[02:13] How can I test Unity 8 on 15.10? [02:14] g0twig: sudo apt-get install unity8-desktop-session-mir [02:14] Mir aha [02:14] And then log out and choose it on the login screen [02:14] I use nvidia/nouvou [02:15] are unity next iso images available, I dont have ubuntu installed [02:15] g0twig: The above command works on plain Ubuntu 15.10 [02:15] You don't need special images [02:15] so you suggest me to download the normal ubuntu [02:15] and than install the packages [02:16] but I dont think unity 8 is fun yet [02:16] g0twig: Yes, because Unity8 is not yet complete. You likely won't want it for every-day use yet [02:16] hm ok [02:16] On the login screen you can choose what shell to use [02:16] hm I dont know if its worth it [02:17] downloading ubuntu for this and than just testing it 3 mins and quitting [02:17] g0twig: Oh if just testing then yeah use the special images (boot from DVD/USB) [02:17] where are the special images [02:17] Haven't tried them myself [02:17] Hang on [02:18] i think they are not longer available [02:19] g0twig: Oh, you're right! They're gone. [02:19] https://wiki.ubuntu.com/Unity8DesktopIso [02:19] but you dont think its worth it right? [02:20] g0twig: Ubuntu 15.10 is by itself :) [02:20] but I use arch^^ [02:20] You can then log out and switch back to Unity7 on the login screen [02:20] g0twig: OK, maybe wait till more convenient images are available then [02:21] too bad unity 8 doesnt hit ubuntu 16.04 :/ [06:29] duflu, afternoon, can you please uncommit/overwrite your latest trunk merge on the unity8 branch? train gets scared when there's new commits (and the silo is under testing for a day or two already) [06:58] Saviq: Didn't know it landed (!?) [07:00] Saviq: It didn't land. Apparently. What branch do you mean? [07:00] Oh, you mean my private branch [07:03] Saviq: Done [07:03] The new revision didn't have any effect in waking Jenkins up anyway [07:34] Saviq: so are we supposed to switch to xenial already' [07:34] ? [07:39] duflu, that's because it was Approved already [07:40] tsdgeos, xenial == wily+overlay today, really [07:40] tsdgeos, so wily+overlay is no more, I won't tell you to *move* to xenial, but you need a way to build/test for it [07:41] so a chroot at least [07:41] duflu, thanks [07:41] i can do xenial for now and let's hope distro people don't make it explode D: [07:42] tsdgeos, well, it will get a bulk-sync from Debian ;) [07:42] so caution advised [07:42] but yeah /me moves to xenial too [07:44] tsdgeos, ppa-purge overlay first, though [07:46] right we don't need the overlay from now on [07:50] Saviq: otoh xenial is old? [07:50] http://paste.ubuntu.com/12900852/ [07:51] tsdgeos, we don't have britney in action yet, so things are in xenial-proposed [07:51] e.g https://launchpad.net/ubuntu/+source/unity8 [07:51] https://launchpad.net/ubuntu/+source/unity8/8.11+15.10.20151009-0ubuntu1 [07:52] is britney a person? [07:52] * tsdgeos clueless [07:52] should settle down in the next days [07:52] no, it's the proposed migration tool [07:52] the one that runs tests&co to validate updates or block them [07:52] so i guess it was a bad call to switch :D [07:52] https://wiki.ubuntu.com/ProposedMigration [07:52] seb128: "next days" == 1-3 or 10-15? [07:52] well, at this time xenial = wily [07:53] so switching is not a bad call [07:53] just not useful [07:53] next days is today or monday [07:53] still need the overlay [07:53] ok [07:53] they are bootstrapping the new serie [07:53] tx [07:53] yw [07:54] oh thought things are already all in place since the train already switched [07:54] well the serie is open [07:54] the migration scripts are not active yet [07:54] so still some work [07:54] yeah, sure [07:56] the release team is in London, now that wily is out I guess they are going to focus on getting xenial going today [08:30] hello! how do I run unity8 in a container on wily? [08:32] Saviq: don't understand what you mean with the binding [08:32] in https://code.launchpad.net/~aacid/unity8/new_and_old_audio_role/+merge/275365 [08:35] tsdgeos, property Item audio: { try { } catch() { } } [08:39] Saviq: fpff, not sure it's more readable tbh [08:43] tsdgeos, I'm wondering if we should increase the touch area of Clear All in recent searches [08:43] tsdgeos, on krillin is quite a small target [08:43] tsdgeos, I just hate onCompleted, but won't push [08:55] cimi: the vertical one? [08:56] it's like the others, small because the mockup had this === davidcalle_ is now known as davidcalle [08:56] but once ew give for designers to test they can comment on it [08:56] it's quite