08:00:35 #startmeeting 08:00:35 Meeting started Thu Jun 14 08:00:35 2018 UTC. The chair is Jaymzz. Information about MeetBot at http://wiki.merproject.org/wiki/Meetings. 08:00:35 Useful Commands: #action #agreed #help #info #idea #link #topic. 08:00:38 hi 08:00:45 #info Sailfish OS, open source, collaboration – June 14th 2018 08:00:56 #info meeting information and agenda can be found here: http://lists.sailfishos.org/pipermail/devel/2018-June/008412.html 08:01:05 I am the meeting’s chairperson today and will be doing my best to keep time and order. Please behave, respect the timings and be gentle. 08:01:17 #topic Brief introduction (5 min) 08:01:50 #info Please prefix your name/handle with # info 08:01:58 #info Chris Adams, developer at Jolla 08:02:01 #info James Noori - Sailor @ Jolla 08:02:09 #info Leszek Lesner, Community & Dev 08:02:10 #info Lewis Rockliffe, community developer 08:02:15 #info Matt Austin, Community/developer 08:02:24 #info Martin Kolman, community & modRana development 08:02:28 #info jakibaki, community/developer 08:02:29 #info David Llewellyn-Jones, community developer 08:02:36 #info Louis, Sailfish fan 08:02:43 #info Simonas Leleiva @ Jolla 08:03:27 #info Raine Mäkeläinen, developer @ Jolla 08:03:28 #info Erik Lundin, community 08:04:23 #info Frajo Haider, developer @ Jolla 08:05:39 #info Kaj-Michael Lang, Community dev 08:06:13 #topic Update camera driver on Xperia X to fix white balance issues. (asked by louisdk – 5 min) 08:06:17 #info Current Sailfish X build is based on an old 8 mpx driver which do have serious white balance issues when using the back camera. As current Open device program adaption using kernel 3.10 utilizes a newer driver featuring 20 mpx support and have better white-balance I'll suggest a driver update to mitigate this issue. 08:07:45 sledges ping 08:07:53 #info David Greaves, Sailor and Mer guy 08:07:58 Any Jolla devs have something to say about this? And if this is not to be done by Jolla can I community driver port be accepted into Sailfish X? 08:08:19 Isn't the driver closed source anyway and a android blob? 08:08:19 louisdk: in which Android base is the camera fix in? 08:08:20 (Sorry I'm late - my sailfish calendar told me it was cancelled :D :D ) 08:09:06 #info Jonatan Walck, community (and late) 08:09:08 lbt that's odd :D 08:09:11 sledges: both in latest 3.10 build and 4.4 build for Xperia X: https://developer.sony.com/develop/open-devices/get-started/supported-devices-and-functionality/current-platform-functionality-maintained/ 08:09:29 louisdk: which android build though? 6? 7? 8? 08:09:31 louisdk: yes, but which Android base? 08:09:38 #link https://developer.sony.com/develop/open-devices/get-started/supported-devices-and-functionality/current-platform-functionality-maintained/ 08:10:40 3.10 kernel drivers for Android 6.0/Android 7.1 and 4.4 kernel drivers for Android 7.1/Android 8.0/Android 8.1 08:11:06 louisdk: you're misunderstanding the question, the blobs arent in the kernel 08:11:15 louisdk: we're on 3.10 android 6, so the hw adaptation would have to be updated to android 7 to support the 4.4 kernel 08:11:23 we already use the driver of 20 mpx (limited to 18mpx in sailfish x) 08:11:52 abranson: he says "using kernel 3.10 utilizes a newer driver featuring 20 mpx support and have better white-balance" 08:12:15 so if it's in Android 6, then it's a simple update. otherwise it's the opposite effort and tech-debt-wise 08:12:28 sledges: is there any plans to rebase on android 8 for f5121? 08:13:02 oh no, that's not right about 8mpx driver etc. plus the android 7 blobs might have different bugs. 08:13:05 r0kk3rz: let's discuss this one when we have general discussion please 08:13:09 we're focusing on XA2 first, will be planning whether to rebase or not thereafter 08:13:13 In terms of better white balance. This can be worked around in the camera app. Making it automatically switch to tungsten WB when activating flash for example 08:13:16 the time is limited for this topic 08:13:30 sledges: in that regard the issue might already be fixed since last time I checked. Sorry for the misunderstanding. 08:13:38 Jaymzz: that question was on topic, new base == new camera blobs :P 08:13:48 r0kk3rz: ah, my bad sorry 08:13:49 sledges: when did you start using the 20mpx driver? 