [09:33] our translators are able to do it directly using the UI, once we fix that bug with not parsing manifest files [09:34] carlos: in the end i need a way to produce diffs that we can submit/discuss with upstream [09:34] (VISION) [09:35] hmm [09:35] and incorporate their feedback after that kind of review round [09:35] asac: Is a diff between two .po files enough? [09:36] not sure about the format yet ;) [09:36] ok [09:36] we have the information in our database [09:36] we even have a way to filter all messages [09:36] and show differences with upstream [09:36] in the web UI [09:36] yeah. in a perfect world we could show them diffs against their translations in CVS [09:37] so they could apply them easily [09:37] but thats nothing for right now :) [09:38] i think in the long run we need a full time upstream translation liason that manages this :) [09:39] until upstream finally switches to rosetta ;) [09:45] :-P [09:45] Mirv: btw, I don't find the email you sent with instruction about how to test firefox lang packs [09:45] carlos: puuh. the access key thing is really hard to explain.. https://bugs.launchpad.net/rosetta/+bug/217626 [09:45] Launchpad bug 217626 in rosetta "Firefox / Xulrunner better explanations for command / access key translating" [Undecided,New] [09:46] Mirv: I know :-P I tried and failed already... [09:46] Mirv: I remember I saw your instructions (I think you sent it) [09:46] but I don't remember were [09:46] hmm maybe planet ubuntu? [09:47] carlos: no, I think you're thinking about someone else? I think asac was planning a blog post about it. [09:47] (and maybe did so) [09:47] Mirv: i plan a blog post about how to produce test translations from .po file exports [09:48] i am not sure how to properly describe the accesskey mystery in a way that anyone would understand [09:48] suggestions welcome ;) [09:48] Mirv: you did it right now, so maybe you have a better idea about this as an "outsider" [09:48] i mean, i probably know to much about firefox internals to describe it in a good fashion [09:49] asac: yep, I tried now to have a semi-short explanation in that bug report I just referred to, so Carlos could put something like it in Rosetta [09:49] Mirv: I think the description is fine [09:49] Mirv: thanks [09:49] * asac looking [09:52] Mirv: ok, found [09:52] Mirv: I was talking about the text you sent to ubuntu-translators yesterday [09:52] :-P [09:53] ubuntu-translators IRC channel [09:53] carlos: ah, yes that one you're correct. that's the only place I wrote about it :) [09:53] blog/irc/e-mail, all the same [10:07] (ok I should probably do now my real work again ;)) [10:07] Mirv: thanks [12:19] fta: doors closed. last package up. me on holiday ;) [12:20] how much i hoped that all points were true ;) [12:24] * asac lunch [13:57] hi all > fta > ubulette ? [14:00] *ping* [14:25] plun: ? [14:25] hello ;) [14:25] plun: whats your question? [14:26] I have a challenge... a Ubuntu killer started in my country, this is a new news channel, I cannot figure out how the streams are built.... [14:27] URL http://anytime.tv4.se/webtv/?progId=427494&treeId=901101&renderingdepartment=2.757 [14:27] Ladda ner Windows Media Player här. [14:27] Well... ;) [14:28] tested with totem-mozilla and mozilla-mplayer [14:28] and vlc? [14:28] Yup VLC brakes and also mplayer from CLI [14:28] brakes? [14:29] breaks? [14:29] Yup ;) [14:29] INET6 error bla bla [14:29] plun: how do you get the stream from CLI ? [14:29] whast the URL for the stream= [14:29] ? [14:32] plun@dunder:~$ mplayer mms://anytime.tv4.se/webtv/metafile.asx?MSG=mc1lSbUlityW0cJiqRrJ(xfyWIKhyDr4FyDJrUewlu8RtZiUhdDWyxd8hFg3ymUtDI0KqynoVaj0NAC23mivSX1HlWyFdo2RlRfbRjb0qNM! [14:32] bash: syntax error near unexpected token `(' [14:32] plun: please start firefox from command line [14:32] plun: and post the output you get there [14:32] OK [14:33] plun: when you use the mplayer, put the url like this: mplayer 'mms://any...NM' [14:33] ie use ':s [14:34] OK