08:01:13 #startmeeting Sailfish OS, open source, collaboration – May 17th 2018 08:01:13 Meeting started Thu May 17 08:01:13 2018 UTC. The chair is Jaymzz. Information about MeetBot at http://wiki.merproject.org/wiki/Meetings. 08:01:13 Useful Commands: #action #agreed #help #info #idea #link #topic. 08:01:21 #info Meeting information and agenda can be found here: https://lists.sailfishos.org/pipermail/devel/2018-May/008381.html 08:01:28 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. 08:01:39 #topic Brief introduction (5 min). Please prefix your name/handle with # info 08:01:47 #info James Noori, Sailor @ Jolla 08:01:57 #info Mister_Magister, Sailor @ VerdandiTeam 08:02:06 #info Leif-Jöran Olsson, community 08:02:15 #info Chris Adams, developer at Jolla 08:02:23 #info schmittlauch, community 08:02:25 #info Pekka Vuorela, developer at Jolla 08:02:26 #Lewis Rockliffe, community 08:02:38 #info John888, community 08:02:49 #info Lewis Rockliffe, community 08:02:52 #info Frajo Haider, Sailor @ Jolla 08:02:53 ups :) 08:03:07 #info Andrew Branson, sailor 08:03:23 #info David Llewellyn-Jones, community 08:03:39 r0kk3rz: you have wonderful lastname 08:03:46 #info Simonas Leleiva, developer at Jolla 08:03:57 #info martonmiklso, community 08:05:40 #info Leszek Lesner, community and dev 08:06:25 #info Raine Mäkeläinen, developer @ Jolla 08:06:55 #topic Sony Xperia XA2 model support (5min – asked by carmeloferso ) 08:07:07 #info It was already disccussed on the meetings that the models that Jolla is working are H3113, H4113 for the Sailfish X image. I would like to know if the others models as H3123 (my current phone), H3133 and H4133 will be supported also, as I was comparing and what I can notice is that only LTE bands are different and also if supports a dual sim card or not. 08:07:43 #info Unfortunately, only the ones in Sony Open Devices Programme can be considered to either become supported or not: https://developer.sony.com/develop/open-devices/get-started/supported-devices-and-functionality/ 08:07:53 your mentioned models aren't amongst them 08:08:41 Followup question to that topic: Dual SIM models: 08:08:41 H4113 (EU region) 08:08:41 H4133 (International - most bands supported, including TD-LTE) 08:08:41 Single SIM models: 08:08:41 H3113 (UAE+EU region) 08:08:42 H3123 (CA+US region) 08:08:44 H3133 (International - most bands supported, including TD-LTE) 08:08:50 any of these are going to be supported? 08:09:49 thanks for detailing the variants, yet they first need to have Sony's AOSP support as mentioned above 08:10:08 hi. Am I later? 08:10:22 m4g0g: a bit 08:10:29 first topic is running 08:11:09 TL;DR to answer that question for good: Only these models are going to be supported: Xperia XA2 (H3113, H4113), is that correct? 08:11:12 From my understanding until now, we will have support for the dual-sim variant of the XA2. About the exact model numbers, as sledges mentioned, those that are supported by Sony's open devices program will be on the list. 08:11:36 Jaymzz: correction, only a selection from that list 08:11:39 not all:) 08:11:46 That is correct John888 08:12:04 Yes, bad wording from me :) Thanks sledges 08:12:10 technical info: at the moment, AOSP lunch build combos are available for aosp_h3213-eng, aosp_h4213-eng, aosp_h3113-eng, aosp_h4113-eng 08:13:04 matches the list online basically, and grepping the sources for e.g. 3123, doesn't return a thing 08:13:22 Thank you Jaymzz - So to summuarize it: Supported is going to be only: H4113 (dual sim, EU region) and H3113 (single sim UAE+EU region) 08:13:41 that is the current line-up, yes 08:13:45 John888: Indeed. 08:13:57 We have to move on now as the time is up for this topic 08:14:05 * ApBBB still bummed about hw selection 08:14:10 #topic Wayland update (15 min – Asked by r0kk3rz) 08:14:11 Sure, answered, thank you 08:14:30 #info Lately more projects are releasing working Wayland support (eg. Firefox) but they require a newer version of Wayland than we have available. Wayland 1.6 is quite old and before any of the APIs were declared stable. Mal has done some prelimiary work here: https://build.merproject.org/project/show/home:mal:update Does Jolla have any immediate plans to update Wayland? 08:14:40 #link https://build.merproject.org/project/show/home:mal:update 08:14:57 I think wayland needs to be updated for Qt 5.9 anyway 08:15:02 mal alwayas do everything xD 08:15:04 so for the upcoming SFOS 3 08:15:23 Ok so, I was looking into building firefox-wayland and starting with GTK3, i got it built but it really wants at least wayland 1.9 08:15:31 firefox wayland is not there yet. they are working on it and as we speak they are solving the EGL support related bugs. 08:15:44 obviously, but thats not the point :P 08:16:17 r0kk3rz: like I said if SailfishOS wants Qt 5.9 they need to update to a newer Wayland anyway as otherwise it will get a crashy mess 08:16:18 yes it was just a notice. i am waiting firefox wayland 08:16:23 for the desktop 08:16:50 leszek: yes it seems we arent using qt 5.6 qtwayland yet... 08:17:21 uff we aren't? That I did not know 08:17:52 wayland update hasn't been discussed, but i hope it's just a simple version bump. it's not patched or anything. 08:18:12 and yea, it's getting old so maybe should just do it. 08:18:15 maybe then there is a reason there was no wayland update 08:18:22 pvuorela: in the link posted mal was testing an updated version, and had some issues with it 08:18:33 so that explains it 08:18:35 i forget exactly what 08:18:42 pvuorela: i would be happy if packages were updated every release not because "its getting old" :) 08:19:09 Mister_Magister: it requires packaging work and testing, its not magic, someone needs to do it 08:19:11 #info wayland update hasn't been discussed, but i hope it's just a simple version bump. it's not patched or anything. and yea, it's getting old so maybe should just do it. 08:19:23 Mister_Magister: we are not going to sync every package for every release :) 08:19:26 r0kk3rz: i know but still other distributions manage that somehow 08:19:50 r0kk3rz: QtWayland API changed significantly from 5.4 to 5.6, so we'd have to rewrite our compositor, probably quite a bit of effort there. 08:19:52 Mister_Magister: they have more people like mal who do things 08:19:57 SailfishOS has a CI system so it should be possible 08:20:18 chriadam_: ah i see, this would be on the open lipstick side? 08:20:23 no idea. 08:20:26 chriadam_: yep and from 5.6 to 5.8 it broke completely. So when rewriting think maybe about 5.9 08:20:29 like openal not working and not updated since 5 years xD 08:20:44 r0kk3rz: yes. 08:20:46 5 minutes left on this 08:21:03 but I think it is kinda been answered already 08:21:17 leszek: I personally agree that 5.9 makes sense as next target 08:21:46 yeah in closing, imo it should be on the list for sfos 3 / qt 5.9, but we can do some testing in the community :) 08:22:16 anything the community can do to help (preliminary effort, testing, smoothing bumps in the road, even just identifying specific issues) would be greatly appreciated. 08:22:28 r0kk3rz: pvuorela: can we bump wayland itself using mal's effort already now-ish? and then only QtWayland will be left to deal with as and when? 08:23:41 iirc he said lipstick was crashing :D 08:23:54 ah minor detail right :D 08:24:08 who really needs that anyway, its not important 08:24:12 if its in the opensource part it can be fixed 08:24:29 sledges: haven't checked. i'd expect wayland being backwards compatible. 08:24:37 might be a max version of wayland that the current version of qtwayland will work with? 08:25:10 abranson, that's sounds like right to me 08:25:13 wayland should be backwards compatible. QtWayland is another thing 08:25:22 qt wayland up to a point was using an old protocol and not xdg_v6. no idea if this matters 08:25:42 was causing a lot of pain in the desktop side of things 08:25:58 yep I remember it for Plasma 08:26:08 Qt 5.8 basically was unusable for wayland 08:26:43 as far as i remember lipstick was crashing only in very specific circumstances for mal 08:27:16 AsteroidOS has newer version of Qt and Qtwayland working with asteroidos' lipstick. 08:27:37 eekkelund: good to know! 08:27:41 oh interesting 08:27:50 so fixes can be imported maybe 08:28:48 guys better wrap up, not mich left on this one 08:28:54 yeah i'll chat to kido and see if he remembers anything 08:29:05 Jaymzz: i think we can move on 08:29:17 alright thanks r0kk3rz 08:29:21 #topic Everything being rendered on CPU instead of GPU (20 min – asked by Mister_Magister) 08:29:31 #info Tested on 10 phones including jolla phone and xperia x. When scrolling in browser or webkit there is no gpu usage but there is high CPU usage. There is no GPU usage at all when watching video. Test method: play video in lls player (it uses sam stuff as gallery) and check gpu usage in aida64 08:29:55 now the important question :D Also changing gpu frequency to lowest chagnes nothing other than page transition fps 08:30:42 yeah either the tools for watching over gpu usage are broken. libhybris is masking something, or almost everything is rendered over cpu which would explain the choppiness on most community ports 08:31:01 erm, gpu != hardware video decoder 08:31:02 leszek: tools arent see my frequency changes messsage 08:31:32 tbr: we do not talk about video decoding 08:31:32 tbr: yep gpu is used for decoding but should be used for rendering aswell 08:31:49 how was the gpu usage measured? aida64 doesn't update the cover with the current gpu usage when it is minimized 08:31:53 leszek: i dont even know if its used for decoding 08:32:11 krnlyng: play video in lls player in background and then check aida 08:32:21 Mister_Magister: I think it is. At least the gpu has that built in 08:32:25 or simply change govenor of gpu to powersafe and nothing changes 08:32:33 powersave* 08:32:37 Mister_Magister: which version of aida were you using? droid or sailfish? 08:32:43 ofcourse sailfish 08:32:55 i tried this on my xperia, and saw gpu usage when scrolling aida 08:33:01 or run htop and playback a video, scroll the interface or do anything and you'll see high cpu usage which should not be 08:33:29 and also frequency changes change nothing… ui is same videos lag as ealier 08:33:40 abranson: xperia x might be not affected. But Jolla1, JollaC and ports for Moto X2 and others are 08:34:00 this wasnt a problem cause cpu handled HD FHD just fine but QHD is a problem 08:34:04 the sailfish version doesn't update when peeking, but using the droid version I saw gpu usage when LLs video player was playing on its cover and I peeked away aida 08:34:10 Mister_Magister, but then aida64 is in the foreground, thus there is no need to render anything by the video player, so no/low gpu usage is kind of expected? 08:34:11 engines do image rasterization on cpu side 08:34:26 krnlyng: then feel free to change frequency 08:34:36 ^ engines that we're using 08:34:55 Mister_Magister: might be good to check the capabilities of the video codec in the /system xml. some of them have max sizes that they will handle 08:34:57 id be really surprised if thats the case on Jolla 1, a lot of effort was put in to getting that smooth 08:34:59 rainemak: so that explains the high cpu usage whenever scrolling an app? 08:35:08 abranson: it even on jolla phone 08:35:17 for instance we have a couple that refuse to do > 1080 08:35:21 remember all the transparency effects sailfish used to have? you sure all that was on CPU? 08:35:33 r0kk3rz: you can try it out. Scroll browser or webcat and see high cpu usage with htop 08:35:57 Mister_Magister: I only tested on the xperia - that was mentioned in the question. can look at the Jolla1 later. 08:36:00 and no gpu usage leszek 08:36:08 r0kk3rz: I am not sure it is on CPU but the CPU usage is too damn high when performing those 08:36:13 abranson: we tested that on 10 phones already 08:36:24 leszek, explains a bit at least in case of browser and webkit 08:36:28 if you dont trust aida change frequency of gpu to lowest 08:36:37 gpu run the same, videos play the same nothing changes 08:36:49 there is no gpu usage at all it doesnt go to max frequency 08:37:21 yep thats also kind of another issue. The gpu frequency never goes to the max 08:37:28 for moto x force with QHD screen doing something like switching apps causes audio to stutter of cpu usage 08:37:31 and it should actually 08:37:46 what about games? using sdl2 or something? 08:37:49 hello, sorry I'm late 08:37:56 if it was rendering on GPU then setting frequency to lowest should change something right? 08:38:20 r0kk3rz: would need to compile some sdl demo but i think it would use gpu as it uses gles libraries directly 08:38:43 directly ... you mean through hybris 08:38:57 just launched aida on my jolla1. scrolling the display page shows gpu usage. 08:39:00 hey stephg, no probs! welcome 08:39:02 if you think about it… 08:39:33 abranson: try playing video in background 08:39:38 or in browser 08:39:56 even if it shows some gpu usage it shows too much cpu usage 08:40:00 also xperia had some bad color conversion issues on the webkit side (bgra) that were expensive 08:40:08 I also can't believe everything is rendered by cpu but some stuff at least is as it seems 08:40:08 a bit better on 2.2.0.x 08:40:11 what do you mean in the background? if it's not being displayed then it might not use the gpu 08:40:23 for cpu usage, it needs to be compared to something else, like android 08:40:27 what makes you say its too much? have you tried conducting similar tests on android? 08:40:30 rainemak: yeah I worked on that fix with the qtwebkit main dev :P 08:40:30 Mister_Magister, in general if libhybris works then hw acceleration should be active, since libhybris doesn't load androids pixelflinger (sw rendering) but actual hw drivers, the performance issues probably have a different root cause. 08:40:32 abranson: then change frequency and go watch some video 08:40:34 we don't know what is too high without it 08:41:22 krnlyng: but it doesnt render on gpu 08:41:26 it only partially uses it 08:41:38 i wouldnt expect much gpu usage with watching a video, the hardware decoders take care of the heavy lifting 08:41:55 abranson: scroll qtwebkit based browser on android and do it on sailfishos. You'll see the difference 08:42:01 r0kk3rz: sure, but would you expect much cpu usage in that case? 08:42:14 Simply: change gpu frequency to lowest and notice no change. change cpu frequency to lowest and oh boi 08:42:29 Mister_Magister: please stop repeating yourself :P 08:42:35 r0kk3rz: i have tp 08:42:39 no, you dont 08:42:39 to* 08:42:49 yes i do cause nobody seems to notice 08:43:53 anyway, sounds like this could do with some more investigation to me, maybe theres some benefits to be gained 08:43:57 10 minutes left on this topic 08:44:10 Jaymzz: and topic was barely explained xD 08:44:12 certainly comparing with android performance would be a start 08:44:40 there is also problem with aliendalvik but that is not directly sfos problem 08:45:16 this needs a more precise definition. there are too many factors flying around here. 08:45:18 r0kk3rz: for example that audio doesnt stutter on android but it does when switching apps 08:45:26 abranson: what do you mean 08:45:36 Mister_Magister: there is always time furing next meetings ;) 08:45:47 Jaymzz: naah :P 08:45:48 r0kk3rz: one test that shows that something is off is scroll the app overview and have htop opened in the background. At least one core of the 2 the Jolla1 provides is at 100% and should not be for that kind of simple action 08:45:53 audio gets shutters when ie. locking the screen 08:45:57 J1 08:46:24 abranson: already told you precise definiton 08:46:39 leszek: im not saying you're wrong, simply trying to be a bit more scientific about it 08:46:41 [10:42] Simply: change gpu frequency to lowest and notice no change. change cpu frequency to lowest and oh boi 08:46:41 leszek: qtwebkit still has places where it uses bgra, so no the best test candidate given the problem (main browser, videos) 08:47:10 r0kk3rz: I know I know. It is just a observation we hoped someone @jolla could do further investigation on this 08:47:31 sledges: the Jolla1 has also bgra issues? 08:47:34 Mister_Magister: cpu does a lot of work even with video playback. that doesn't really tell you anything. we need something concrete. 08:47:49 abranson: it is concrete 08:47:50 I played a video with LLs player, and saw gpu utilization 08:47:51 leszek: mainly xperia, unsure about jolla1 08:48:15 abranson: and i didnt at all on 5 phones includfing jolla 1 08:48:29 abranson: change frequency 08:48:41 sledges: the problem is not on the Xperia X soo much. more on jolla c and ported devices like Moto X2 or Nexus 5 were it stutters with bgra enabled and with it disabled (+patched qtwebkit which disables it) it stutters even more 08:49:00 if it were to use gpu setting frequency to lowest would matter but it doesnt 08:49:15 leszek: and jolla phone 08:49:17 Mister_Magister: just tried it on jolla1 and saw gpu being used 08:49:34 like for 1 second? 08:49:48 change frequency 08:49:57 leszek: from what i heard, there are many different places in the pipeline that utilise bgra, it's hard to get to all of them, and hasn't been done yet 08:50:22 Mister_Magister: Time is up. Do you need 5 minutes more? 08:50:30 yes please 08:50:33 yeah, but I wouldn't expect the gpu to keep going on the video once aida is in the foreground 08:50:37 can't give you more than 5 as there are other topics waiting 08:50:45 sledges: still does not explain why its not running smoothly on jolla1, jollac and others that don't suffer from bgra driver issues 08:50:46 #info added 5 mins to this topic 08:50:54 abranson: then change frequency x10 and go around gui or play video 08:51:07 i told already 10 times to change frequency 08:51:22 we know :P we saw 08:51:28 right, but that's not something I'm going to do during the meeting is it. 08:51:38 why not 08:51:39 gpu frequency doesn't matter if the gpu is capable of running the task you're making it do at the lowest frequency just fine. the performance issue could be explained by a problem with cpu usage alone 08:51:54 i'll try later, but like I said, this needs a concrete definition and comparison to other playforms 08:52:21 -> so it needs to be clarified/determined what the actual issue(s) are 08:52:27 yep what worries me is the cpu usage. 100% for just scrolling on the app pages is not very efficent 08:52:29 krnlyng: it is possible that its rendered on gpu and while doing stuff its also causes high cpu usage and the thing that causes cpu usage is different 08:52:40 but i would expect something more on gpu really 08:53:01 maybe its doing some copying with CPU while rendering on GPU 08:53:27 2 minutes left. Please try to wrap up as I won't be waiting (lots left to discuss) 08:53:31 maybe sfos gui is light enauygh to render it on lowest frequency 08:53:37 but cpu usage is bad 08:53:39 I remember that being the case for high cpu usage on the pinebook with plasma where we needed to patch the opensource part of the blob to make it somehow work smoothly 08:54:00 mobile gpus are pretty tanky these days, id expect them to run the homescreen basically on idle 08:54:04 but that is all guessing right now. 08:54:05 there was a lot of work done on zero-copy as far as the video buffers were concerned, to avoid copying on that side 08:54:14 but there's still demuxing etc that will take cpu 08:54:33 Basically what we want is someone at jolla taking a look at that. We can assist on helping describing the issue in more details and such if necessary 08:54:33 Can we put together what test cases are needed to further investigate this issue? 08:55:00 on 2.2.0 there were also small improvements to the HWC utilization 08:55:06 abranson: krnlyng: i would be happy to debug that further but i dont have enaugh knowledge for it 08:55:20 *enough 08:55:21 Mister_Magister: time's up man 08:55:32 cpu usage when watching video and playing apps is really bad thats all i know 08:55:32 krnlyng: yep I know already applied the new config on Xperia X and Moto x2 (partly just the buffers set to 3) 08:55:34 amen 08:55:39 Jaymzz: go close 08:55:42 moving on 08:55:44 #topic Share contact details via SMS (10 min – asked by martonmiklos) 08:55:51 hey all 08:55:56 title says all 08:55:56 #info Recently I started to came across this issue or missing feature weekly: https://together.jolla.com/question/7695/share-contact-carddetails-via-sms/ and it started to hit my stimulus threshold, to look for some solutions. Before starting to implement a plugin I would like to doublecheck that this not on the short term roadmap. 