08:13:59 louisdk: around New Year 08:14:02 ok, just to say again - all the f5121 hw adaptation is on github, so if anyone wants to try our droid 7 and 4.4 kernel, it's all there :) 08:14:34 and we're in #sailfishos-porters to help :) 08:14:49 of course upgrading the blobs would mean reflashing that oem 08:15:00 #info community is encouraged to take a look and develop on the F5121 adaptation which is available on Github if they want to try out the Android 7 and 4.4 kernel 08:15:05 i did actually try the droid 7 blobs with our android 6 base for a while - they're not that different 08:15:24 louisdk: Let's give these questions 10 minutes or more next time mate ;) 08:15:38 unless you want to use the front camera - everything goes a bit psychedelic with that :D 08:15:55 #info it needs to be doublechecked if currently as we already support 20mpx driver since a while on Android 6 "In terms of better white balance. This can be worked around in the camera app. Making it automatically switch to tungsten WB when activating flash" 08:16:19 sledges: I'm a bit unsure if the white balance issue is fixed in the current driver. Anyway I've got my question answer for now. Already using lastest camera driver possible (20 mpx). 08:16:41 alright moving on - we are way over time on this one 08:16:52 * sledges :thumbsup: 08:17:03 #topic Update alien dalvik on Jolla Tablet (asked by louisdk - 10 min) 08:17:12 #info Current alien dalvik implementation is based on Android 4.4 and should be able to run on Jolla Tablet. Are there any obstacles to this apart from code recompilation? 08:17:24 louisdk: I changed this to 10 minutes just in case 08:17:47 Jaymzz: Thanks and sorry. 08:17:54 louisdk: no problem :) 08:18:51 ping krnlyng 08:19:05 We will try updating aliendalvik on the Jolla Tablet to the same version as Jolla C 08:19:33 krnlyng: thanks. All I wanted to hear. 08:20:07 great! :) compensated for the last question's time I guess ;) shall we move on louisdk? :) 08:20:08 is it not already 4.4.4? 08:20:53 r0kk3rz, it is, but the Jolla C version contains some small bugfixes that are not included in the Jolla Tablet version at the moment. 08:21:04 ah i see 08:21:11 Jaymzz: Yes. Please do :) 08:21:25 #topic Some remarks on the gallery interface and design direction of SFOS (asked by ApBBB – 10min) 08:21:29 just to clarify, both Jolla C and Tablet are getting an updated version of Alien Dalvik for Sailfish 3, right? 08:21:40 #info Newer interfaces (gallery camera) seems to move away from the usual SFOS design paradigm. 08:22:23 #info The new Gallery design is implemented after we received numerous requests over the past years. It is an overall improvement considering that it is easier to use than the previous one especially if you are new to Sailfish OS UX. It definitely follows our principles with gesture control and it works rather intuitively at that. If you would like to see more improvements and changes, let us know what you think we can improve up 08:22:23 Further, split view has been seen as problematic to use and causing troubles for people. Hence, it got replaced from Gallery image details and Camera roll. There are views in Gallery that are still using split view such as image edit and those will be updated in later phase. Also other apps using split view will get similar treatment. 08:22:37 ApBBB: Does that answer your question here? :) 08:23:04 overall the change made the gallery app flow better. however you added -as in the camera interface- taps on the top section which are hard to reach 08:24:06 nh1402[m], i think you should file that as a separate question for an upcoming meeting. 08:24:15 ApBBB: are you talking about the X on top? isn't that solvalble with a gesture? or am I missing something? 08:24:35 for the record, I want to say I like the redesign & the previous paradigm always seemed rather clunky and rather confusing 08:24:37 imo the new gallery screen is nice, much more discoverable than the previous one 08:24:38 Yeah the gesture swipe down or up closes the window anyhow 08:24:43 ApBBB, we tested with different users which one should go with top and which to bottom => regarding close on camera roll, we're working on it to get it to the same level as in gallery 08:24:46 i am talking about the info (top left). Also the ambience button on the bottom 08:24:52 yea, gesture is been planned. 08:25:04 Gesture swipe up/down doesn't close in the camera picture viewer. 08:25:11 the up swipe is good 08:25:17 what i didnt like, was there was no confirmation on the ambiance button 08:25:30 some of the buttons coulb be implemented in a pulley 08:25:37 and not use the down swipe 08:26:01 ambience and info buttons should go in a pulley IMO 08:26:05 r0kk3rz, noted 08:26:17 As in speed (I read the TJC thread about it being slow). Is it hiding the overview in the background after the image is shown on top? I know in Webcat I need to do this for the video player to gain more speed by hiding the underlying webview. Not sure if that is a webcat/qtwebkit only thing or something that is a general issue at least on the Xperia X 08:26:22 what r0kk3rz said is also an annoyance 08:26:51 Should other community apps move away from SplitView? 08:26:57 ApBBB: r0kk3rz thanks for your input, as said above, there is always room for improvement. Your input is noted and will be taken into consideration! 08:27:26 flypig: If they want to follow Sailfish design language, I would say yes. 08:27:30 Jaymzz: the thing is that even in the same app UI looks all over the placve 08:27:41 Jaymzz, thanks, noted. 08:28:00 ApBBB: how so? 08:28:18 ie edit looks different 08:29:02 i've been talking about inconsistencies allover SFOS for quite a while now 08:29:10 I guess in that regards rainemak should be the guy to talk to, as it is going deeper and he is the expert :) 08:29:25 leszek: can use QSG_VISUALIZE=overdraw to check if alpha blending is occurring 08:29:49 Jaymzz: basically the designer is who i am supposed to speak to i think 08:30:19 and even if i present rainemak with all my gripes on sfos i doubt much will change :/ 08:30:27 ApBBB: He is unfortunately not available at this meeting :( (side note: 1 minutes left from the time dedicated to this topic= 08:30:31 ) 08:31:07 ApBBB, a lot will happen still during this year 08:31:14 hello 08:31:27 rainemak: in which direction i don't know. 08:31:35 yeah designers are opinionated, thats kinda their job :P 08:31:56 everything that was updated was in a way unsailfishified 08:32:08 chriadam_: testing. ssh is slow as hell on the xperia x. Needs 10 retrys to connect finally 08:32:15 ie the iso settings on the cammera 08:32:22 well, the design is supposed to advance over time 08:32:36 ApBBB: Time's up for this now, do you want to add another 5 minutes or continue later at general discussion or even continue during the next meeting? 08:32:52 2mins 08:32:54 fwiw feedback is good. if you can email me a list of your gripes / wishlists I can forward it to the designer. I can't promise that he'll agree with your points, obviously. 08:33:24 abranson: advance with what user in mind? the one that is used to android apps or the sfos design paradigm 08:33:58 hopefully both :D 08:34:09 #info adding 5 more minutes to this topic 08:34:10 abranson: that wont work 08:34:31 the barrier to entry should be as low as possible though 08:34:52 otherwise it becomes too inaccessible to new users 08:35:05 chriadam_: not sure how I can see if alpha blending appears. It shows a small 3D animation with red ontop of webcat. Jolla-Gallery isn't working for me it is all black 08:35:09 leszek: try USB2 port for ssh 08:35:23 abranson: if it looks cool and works users will come 08:35:25 luckily android and iOS are slowly becoming more swipey and gesturey 08:35:28 yeah that usb cable works faster 08:35:53 Jaymzz: i think we can move on. i'll probably make a list and mail it to chris at some point 08:36:00 leszek: it's documented at http://doc.qt.io/qt-5/qtquick-visualcanvas-scenegraph-renderer.html 08:36:02 it will take some time though 08:36:03 ApBBB, regarding ambience selection you'll get a fix in next release 08:36:09 thx 08:36:12 ApBBB: sounds good. 08:36:19 in any case I think we can agree the split view has not proved itself good enough for mainstream acceptance, even just for discoverability reasons 08:36:19 #info regarding ambience selection you'll get a fix in next release 08:36:36 M4rtinK: very true 08:36:51 moving on 08:37:01 #topic freedesktop standards vs lipstick (asked by tortoisedoc – 15 min) 08:37:11 #info currently freedesktop standards is partially adhered to by lipstick. Mainly user-defined conifiguration is not prioritized over system defaults, as requested by the spec in an obscure way here ; in particular "The base directory defined by $XDG_DATA_HOME is considered more important than any of the base directories defined by $XDG_DATA_DIRS. ") is currently not adhered to by lipstick, which searches both locations (system & 08:37:11 er). The fix as such is "straightforward", but carries serious security implications with it. The idea is to brainstorm workarounds to harden the security of this "feature" and come up with a solution. 08:37:52 I don't see tortoisedoc available though... 08:38:10 and he has noted " if i am not around move it to the next" 08:38:21 so I will move on in a few seconds. 08:39:07 #info topic dismissed due to the author's request and unavailability. Will retain the topic on TJC for the next meeting and if the author is unavailable by then topic will be removed. 08:39:07 If it goes to next meeting, any chance of a summary of security implications beforehand? 08:40:10 rainemak: ^^ 08:40:17 I guess it's that the handler application for some mimetype could be overwritten if you can inject a config file into one of the searched paths. but not 100% sure. 08:40:59 maybe I'm misunderstanding which part of freedesktop standards he's referring to here. 08:41:32 chriadam_ needs clarification from tortoisedoc 08:41:34 chriadam_: Yeah that was my concern too that this question might need extra explaining 08:41:43 +1 08:41:46 alright moving it to next meeting then 08:41:51 moving on to the next topic 08:42:03 #topic Has building on top of another distribution been considered as alternative to Mer? (asked by luen – 10 min) 08:42:13 #info Jolla developers spend a lot of time maintaining middleware packages in the Mer core distribution, but still many of them are very old (the packages, not the developers...). Has it ever been considered to instead piggyback on some more general distribution like Debian or Fedora, which each has thousands of developers who continuously work to keep packages up to date? Would it be feasible to create a niched distribution like 08:42:13 ilfish OS on top of something that general? 08:42:51 There have been proposals in the past but Mer is essentially a collection of standard upstream components. Over time the boundary between Mer and SailfishOS has blurred too - there's no practical difference any more. 08:43:07 The packaging is greatly simplified to avoid the complexity of supporting alternatives (eg NetworkManager and Connman) and building for old/multiple versions and to work with things like SSU and image creation which are different to most distros. 08:43:40 I think the main issue would be also another distribution needed to be snapshots otherwise it moves too fast for SailfishOS :P (Like Qt update could break the UI completely) 08:44:01 lbt you can start your sentences with #info to avoide repetition from my side ;) 08:44:02 *snapshotted 08:44:04 We actually make extensive use of Suse and Fedora too 08:44:05 There would be a significant cost of changing to another core and it's questionable what benefit that would bring; especially given our investment in tooling around OBS etc. 08:44:12 Jaymzz: oops - will do 08:44:16 cheers 08:44:29 well for most packages, theres no patches and updating them is fairly easy, as is adding new packages 08:44:36 *snapshot 08:44:40 for major things like Qt, theres many patches, and so migrating them takes time 08:45:00 leszek: I was going to say it would be interesting to think how well SFOS would haveWhilst that's valid it would be interesting to think how well SFOS would have fared it it had *had* to follow an upstream at their pace - anyway fared it it had *had* to follow an upstream at their pace - anyway 08:45:14 except without the bad cut'n'paste 08:45:22 Yeah, no doubt there would be a cost involved in changing, but I just wonder if it has been considered long-term. 08:45:22 Whilst that's valid it would be interesting to think how well SFOS would have fared it it had *had* to follow an upstream at their pace - anyway 08:45:40 #info Mer core is open to community contributions and has a clear Maintainer setup. We do have architectural (and legal GPL3) issues wrt to upgrades which need to be factored in though 08:45:57 #info Feel free to talk to a maintainer about upgrading a package and get some guidance on how to go about it (eg grab a copy of the latest Fedora src and tweak our packaging) 08:46:13 #info In general we need to be lead by SFOS architectural needs 08:46:18 my experience with updating packages has been pretty horrendous in the past 08:46:20 #info See https://git.merproject.org/mer-core/Maintainers/blob/master/maintainers.yaml 08:46:29 if you want help submitting a new package to mer, you can talk to me about it 08:46:32 M4rtinK: you are right 08:46:48 M4rtinK: ive recently submitted a number of changes, its been fine :P 08:46:48 just to get new PyOtherSide update in took a couple months and nagging on community meetings 08:47:02 M4rtinK: it's time we got that kind of stuff improved 08:47:05 I really hope it improved since then 08:47:14 a couple of updates, and like 4 new packages 08:47:23 M4rtinK: So I think we're going to touch on that in the next topic :) 08:47:27 also still no way for non-Jolla people to maintain packages 08:47:37 *cough* yet 08:47:45 M4rtinK: So I think we're going to touch on that in the next topic :) 08:47:45 :D 08:47:46 Considering that quite a lot of work seems to be spent on maintaining mer core packages, and still many of them are very old 08:48:06 luen: specifically? 08:48:21 luen: the physical updating of the packages isn't the tricky part - it's the testing against the rest of sailfish. that would be the same if we switch to something else. 08:48:34 across the entire device portfolio 08:48:45 and multiple vendor patterns 08:48:51 sailfish has a lot of custom middleware and a unique front end. that stuff usually needs tweaking when we update base packages 08:49:04 I work on Fedora & can see all the work going into package maintenance, so seeing the same work being done more or less again for slightly different package versions on the Mer side seems kinda bizzare, especially given the already pretty overworked Mer/Sailfish devel team 08:49:15 chriadam_: i think that you are too small for a lot of devices :/ 08:49:30 not you as in chris as in jolla 08:49:47 #info abranson has volunteered(!) to be the main contact to discuss upgrades with if you can't get a response from a Maintainer 08:50:09 way to go abranson! 08:50:14 D: 08:50:15 It can be done I think. It just needs good automated tests and CI work. And yep this is the hard part to get going 08:50:15 r0kk3rz: Don't have any example at hand right now, but when I had a look yesterday, it was easy find examples in the mer core git 08:50:17 cool, thanks! :) 08:50:20 abranson: OK 08:50:40 luen: its best to have examples, some things can be updated, some things are tricky 08:50:51 * M4rtinK thinks about updating PyOtherSide to the latest minor bugfix version 08:50:55 in general, you can submit updates yourself if you package them 08:50:56 ok if no one wants to list it. I can. QtWebkit is horribly outdated :P 08:51:04 leszek: we do have fairly solid CI and QA internally - but not enough community contributions to persuade management to invest in rebuilding the Mer CI/QA 08:51:13 this is sadly catch 22 08:51:40 #info we do have fairly solid CI and QA internally - but not enough community contributions to persuade management to invest in rebuilding the Mer CI/QA. Catch 22 08:51:46 lbt: exactly - I can't really see any meaningful contributions being possible with the current setup 08:52:14 M4rtinK: part of that problem is that the full CI needs a build of the closed bits too :/ 08:52:35 r0kk3rz: OK, just as an example, libjpeg (just picked one at random). It's several years old. 08:52:44 well, the closed bits have been a problem from day one & will continue to be as long as they are closed... 08:52:48 luen: do you need more time? time is up for this topic. I can give an extra 5 minutes to wrap up 08:52:53 if needed? 08:52:56 luen: that one can probably be updated :P 08:53:30 Jaymzz: luen: I think we're conceptually on the next topic anyway 08:53:42 Jaymzz: If it's possible a few more minutes would be fine to finish the ongoing talk 08:54:40 So it feels like we're back at the 'how to contribute to mer:core in a meaningful way' 08:54:45 also Python 3 - we have 3.4.3, released February 25th, 2015 08:55:07 #info 5 more minutes added to this topic 08:55:16 abranson (and I'll help) are going to work on fixing the Mer OBS build of mer:core 08:55:23 while I have not hit any major issues with the 3.4 series just yet, it might be good to at least thing about updating to a more recent version in the future 08:55:48 btw you guys definitely need to ask for more time than you anticipate it is going to take! 08:55:52 there are many fixes and improvements in newer versions 08:56:05 lbt: Contributing feels more meaningful if you don't see the same work already done in other places. 08:56:14 luen: in what way? 08:56:22 Jaymzz: tbh we can move on i say, the question was 'why dont you use debian/fedora' which was been answered 08:56:37 BTW, does Python even have a maintainer in Mer/Sailfish ? It seemed kinda orphaned since THP left. 08:56:55 r0kk3rz: yeah - we're on 'contribution' now - which is why M4rtinK sounds like he's on the next topic :D 08:56:57 r0kk3rz: I noticed too, but the author asked for a few more minutes. I'll move on soon(TM) 08:57:06 lbt: I mean the packaging work already made in other dists 08:57:23 luen: so feel free to merge in packaging changes 08:57:26 luen: yeah my starting point for any package is src.fedora.org 08:57:29 luen: getting a package upgrades can use work from other dists as a start, but it's never that simple 08:57:30 we tend to use Fedora style 08:57:51 *src.fedoraproject.org 08:57:52 but they do a lot of work on making things work with all kinds of #ifdefs 08:58:01 we generally don't need that 08:58:06 lbt: good good :) (I have seen some pretty horrendous things in OpenSUSE packaging) 08:58:16 lbt: and opensuse, because of obs! 08:58:17 yeah they have a more complicated setup that we need 08:58:17 M4rtinK: haha 08:58:26 abranson: exactly 08:58:50 r0kk3rz: s/that/than/ ??? 08:58:51 just don't ever look at the texlive package in Fedora and you will be good :) 08:59:31 luen okay time is definitely up now. Moving on to the next topic 08:59:33 lbt: yeah than :) 08:59:40 crucial letter there 08:59:46 M4artinK: I believe Python 3.6 is 'long term support'? (correct me if I'm wrong) Might be a good candidate. 08:59:47 Jaymzz: Alright 09:00:00 luen: ty - good topic 09:00:16 #topic Allowing usage of newer versions of PyOtherSide in Jolla Store/RPM validator (asked by M4tinK – 5 min) 09:00:20 mattaustin: let me check 09:00:27 #info The PyOtherSide library, which makes writing of Python 3 + QtQuick 2.0 applications possible has been updated to version 1.5 in Sailfish OS quite some time ago. The new PyOtherSide version brings new features and fixes some rather annoying bugs. But apparently the RPM validation tool used for gating Harbour acceptance only allows the QML import for >=1.3 only. This looks just like an omission (the higher version of PyOth 09:00:27 ide has been in official repos for ~year) and I've prepared a PR fixing the imports: https://github.com/sailfishos/sdk-harbour-rpmvalidator/pull/109 Could the PR be merged so that developers can benefit from the improvements introduced since PyOtherSide 1.3 ? 09:00:39 #link https://github.com/sailfishos/sdk-harbour-rpmvalidator/pull/109 09:01:26 expecting that's quite ok, but didn't yet get onto it. 09:01:45 later today, i think. 09:01:49 :) 09:02:00 so basically, is this just and oversight & can we fix it by merging my PR ? :) 09:02:31 So I think one answer here is to improve the Maintainer situation 09:03:02 As I mentioned above SFOS uses https://git.merproject.org/mer-core/Maintainers/blob/master/maintainers.yaml for the packages in mer:core 09:03:10 technically, users should be already able to benefit from the bug fixes even if lower version is imported, but the newer features are AFAIK guarded by a version check 09:03:58 we have master and developer roles - the master role is essentially to coordinate with SFOS archicture and they work with developers on getting stuff in 09:04:35 So for python3 and pyotherside pvuorela has volunteered(!) to be the master 09:04:54 woo! 09:04:57 cool! 09:04:58 M4rtinK: would you care to take on an official developer maintainer role? 09:05:21 lbt: definitely for PyOtherSide 09:06:09 great - do you think it would help with python3 as we have limited resources there 09:06:44 #info pvuorela to be the master for python 3 and pyotherside. He will check everything later today. 09:07:00 We can leave it with just pvuorela as master and review PRs as they arrive 09:07:10 M4rtinK: I assume you got your answer right? 09:07:52 lbt: do you mean maintaining Python 3 itself ? I'm not sure I'm that well versed in Python internals to be the main maintainer of Python 3 on Sailfish OS 09:08:37 M4rtinK: OK - no problem. I assumed that pyotherside would have interactions there and it'd mainly be a packaging and CVE thing 09:09:30 I would be willing to be a co-maintainer or something similar, but really not the primary maintainer for Python 3 09:09:46 for the record, I'm the maintainer of PyOtherSide on Fedora 09:10:14 yes - you'd be a developer (ie could merge stuff to branches) and pvuorela would be master (ie makes sure SFOS is OK with changes) 09:10:27 M4rtinK: Time is up man, how much more time do you need? 09:10:37 M4rtinK: you could also coordinate community contributions too 09:10:50 ie ensure they went to the right devel/testing branches etc 09:11:19 lbt: I think I would be fine with being listed as in the developer role as long as there is also someone else 09:11:25 then you and pvuorela would work to getting it into the master and into a release 09:11:27 OK 09:11:45 & definitely want to maintain PyOtherSide 09:12:18 also, about the original question - can someone review https://github.com/sailfishos/sdk-harbour-rpmvalidator/pull/109 ? ;-) 09:12:19 #info M4rtinK to be an official developer for PyOtherside and Python3; working with pvuorela 09:12:29 So, is the Master the person to get changes to rpmvalidator through? That's one of the issues with the original question if I understood correctly. 09:12:54 flypig: essentially yes 09:12:54 12:01 < pvuorela> expecting that's quite ok, but didn't yet get onto it. 09:12:54 12:01 < pvuorela> later today, i think. 09:13:46 This is going way over time. You guys either need to ask for more minutes when creating your topic or respect the timing 09:13:55 I'm miving on to general discussion 09:14:05 #topic General discussion (15 min) 09:14:27 it's difficult to know in advance how long a discussion will take. 09:14:52 hybris-15.1 ? 09:14:53 any updates on the socket activation and keyborads? 09:14:55 lbt, thanks. It's helpful to know who to target for other things too :) 09:15:02 chriadam_: that is why I always say "ask for more time than you anticipate the discussion will take" and also "never ask 5 minutes because nothing is discussable in 5 minutes!" 09:15:14 also the emoji KB. are we going to get more emojis on the next release? 09:15:51 Mr_Doge: in the works, please help testing https://github.com/libhybris/libhybris/pull/388 https://github.com/mer-hybris/droid-hal-device/pull/185 09:15:51 Jaymzz: originally I was just expecting a ACK/NACK for that PR in my topic ;-) 09:16:55 M4rtinK: fair enough, but that happened to almost all of the topics today and it is not OK in my opinion. I will ask everyone again on my minutes email to ask for more time than anticipated 09:17:22 ApBBB: my current understanding: presage is licensed as GPLv2, which means jolla-keyboard will have to be opensourced (and that's not an option at the moment), unless presage uses process separation. cc rinigus martonmiklos 09:18:08 ok 09:18:19 ApBBB: allowing socket activation haven't gotten round after the break yet 09:18:33 thanks 09:18:43 thanks for info 09:18:55 BTW we need a gesture to switch kbd layout fast. 09:19:07 ApBBB: that has been mentioned internally, so hopes up ;) 09:19:08 the press space one doesn't cut it 09:19:19 and eyes on sfos3 09:19:26 sledges: swipe from the keyboard edge and i';; be fine 09:19:49 which further deters any chances for SwypeUI kbd 09:20:28 sledges: i don't think they conflict. if the swipe is from outside the screen you can swipe as much as you want in the screen 09:20:37 if you're going to have swipe on keyboard edge then there should be a similar thing for Alien Dalvik to get rid of that stupid navigation bar. 09:20:41 from outside screen is minimising app 09:20:59 Does Jolla have any plans to sell community version of Inoi tablets? Just like Jolla C was a community version of AquaFish. 09:21:08 and we already have hide-keyboard-with-swiping-down gesture 09:21:14 sledges: it shouldn't be when the keyboard is active though 09:21:17 swipe up on keyboard? 09:21:28 sledges: thats what i am saying. if the kb is up & swipe from outside in the kb area change layout. if not minimize 09:21:32 Pekkeli: Not at the moment 09:21:35 sledges: this hide keyboard when swiping down could be changed or made configurable maybe 09:21:43 r0kk3rz: would still conflict with a possibility of SwypeUI-like feature in future, is what we're on about:) 09:21:48 why not have a dedicated button next to the prediction stuff 09:21:52 long press space bar to hide keyboard for example 09:21:54 sledges: n9 had it like that i thinkg 09:22:02 sledges: yeah like that will ever happen anyway 09:22:17 #justsayin 09:22:20 Ok, by the way do Inoi tablets have Android support? 09:22:31 bb10 used to have swipe down on keyboard to switch to second level. Though I think we can make it switch layouts and use long press to hide keyboard (like bb10 also did) 09:22:41 leszek, long press on space bar is already in use for language switching 09:23:07 rainemak: yeah I want it to go away that is the proposel 09:23:09 how often do you ned to change the kbd? isnt that usualy max once per conversation/topic/usage/context? is it really that bad to longpress spacebar? 09:23:14 and only use swipe down or a dedicated button 09:23:25 leszek, noted 09:23:35 rainemak: also space bar needs to be bigger. I am hitting the wrong keys all the time (. in this case) 09:23:37 phlixi_o: with emoji support this has become more an issue even for non-bilinguists 09:23:45 or swipe on space 09:23:52 so the lesser used keys need to be a little smaller 09:24:05 leszek, same here: always hit . instead of space 09:24:42 btw any thought of newer HW after you are done with the xa2 09:24:43 in general I would love a layout of the BB10 virtual keyboard. That is the only virtual keyboard I really could use for typing with less than 10% typos 09:25:16 * ApBBB in desperate need of a compact unit 09:27:13 ApBBB: unfortunately the longer you wait the less compact the devices will become :P 09:28:20 5 minutes left on general discusson 09:28:22 just got one, works fine (beside of not beeing stuck on 2.1.3.7 and lack of audio in sip calls) 09:28:23 the XZ2 Compact for example has a bigger screen 09:28:24 leszek: i'd be ok with a xz2 compact 09:28:49 its less wide than the J1 albeit a bit longer 09:28:58 the xz-notsocompact with no headphone jack 09:29:14 mattaustin: so looks like upstream support for Python 3.6 is planed till the end of 2021: https://www.python.org/dev/peps/pep-0494/#and-beyond-schedule 09:29:15 ApBBB: yeah that would be a flagship device 09:29:29 despite the fact Sony make many wired headphones 09:29:53 r0kk3rz: if i can get a quality usbc headphone pair by ie sennheiser rha etc i'd be ok with no headphone jack 09:30:04 is it possible to upload a version of webcat in harbour for older sfos and for newer sfos or does it needa newer version? 09:30:40 so 2.9.9 for sfos <2.2 and 3.0 for sfos >=2.2 09:30:46 what if you wanted to charge your phone at the same time 09:30:58 well you dont, obviously 09:31:21 mattaustin: for the Python 3.4 series, the latest security fixes-only release has been released in February 2018 09:31:27 mattaustin: https://www.python.org/dev/peps/pep-0429/#release-schedule 09:31:44 mattaustin: 3.4.8: https://www.python.org/downloads/release/python-348/ 09:32:20 so a good first step could be at least updating from 3.4.3 to 3.4.8 to get all the bug & security fixes 09:32:52 leszek: we need a flagship 09:33:31 pketo: could you comment on leszek's question please? ^^ 09:34:00 sledges: I suspect pketo is having lunch now as it is lunch time in Finland 09:34:05 go straight to the XZ3 as soon as that's announced 09:34:13 Time is up everyone, 09:34:33 I'm moving soon on to the time & date for the next meeting 09:34:51 #topic Next meeting time n date (5 min) 09:34:57 Thursday 28th of June at 08:00 UTC? 09:35:04 gimme +1 :D 09:35:11 +1 09:35:22 +disco 09:35:28 XD 09:35:36 /o/ \o\ 09:35:57 #info Next meeting will be held on Thursday, 28th of June 2018 at 08:00 UTC 09:36:57 @sledges: one question regarding the translation: the coordinator for the bulgarian section seems to be inactive, and Id like to overtake the role together with spaceumko 09:37:24 #info for everyone reading the logs: please request more time for your topics than you anticipate. I have asked this before and it was cool for a while but now we are back at having over-time for every topic. Try to ask more than 5 minutes as reading the # info takes longer than that usually 09:37:38 +1 09:37:40 HeinrichJolla: you'd need to announce your wish on the TJC just like everyone else 09:38:03 @sledges: I already did this ... but so far, no reaction 09:38:18 Alrighty, time to end the meeting. Thanks everyone for participating this time 09:38:19 patience 09:38:22 HeinrichJolla: i haven't received a notification, just checked sorry this email alert system has been playing up recently https://together.jolla.com/question/134467/official-announcement-translate-sailfish-os-elections-of-community-languages-linguists/?answer=134491#post-id-134491 09:38:40 Meeting minutes will be sent to you shortly 09:38:47 #endmeeting