testing.... I am using Chatzilla so I cannot close my browser.. ;) [14:35] plun: try: [14:35] totem 'http://anytime.tv4.se/webtv/metafile.asx?MSG=mc1lSbUlityW0cJiqRrJ(xfyWIKhyDr4FyDJrUewlu8RtZiUhdDWyxd8hFg3ymUtqZas4oE1y6ft2I)1FlATFCALGsAD8kTOICOIvr72a0c!' [14:35] that works for me [14:35] it asks me to install missing plugins [14:35] (which isn't working in totem plugin unfortunately) [14:35] but after installing codec it should work [14:36] hmm but doesn't for me :) [14:36] now totem locks up [14:37] Ok, here is the error output [14:37] http://paste.ubuntu-nl.org/63311/ [14:37] '***' was needed [14:38] plun: did you manage to install the codec like above? [14:38] why mms? for me http://... triggers the proper codec install [14:39] It is 2 commands first with mms then with http. [14:39] I just copy the adress from mozilla-mplayer GUI [14:40] plun: i think the url is just dead [14:41] No they are not dead.... dead for Ubuntu and Linux maybe. [14:41] Tested on a Windows PC an hour ago. [14:42] This is real sabotage against open source and makes users to abandon Ubuntu. [14:42] plun: does it work with firefox 2? [14:43] I dont know.... I uninstalled FF2 long time ago... ;) [14:43] plun: that site is crap anyway. you should send email and complain [14:43] i mean: Due to rights limitations, this video cant be offered in the country or region where you are located. [14:43] i wouldn't be shocked if that site uses DRM mechanisms as well [14:44] who knows if its the same issue :) [14:44] Well, mostly all commercial sites using DRM and the world is indeed commercial. [14:44] Just to find a way around this.... :) [14:45] plun: ? [14:45] thats not true anymore [14:45] lots of companies have reverted DRM [14:45] as they found that noone wants that [14:45] i think time-warner and universal have stopped using DRM [14:46] Well, thats just a theatre, TPM is coming instead... [14:46] [14:46] the idea is to point other companies to such examples and tell them that they have no clue [14:46] TPM won't come i guess [14:47] anyway ... no time for such things right now ;) [14:49] Well... it will 100% sure... all security agencys already uses it, goverment PCs and son on. [14:49] Every mobile we buy probably includes that terrible chip.... ;( [14:49] Intel is coming with it Built-in, former called LaGrande. [14:51] Back to breaking the stream... :) [14:51] Thanks for answering ! [16:15] carlos: i think there will be a bogus en-US.xpi (missing install.rdf) going up for firefox-3.0 package that comes now. can we overload that with the one we currently have in case it really is catched? [16:15] asac: I can remove it from the queue, yes [16:15] asac: firefox and xulrunner [16:15] or just firefox? [16:16] carlos: just firefox [16:16] the next upload will fix it [16:16] carlos: i don't think its already in the queue ... do you see? [16:16] let me check if the package has been build yet [16:17] (we had some buildd issues an hour ago or so) [16:17] asac: no, it's not there yet [16:17] https://translations.edge.launchpad.net/ubuntu/hardy/+source/firefox-3.0/+imports [16:20] ok ... lets hope we are awake when the buildd's are cured again [16:30] carlos: with some luck OOO lands before firefox :) ... i guess that should help to digest the import queue ;) [16:30] 000 lands? [16:31] carlos: well https://edge.launchpad.net/+builds ... at least its building [16:51] asac, I have the black rectangle bug on ati [16:52] fta2: which site? [16:52] I open a png in a tab: perfect, change tab, return to the png: it's a black rectangle, changing tabs again doesn't solve it [16:52] http://upload.wikimedia.org/wikipedia/commons/2/20/Keyboard_layout_Chinese_Traditional.png [16:53] let me upgrade everything here [16:54] fta2: cannot reproduce here with fglrx [16:54] hmm [16:54] refresh help [16:54] let me check once everything is updated [16:54] but it's broken again if i change tabs [16:54] fta2: maybe due to extensions? [16:54] oh, reset zoom fixed