easy to change [08:58] tsdgeos, so we don't have design approve? [08:58] no [08:59] still need to be shown to them for the final bits [08:59] oki [09:00] there's still one thing pstolowski has to finish so design can complain on how ugly it looks :D [09:28] cimi: can you update https://code.launchpad.net/~cimi/unity8/sdk1.3_newUbuntuShape/+merge/271610 ? [09:29] tsdgeos, sure I can [09:32] tsdgeos, we are still aiming at 1.3 for OTA8? [09:32] yes [10:07] tsdgeos, ok pushed the first one, let's see if tests break on CI [10:08] doing new shadows 1.3 now === broder_ is now known as broder === \b is now known as benonsoftware [10:16] done too [10:22] Saviq: so i went to a binging beucase the onCompleted was getting confused with other oncompleted in the parent [10:23] tsdgeos, had trouble with the "binging", thought you were talking about some TV series [10:24] tsdgeos, but yeah, another reason why I hate onCompleted for internal bits (could alternatively do it on a priv object) [10:24] lol [10:24] sorry === tjaalton_ is now known as tjaalton [10:37] cimi: do you think you could also have a look at https://code.launchpad.net/~aacid/unity8/dash_reset_instead_of_fatal/+merge/274363 or want me to find someone else? [10:38] tsdgeos, I can [10:38] cool === davmor2_ is now known as davmor2 [11:50] tsdgeos, another small bug, the "Cancel" button in the header does not follow the color for the foreground, try with CNet [11:50] let me see === alan_g is now known as alan_g|lunch === popey_ is now known as popey [12:07] Hello everybody, [12:07] I work for Mycroft project and we are starting some demo integration for Unity 8. [12:07] Therefore, we'd like to know what is the best way to start playing with it. [12:07] We've followed this tutorial: https://unity.ubuntu.com/getinvolved/development/unity8/ [12:07] However, it seems to lack more information for unity development. [12:07] 1. Besides that webpage, is there any other Unity 8 documentation or tutorial for developers? [12:07] As Mycroft client is in Python it'd be important for us to run it natively in mobiles. [12:07] 2. Anyone knows if Python will be natively supported for Ubuntu mobiles versions? Are there any restrictions or particularities for that to happen? [12:07] Finally, for our POC, we would love to autostart Mycroft client as a background process as soon as Unity is started. [12:07] 3. Any tip on how we could run some python or shell scripts during the start of Unity 8? [12:07] We are very excited to integrate Mycroft with Unity and make Ubuntu to become the most intelligent opensource OS out there. [12:07] So any help from the Unity developers and community will more than appreciated! [12:07] Thank you a lot. [12:08] jdorleans, I can guarantee Python is supported on Ubuntu, regardless of the device (I'm already relying on it for some stuff on the phone) [12:09] great! that's very important to know [12:13] jdorleans: hey, documentation on developing unity8 pretty light. What kind of development are you considering? Are you working on an application for unity8? [12:15] jdorleans: on point 3, the process manager (upstart for us currently, but transition happening to systemd eventually) allows you to run scripts after certain events - one event being unity8 started, another event being after the user session has started [12:30] oh, I see... I'm glad to know that. For the development in unity we haven't plan nothing special for now. However, it'd be very cool to start studing how unity pieces work. [12:35] greyback, do you know where I can find docs or maybe example on how to listen for these events so I could execute my process after that ? [12:37] mzanetti: do you want to approve https://code.launchpad.net/~mterry/unity8/shutdown-dialog-on-resume/+merge/275240 or should i? [12:48] jdorleans: the upstart documentation is quite comprehensive: http://upstart.ubuntu.com/cookbook/ [12:48] jdorleans: /usr/share/upstart/sessions contains a bunch of upstart config files you can have a look at [12:49] jdorleans: something like unity8-dash.conf might interest you, as that launches straight after unity8 === alan_g|lunch is now known as alan_g [12:59] tsdgeos, I can give it another test [13:00] ah you did already [13:00] feel free to approve then [13:00] tsdgeos, ^ [13:00] oki [13:01] cimi: fixed the color thing [13:08] tsdgeos, ugh, why the if (realAudio === null) hoop? [13:08] tsdgeos, if you've a priv object, could very well do onCompleted there, if we need to have an indirection like that? [13:10] the oncompleted was there already and it wasn't working [13:10] i am not sure we need the realAudio thing [13:10] but it makes it much more easy for my mind to understand [13:10] :D [13:11] more difficult for me than a comment ;) [13:11] tsdgeos, onComplete was in the priv, and wasn't working? [13:11] it's strange but the tests were showing the onCompleted of the parent being executed first [13:11] i can try again if you want [13:11] tsdgeos, I'd just go for a single property + comment [13:12] tsdgeos, thought that you had to go to === null to get it working in the first place (and that onCompleted on the top component was where the problem existed) [13:12] (in the first place → with the binding) [13:13] dandrader, hey, I've just tested the noStrechOnResize branch on top of qtmir, and I've noticed that the mouse speed has decreased *a lot* [13:13] I don't think it is because of that branch but must have happened recently [13:13] mzanetti, it's the mousePointer branch [13:13] can we fix that? it's unusably slow now [13:13] Saviq: ok, i'll retest again if you don't like this solution either [13:14] I need to move 3 times over the complete touchpad in order to move the cursor half the screen [13:14] tsdgeos, wait, since we have three competing solutions :) [13:14] tsdgeos, onCompleted, two props with === null, one prop with comment [13:14] tsdgeos, I thought you went for the second because the third didn't work [13:14] mzanetti, I thing anpok_ has plans for fixing that ^^^ [13:14] *think [13:15] tsdgeos, if the third works, I'd go for that, you wanted onCompleted, and I'm fine with that too :) [13:15] ah ok. anpok_, any details? [13:15] anpok_, talking about the cursor being slow once we use relative movement axes from the event [13:15] isn't it because of DPI? [13:16] i guess it is slow because it isnt acclerated [13:16] Saviq: i didn't try just one property because it makes me uneasy it will be evaluated again somehow [13:16] anpok_, yes, have to apply the acceleration also to the relative_x and relative_y axes [13:16] and as soon as you switch to input-evdev.so as the input platform you get accelerated movement [13:17] tsdgeos, believe in the force ;) [13:17] dandrader: the acceleration inside the android stack always was kind of broken for mice .. and i believe it was missing for touchpads [13:17] tsdgeos, there's nothing to trigger the binding to get reevaluated [13:18] tsdgeos, would property int foo: { if (true) return 1; else return 2 } make you as uneasy? [13:19] Saviq: no, but that's not creating new items [13:19] tsdgeos, if it would get evaluated multiple times, that's a Qt bug, wouldn't you say? [13:19] tsdgeos, but the binding doesn't depend on anything [13:19] i know [13:19] i mean you can try it right now with --platform-input-lib /usr/lib/arm-linux-gnueabihf/mir/server-platforms/input-evdev.so as runtime parameter for usc % typos [13:19] tsdgeos, ;) [13:19] but beware.. on mx4 you get jittering touch coordinats in return [13:20] we'd still get blamed for the memory leak and notifications stopping playing [13:20] lol :) [13:20] but you win [13:20] i'll quote you when if it happens [13:21] tsdgeos, let's try, please, if you can see it run more than once, we'll go for your approach (unless you can get onCompleted work, which would IMO be preferable in that case) [13:21] pushed [13:22] tsdgeos, and sure, I take the blame [13:53] tsdgeos, I've a feeling you need to alias state from the real Audio object [13:53] ah playbackState [13:53] as you were [13:53] * tsdgeos is [14:13] tedg: hey man, any word about getting the HUD back into Unity 8? === popey_ is now known as popey [14:15] mhall119: There is discussion and apparently JohnLea has some ideas that are moving forward with it. [14:16] mhall119: I don't think it's even to the point of being assigned to the U8 folks though, Saviq could comment on that. [14:18] tedg: awesome, I saw jdorleans in here earlier talking about Mycroft integration, it would be cool if it could power the HUD's voice controls [14:21] mhall119: Yeah, it would be great. Not sure that the people who are going to get assigned the work (probably pete-woods and tsdgeos) are as excited, but I want it back :-) [14:23] I do too, there are many things I would like to do with it in uReadIt [14:32] mhall119, I've not been told anything about the HUD for a while now [14:42] :/ === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [16:08] Saviq, so silo 022 is landing today? [16:13] dandrader, yeah, it's QA-approved, there's a bit of a delay because proposed migration for xenial isn't working yet, should be Monday latest [16:14] Saviq, awesome [16:29] Saviq, dandrader: hey! So I'm looking at silo 22 now... did the addition of dmz-cursor-theme as a dependency get a +1 from the product team? [16:29] Just asking as it's 3.5M more on the image [16:30] sil2100, hmm we didn't ask [16:30] I remember there was a discussion regarding that in the past, since there was a though of adding it to the seeds [16:30] kgunn, do you know ↑? [16:30] sil2100, I don't think so. I added it as a untiy8 dependency as an interim solution and ended up forgetting the matter [16:30] But the size was a concern [16:31] I mean, we freed up some space so it should be good [16:31] But still, I would prefer hearing from kgunn and pmcgowan before ACKing [16:31] (we anyway need an archive admin ACKing the landing) [16:33] sil2100: i don't recall, not sure i was part of that discussion [16:33] sil2100: in general i'm +1 as i'm anxious for silo22 to land [16:33] pmcgowan: hey! You remember the discussion we had about dmz-cursor-theme? It's 3.5M more on the image, I remember there was an idea to split it to two smaller packages [16:33] But I suppose it's fine anyways [16:34] dandrader: question on the dmz-cursor-theme....could we make it a PD only thing ? [16:34] unsure how that would work [16:34] kgunn, I don't know what it means to e a "PD only thing".... [16:35] dandrader: well, we have seperate channels & images actually for pd [16:35] so in theory, we could include that package only for pd....if space is really an issue [16:35] kgunn, and remove the dependency from unity8 [16:36] dandrader: well...that's the "not sure how'd it work" part :) [16:36] kgunn, we need it for non-PD, too [16:36] kgunn, what if you connect a mouse to your phone [16:37] Saviq: well sure...it's a choice [16:37] curious if it's really an issue [16:37] the size [16:37] if it is, then maybe we have to figure something [16:37] Saviq, we have a baked-in fallback pointer (written in svg) in qtmir. the drawback is that there's only the pointer cursor, the all the window border versions etc are lacking [16:37] kgunn, we could split the package (black'n'white theme) [16:38] dandrader, right, so maybe we shouldn't depend on it but seed it instead [16:38] and split it to keep small [16:38] sil2100: is this as easy as spin an image and see if it's an issue ? [16:38] kgunn, I don't think sil2100 means it's an issue today [16:38] ah ok [16:38] kgunn, rather he's being a weightwatcher [16:38] so we land as is, and then think about it [16:38] +1 [16:39] Saviq, adding it as a hard unity8 depencency sure isn't a final solution [16:39] good to watch your weight [16:39] It's not an issue per se as Saviq mentioned as we freed up space earlier with removal of the apt source lists [16:39] but we should be vigilant, that's for sure [16:39] But I'm always on watch for size issues ;) [16:39] Yeah [16:39] Anyway, for now I suppose we can publish this once we get the binNEW +1 from the archive team [16:40] But let's keep this on our TODO lists to maybe include only what's needed [16:40] sil2100, my vote is land (we don't have a real choice anyway) [16:40] and yeah, think how to reduce (it does seem a lot for a few cursors) [16:41] Saviq, besides the two versions (black and white) it also provides 3 or 4 different sizes for each cursor [16:41] I think the split idea fits best now [16:41] But we can do that separately [16:42] yup [16:44] sil2100, Saviq +1 to split it out when we can, 3.5MB for a cursor? [16:44] Yeah, absurd [16:45] indeed [16:45] pmcgowan, well, a *theme* [16:45] We need to include it as a TODO before OTA-8 [16:45] but still [16:45] I'll try to split it next week [16:46] just the White is still 1.9MB [16:46] ok [16:47] actually 1.7MB, some of them are links [16:48] aah [16:48] 480K /usr/share/icons/DMZ-White/cursors/left_ptr_watch [16:48] 480K /usr/share/icons/DMZ-White/cursors/watch [16:48] so *two* of them are actually .5MB each [16:48] the rest are just 16K [16:48] add a card to backlog [16:48] https://trello.com/c/iG8SmR2Q/219-split-dmz-cursor-up [16:49] yeah, it's animated... as in there's 31 frames in 3 sizes [16:54] Saviq, watch is animated and multiresolution [16:54] dandrader, yup, just saw [16:54] Saviq, so there maybe a hundred cursor images there in total (frames x the different resolutions) [16:54] yeah, it's animated... as in there's 31 frames in 3 sizes [16:55] Saviq, sorry for being mr obvious then :) === alan_g is now known as alan_g|EOW === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === boiko_ is now known as boiko === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader