=== marcusto_ is now known as marcustomlinson [07:35] Saviq, hey [07:35] Saviq, have the Authenticator issue now too. I wasn't up to date [07:56] mzanetti, acl [07:56] k [08:31] ltinkl, can you please merge lp:~unity-team/unity8/mousePointer in lp:~lukas-kde/unity8/liveCaption [08:31] ltinkl, I already resubmitted as https://code.launchpad.net/~lukas-kde/unity8/liveCaption/+merge/273792 [09:58] "any changes that you land into the overlay during the landing gates open [09:58] will have no image to appear on [09:58] " [09:58] why? [09:59] what's the purpose? [10:07] tsdgeos, stop-gap until w+1 opens [10:07] i see [10:07] when's that expected to happen? [10:07] still a few weekd? [10:07] -d+s [10:07] tsdgeos, ~a month I'd say, where did you read ↑↑? [10:08] Saviq: sil2100 sent it to the list [10:08] phone list? can't see it there [10:08] tsdgeos: for OTA-7 [10:08] ah [10:08] tsdgeos: since we're using the rc-proposed channel for the OTA-7 candidate [10:09] sil2100: but that is against what we were doing and it's because of the wily freeze, right? [10:09] tsdgeos: no, wily freeze has nothing to do with rc-proposed being frozen for now ;) [10:10] tsdgeos: we just want to use rc-proposed for spinning OTA-7 candidates, we'll unblock it for daily builds after release [10:10] Or after we validate a candidate at least [11:12] ltinkl: ping [11:30] did our autotests regress? === alan_g is now known as alan_g|lunch [11:34] * guest42315 EWWWW mailing lists [11:37] d'oh [11:37] tsdgeos, autopilot? [11:37] AP is unstable all the time [11:37] qml should be good === alan_g|lunch is now known as alan_g [12:48] Saviq: and one qml too [12:49] let me run them [12:49] here locally === alan_g is now known as alan_g|lunch [13:11] Saviq: and one autopilot seemed to be effected by one of ltinkl's changes too [13:12] tsdgeos, is it the dbus mock thing? [13:12] ltinkl: seemed like something in the wizard [13:12] let me dig [13:12] tsdgeos, if someone's getting reliable qml test failures, that's a problem, same if the autopilot failures are consistent [13:12] yeah i need to wait for the rerun [13:12] we should get green on QML and at worst random AP fails [13:13] agreed [13:14] ltinkl: https://jenkins.qa.ubuntu.com/job/generic-deb-autopilot-runner-wily-mako/483/testReport/unity8.settings_wizard.tests.test_settings_wizard/SkipThroughSettingsWizardTestCase/test_skipping_through_wizard/ [13:14] complains about Object not found with name '*' and properties {'objectName': 'reportingPage', 'visible': 'True'}. [13:14] which is something you touched recently? [13:16] tsdgeos, umm ye but that's something different [13:16] ltinkl: ok [13:18] tsdgeos, ok later, need to pick up kids from school :) === alan_g|lunch is now known as alan_g [14:08] mterry: not what the SDK people told me about 1.3 [14:09] mzanetti: Saviq: right? didn't they say "1.3 is far away" ? [14:10] * mterry tries to find where he saw that [14:10] the release, yes, zsombi said like early 2016... [14:10] however, he wasn't aware of the the issue that creates [14:11] yeah 1.3 isn't coming soon [14:11] OK cool [14:12] ltinkl, I remembered that I have Friday AND Monday off, so I actually won't be able to attend the next OOBE standup [14:12] mterry, ok [14:14] Saviq, tsdgeos: I feel like we should stop shipping it on the phone then... App authors will use it despite it not "being released" and we'll break them [14:14] Or at least ship it with a version number like "import Ubuntu.Components 1.3dontusethis" [14:15] mterry: i do agree [14:16] I'll file a bug against the sdk and see what they say about an alternate versioning in the meantime [14:16] ack [14:17] dednick: ping [14:17] tsdgeos: alo [14:17] dednick: what do you think of http://paste.ubuntu.com/12715158/ ? [14:17] real leak or test-leak? [14:18] tsdgeos: give me a mo [14:19] https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1504161 [14:19] Ubuntu bug 1504161 in ubuntu-ui-toolkit (Ubuntu) "Stop shipping an SDK versioned "1.3" on customer phones before we release 1.3" [Undecided,New] [14:22] confirmed [14:26] I don't think we do change compat for 1.3 things, do we? [14:26] some of core apps use 1.3, it better be stable [14:27] seb128: it's totally not [14:28] our non official 1.3 branch has broken quite a bit because of changes [14:28] argably "dark corner" stuff [14:28] but still [14:28] well, those might be bugs [14:28] the sdk team is pushing for others to use 1.3 [14:29] and some core apps do [14:29] so we better sort that story out [14:29] we speak to different SDK people :D [14:29] also there was the issue of having a newer framework to depends on [14:29] tsdgeos: i'm thinking it's a test issue. but have to make sure. [14:29] some at least appdevs can declare that they use the current version and those apps don't get rolled out on old images not having it [14:30] pmcgowan, ^ do you know what's the story with 1.3 is? should be stable enough that appwriters use it? or should we recommend against using it? [14:32] seb128, part of that bug I filed is to get an official answer from sdk team [14:32] seb128, current situation is confusing to me [14:33] +1 [14:37] seb128, in my view and bzoltans devs can use 1.3, the api may expand but it exisitng api should not change, and we have in ota7 a new framework to reference [14:37] mterry, why did you file that bug? [14:38] that is bad data [14:38] its released [14:38] pmcgowan, because I'm hearing different things [14:38] I can believe that [14:38] pmcgowan, and 1.3 was released in OTA6, but it wasn't officially released [14:38] so there is a term "stable" which means different things to different people [14:38] pmcgowan, so that was at least poor planning. At least OTA7 will have a framework for it [14:38] mterry, true it was not official in 6 [14:39] correct [14:39] pmcgowan, so if this bug is considered fixed by OTA7, great. But it is a valid bug [14:39] ok [14:39] you win :) [14:39] pmcgowan, is the SDK team committed to API stability for 1.3 already? [14:40] (that's another thing I wasn't sure about) [14:40] mterry, acc to bz and me yes [14:41] as I said it may add components and new properties but is not intended to break [14:41] we need to use it now for convergence [14:42] pmcgowan, OK cool. There is uncertainty on my end partly because 1.3 appeared on devices, but there was never an announcement / blog post about it. So no clear message about using it [14:42] Or at least, no clear message about API stabliity [14:42] mterry, yes I agree, will have zoltan clarify [14:43] the website got updated last week [14:46] tsdgeos: hi [14:47] bzoltan_: ho [14:49] bzoltan_: anything you wanted to tell me? I need to get going to the airport soon-ish [14:52] tsdgeos: I would like to clarify few thinsg about the UITK 1.3 [14:52] 1. There was no API breakage in the UITK since 0.1 ... not a single [14:52] 2. When we deprecate APIs we do it gentle, slow and well communicated ... never ad-hoc [14:53] 3. whatever is released in 1.3 can be safely used [14:54] bzoltan_: 1. you guys removed useDeprecatedToolbar in MainView in 1.2 for example, that's an API break [14:54] unstable in our terminilogy means that we keep adding components and change visuals as the design evolves, but it does not mean API breakage [14:54] tsdgeos: errr... dude, have you read the _NAME_ of the property? :D it says ... deprecated :D [14:56] bzoltan_: i know [14:56] don't tell me API was never broken [14:56] because it has [14:56] it may have been slowly done [14:56] fine [14:56] and within 1.3 things broke [14:56] tsdgeos: API deprecation is not breakage [14:56] i just don't have the link at hand at the moment to prove it [14:57] tsdgeos: bugs are not API breakage [14:57] but specially around PageHeadConfiguration [14:57] tsdgeos: we have bugs, yes [14:57] anyhw that's not something i complain [14:57] i have not complained about that :D [14:58] tsdgeos: we are refactoring lots of things for the desktop apps and for better performance .. bug happen [14:58] i'm complaining about mixed messages about if 1.3 should we used or not [14:58] which seems to be "yes" from some people [14:58] and "no" from some people [14:58] tsdgeos: I am just a but dissapointed about the gossips and terminology confusion [14:58] bzoltan_: i have no idea how tells you i'm gossiping [14:58] but i am not [14:59] so do not lecture me plese [14:59] tsdgeos: I do not know who are those who say "no" but I assume they do not work on the SDK :) [14:59] bzoltan_: zsombi [14:59] and Saviq can confirm [14:59] tsdgeos: no offense... I do not mean you [15:00] i have a plane to take, we can talk about this tomorrow if you want [15:00] tsdgeos: zsombi never said that 1.3 should not be used. [15:00] tsdgeos: sure, anytime [15:00] bzoltan_: well you were not in the place i was [15:00] and did not hear what i hear [15:00] i may have misunderstood [15:00] but it seemed pretty clear to me [15:01] anyhow, tomorrow more [15:01] * tsdgeos waves [15:01] bzoltan_, we've been asking about when can we switch to 1.3, and use features from it a few weeks back, we've been told it's not to be used yet (as in, it's not gonna be there in any framework in the near time) [15:01] Saviq: who told you that? [15:02] Saviq: The 1.3 is on all phones since May .. it is available for app developers since then. [15:03] bzoltan_, no it's not [15:03] bzoltan_, because there's no framework to depend on [15:03] bzoltan_, so app developers can't use it without their apps breaking on devices that didn't upgrade [15:03] Saviq: 1.3.1510+15.10.20150519-0ubuntu1 [15:04] bzoltan_, and which framework do you use to ensure that SDK 1.3 is on the device your app is being installed on? [15:05] Saviq: well.. the next fw after 19th of May [15:05] bzoltan_, which is? [15:05] Saviq: I do not remember that [15:05] bzoltan_, the only frameworks post May 19th are -dev ones [15:05] Saviq: there were few fw bumps since [15:06] wdym few fw bumps? [15:06] ubuntu-sdk-15.10-dev1 [15:06] ubuntu-sdk-15.10-html-dev1 [15:06] ubuntu-sdk-15.10-papi-dev1 [15:06] ubuntu-sdk-15.10-qml-dev1 [15:06] Saviq: of course ...