it [16:54] he? [16:55] i already tried zooming [16:55] can you reproduce at all now? [16:55] nope [16:56] fta2: do you already have the new xserver? [16:56] which version is that? [16:56] apt-get dist-upgrade doesn't bring me anything new :( [16:56] and zooming the png itself produces black areas [16:56] fta2: try add Option "XAANoOffscreenPixmaps" "true" [16:57] i think the xserver with that fix was not yet rolled [16:57] i'm not uptodate [16:57] anywy ... try that option ... the4 new X server should just set that as default [16:57] nothing else is new [16:58] xorg-server (2:1.4.1~git20080131-1ubuntu8) hardy; urgency=low [16:58] thats the one with the fix [16:58] lets see [17:00] k looks like its already installed here ... let me try to remove that option and see if its still fixed [17:00] ok restarting X [17:05] bummer [17:05] fta2: its not fixed [17:05] i have the feeling our x maintainer made a mistake when porting that patch from fedora 1.5 [17:05] fta2: the setting should fix it though still [17:06] ok let me try to fix xserver [17:08] ok lets see whats going on [17:32] http://ubuntuforums.org/showthread.php?t=755730 [17:38] fta2: n00b ... he didn' tadd that patch to series [17:38] tse [17:38] nothing got applied [17:39] lol [17:39] and the description was not very clear [17:40] are users supposed to set XAAxxx = true themselves ? [17:40] it's supposed to be deprecated [17:44] fta2: yes. now its XaaOffscreenPixmaps "True" [17:44] uploaded fix [17:48] lets see if admins are happy with it ;) [17:48] RMs rather ;) [17:56] fta2: opened bug 186186 for cairo again and milestoned it as "later" (aka intrepid) [17:56] Launchpad bug 186186 in xulrunner-1.9 "web page background render errors" [High,Confirmed] https://launchpad.net/bugs/186186 [18:22] fta2: ok accepted [18:22] (xserver [18:22] ) === jetsaredi1 is now known as jetsaredim === willguaraldi_ is now known as willguaraldi [21:09] newz2000: ok we are waiting for mdke [21:09] oh, its mdke. :-) I should have guessed [21:09] hi mdke [21:09] mdke: newz2000 [21:09] heya newz2000 [21:10] mdke: he probably has all the answers you want ... and you can probably provide answers to him :) [21:10] win-win ;) [21:10] asac: I wouldn't go as far as saying I know how things work... what "things" are we talking about here? [21:10] not sure ;) [21:10] newz2000: ? [21:11] Well, there's these offline HTML pages that show up by default in firefox when you open the browser and haven't customized the start page [21:11] I was wondering how these pages are generated, because we're also going to create online versions [21:11] newz2000: they are translated manually by translator teams and uploaded to ubuntu-docs [21:12] newz2000: except the original, which was just written by us and gets its version number bumped each release [21:12] So is there some script that generates the separate html pages? [21:13] no [21:13] it's just one page for each language [21:14] mdke: does the translation make use of rosetta? [21:14] or just hand-written submissions? [21:14] asac: no, it's done manually [21:14] oh, I see [21:14] So each person copies the html page and then changes the text in the HTML to match their language? [21:15] yes, and the links, generally [21:15] newz2000: maybe we want to do XML in future? [21:15] we avoided during xml because it's much harder to customise the output when building html [21:15] during/using [21:15] we tried it though, iirc [21:15] well, it looks like the pages are all valid xhtml [21:15] what customisations are you referring to? [21:16] so I can use beautiful soup to parse it and get what I need [21:16] asac: if you build html from docbook xml or sgml, you need to customise things a bit to build them [21:17] newz2000: can you explain what it is that you're doing? [21:17] the mechanism to allow firefox to show translations and to show the correct derivative start page was quite delicately constructed, I'm concerned that changing the system at this late