-dev [15:06] which means they're not stable as far as I understand [15:07] Saviq: I do understand you and you are right. from that point UITK 1.3 ios not stable .. it will be once we freeze it [15:08] Saviq: But it does not man that it is unsafe to use [15:08] s/man/mean/ [15:08] well, there's miscommunication then [15:08] is it safe to mix 1.3 and 1.2? [15:09] more like different use of terminology [15:09] as we can't go for 1.3 until it's stable otherwise because the header changes and we can't look different than core apps etc. [15:10] Saviq, you already do, bug #1503498 [15:10] bug 1503498 in unity8 (Ubuntu) "Using both X-Ubuntu-Splash-Show-Header=true and SDK 1.3 looks bad" [Undecided,New] https://launchpad.net/bugs/1503498 [15:10] seb128, right, our bad [15:11] but also stemming from the fact we thought it was safe to switch [15:12] Saviq: mixing versions is tricky ... it should work and I know apps that mix versions without problems. But version separation is a new thing in the UITK, so it can bug [15:12] Saviq, well, it's the other way around in that bug, some apps updated to 1.3 and have the new look, and you didn't ... but unsure how that can be resolved, some apps are on 1.2 and some on 1.3, unsure what the shell can do to know what to display [15:12] Saviq: Why can not we go with 1.3? === dandrader is now known as dandrader|afk [15:13] seb128, right, I think we might be able to get the framework from an app and display one or the other [15:13] Saviq: I would go with 1.3 [15:13] bzoltan_, maybe now we can if you tell us it's fine, but we have to update [15:14] all the apps we have preinstalled on the phone [15:14] let's make that a target for OTA8 then [15:14] that would be good [15:14] Saviq: I have an idea how it should go... I am pushing it for 3 years without much luck :) Do you want to hear? [15:14] we have apps still importing 0.1 [15:15] seb128, right, framework doesn't force you to import 1.3... [15:15] Saviq: We should have a QA dashboard with the staging UITK and whatever unity8/browser/setting/etc ... the dash should show the AP results [15:17] Saviq: seb128: the point is that apps should adopt to the latest available UITK and test their apps against the staging UITK ... that way they could foresee any posble regression and flag out [15:19] sure, we should have autotests running on silos, too [15:19] we'll have Jenkaas instead, but whatever [15:20] Saviq: The thing is that for those app developers who do not want to follow what we are doing the 1.0 and 1.2 are the safe land ... but for developers who are hanging on our IRC and read our blogpost and follow our changes 1.3 is the cheapest way [15:20] seb128: ^ [15:20] right [15:21] it's just that porting code to new version is work [15:21] sure, agreed [15:21] so unless we get that set as an ota goal it's always coming after other work [15:21] Unstable in our terminolgy does not mean that it will crash and eat your dog :D It means that we keep adding new APIs and with a strict process we migh deprecate stuff [15:21] should talk to pmcgowan maybe about having a goal of updating our stuff to 1.3 for ota8 [15:21] good point [15:21] * pmcgowan makes a note [15:22] pmcgowan, thanks ;-) [15:22] seb128: Saviq: pmcgowan: The deal is that once you are on 1.3 you must watch out for possible but very unlikely changes! [15:23] bzoltan_, sure as long as they don't get in the wild on stable [15:23] But API deprecations are super rare things ... [15:23] well, we should at least update to 1.2 [15:23] bzoltan_, as long as you post a deprecation warning, it's fine [15:23] seb128: That is for sure... we should have a check for that in the SDK tools [15:24] Saviq: Of course we do ... first we spam the logs, build logs, runtime logs ... then we send out mails, write blog posts... [15:25] bzoltan_, an email to the app-devel list along these lines would be helpful [15:27] pmcgowan: I write one latest tomorrow === dandrader|afk is now known as dandrader === alan_g is now known as alan_g|EOD [17:56] Saviq: dandrader: how much of the convergence stuff I've been testing out of silo 22 is going to be in OTA 7? [17:58] mhall119, I don't know. And Saviq is past his EOD [17:58] ok, I'll try him tomorrow, thanks dandrader [17:59] dandrader: do you have a Mir bug # for that mouse issue we both encountered? [18:28] mhall119, none of it [18:28] Saviq: thanks, I expect I'll be asked about it once the OTA goes out [18:29] mhall119, it should land within a week in rc-proposed [18:29] maybe for the better, we'll have time to iron some things out [18:30] Saviq: since neither commercial device supports it yet anyway, I don't think there's a rush, I just wanted to know what to tell people if they ask [18:31] kk === dandrader is now known as dandrader|afk === \b is now known as benonsoftware === dandrader|afk is now known as dandrader