stage is a bit risky [21:17] we're creating an online version of the page. It has some customized things on the side to make it interesting and it looks different than the offline page. [21:18] so if the user runs kubuntu will they see a kubuntu page? [21:18] for hardy only FF is affected [21:19] right, but firefox is the default browser for loads of derivatives, no? [21:19] correct [21:19] i thought it was for kubuntu too [21:19] and edubuntu etc [21:19] I don't know about that [21:20] so if the user runs kubuntu will they see a kubuntu page? [21:20] whoops. [21:20] the page has been made to be as generally interesting as possible, there's no ubuntu specific content [21:20] I see. but the current version has the Ubuntu colours, maybe it's just that I haven't seen the new version yet [21:20] it looks like the current ubuntu website with hardy theme elements [21:21] a heron in the top corner [21:21] its pretty simple [21:21] it sounds nice [21:21] but it seems pretty unfair if kubuntu, xubuntu and edubuntu users are going to lose a customised start page because of it [21:22] its exactly the same content as the offline page [21:22] <[reed]> http://www.ideastorm.com/article/show/63082/Preinstall_gNewSense_on_laptops_and_desktops [21:22] <[reed]> lol [21:22] plus a search box, plus some links to interesting ubuntu info and headlines [21:23] don't get me wrong - I like it, and I think there has been a place for an online homepage for some time. [21:24] but if users of branded derivatives used to get a page with their derivative's branding, and now they will get an ubuntu branded page, that needed to be addressed before implementing this [21:24] I didn't see any public discussion of the idea at all, I definitely would have pointed that out if I had [21:24] asac can correct me if I'm wrong, but we're not replacing a branded page [21:24] aiuc [21:25] mdke: it only applies to firefox [21:25] we talked to kubuntu devs and as always they didn't care [21:25] asac: I get that bit. But kubuntu, edubuntu and xubuntu all have firefox, don't they? [21:26] actually they didn't even care that thei local startpage was not properly localized in firefox [21:26] that was the whole point of including different startpages for them [21:26] and they never reported that to me. [21:26] mdke: no ... kubuntu has konqueror [21:26] asac: we've had a bug about it in ubuntu-docs which was recently marked as fixed [21:26] not sure about xubuntu. [21:26] kubuntu-docs, sorry [21:26] mdke: its not fixed [21:26] the kubuntu-docs package has the translated pages, but it doesn't have proper links setup [21:26] asac: ok, that sometimes happens with bugs [21:26] everything is still pointing to index.html [21:27] I see [21:27] i found that while tesitng the localization feature [21:27] I guess the developer thought that including the translated pages would fix it, without testing properly [21:27] most liekly [21:27] mdke: however the same bug was in gutsy i checked that [21:27] before gusty no translation existed [21:27] yeah [21:27] so back to the branding [21:28] i am not sure about edubuntu and xubuntu [21:28] I thought edubuntu was gone [21:28] !! [21:28] newz2000: yes. its an addon cd [21:28] now [21:28] right [21:28] newz2000: please ask ogra if they still patch the startpage [21:28] so left is xubuntu ... thats not a big problem either imo [21:29] point is that all this is shipped by ubufox [21:29] they could easily do xubufox [21:29] edubuntu-docs still has a startpage [21:29] but nobody came to me to ask what to do to make that happen :) [21:29] ok [21:29] same for edubufox and kubufox [21:29] :) [21:29] if there is demand we can do something like that [21:30] edubuntu never patched the start pages [21:30] in worst case only ubuntu will have the online startpage [21:30] to be honest, I only found out yesterday that the functionality for the browser homepage had been changed, and I maintain ubuntu-docs which ships the homepage [21:30]