08:56:08 I would like to share contact details via SMS 08:56:08 hi martonmiklos :) 08:56:23 #link https://together.jolla.com/question/7695/share-contact-carddetails-via-sms/ 08:56:23 doing it manually by copy paste it is a bit cumbersome 08:56:39 I can have a capabilities to solve this issue 08:57:01 but before doing it I would like to get clarification that it is not something that already being worked on 08:57:29 and the fact that I have not found similar on openrepos does not means it not exists 08:57:49 >topic says it all >talks about topic for next 5 messages xD 08:58:15 afaik it's not being worked on within Jolla currently. as for whether it's on roadmap -> rainemak was that part of business features or not? 08:59:06 does it considered business feature ? 08:59:12 sms contacts sharing is not part of business features 08:59:54 and not on roadmap either 09:00:10 thanks for the clarification 09:00:30 I guess if we frame our feature proposals as business feature, the chance of them getting implemented increases hugely :P 09:00:31 martonmiklos: I assume this is different from MMS sharing of contact? because we have a sharing plugin to do MMS sharing already. so I guess you're referring to SMS specifically? 09:00:54 correct 09:00:58 M-schmittlauch: That could actually be a good strategy 09:01:13 my plan includes free SMS, but not MMS 09:01:25 this is why I prefer sharing contacts via SMS 09:01:41 and as I see from the TJC post I am not the only one 09:02:24 martonmiklos: cool. sounds great :-) please don't hesitate to ping me if you strike some roadblock etc. 09:02:41 is this like, sending a vcard over sms? 09:02:47 nope 09:02:59 I imagine this feature similar to the MMS one 09:03:11 pick a contact in People - > share -> SMS 09:03:31 and you get to a message sending screen where you can select the share target contacts 09:03:36 i think it does internally though. the sms text is a vcard. 09:03:48 and the ui hides that 09:03:49 in the case of MMS there is a vcard attachment 09:04:04 yeah, so sending a vcard over sms :P 09:04:17 in my case I would simply generate the contact info to the message text 09:04:24 Is the idea the other phone picks up the contact automatically? Is this supported on other phones? 09:04:41 well 09:04:43 i remember it used to be 09:04:52 martonmiklos: just FYI there is 1 minute left. do you need more time? 09:04:57 nope 09:05:02 I got the answer 09:05:11 I will came up a plugin soonTM 09:05:21 :-) 09:05:25 ah, nokia used to do that 09:05:42 haha, okay, moving on then 09:05:42 :P 09:05:59 #info he got his answer and will come up with a plugin soonTM 09:06:28 For the next topic, for some reason I forgot to implement it into the email I sent to you all. But here we go anyway 09:06:32 #topic SailfishOS for the Youyota tablet; comments from Jolla's perspective (10 min – asked by schmittlauch ) 09:06:45 #info While the Youyota Tablet (an attempt to rebuild the Jolla Tablet) more and more looks like it's also failed, some rumors suggest that this is partly due to higher licensing fees demanded by Jolla in advance then it used to be the case. Understanding the legal circumstances like NDAs, can Jolla sonmehow comment on the state of this project? Going to a meta level, how hard would it be to provide official SailfishX licenses in 09:06:45 case the tablet was released without any OS/ without Sailfish OS 09:07:33 surely os licensing is the least of this projects worries 09:07:51 Not so many comments from Jolla has been released about Youyota. But I can tell you that the rumor is definitely not true about licensing fees. 09:08:10 Jaymzz: that topic was added very late, that's why it wasn't in your mail 09:08:57 So just to take this excuse away from them: How hard woud it be to make a Sailfish X release in case the tablet was released without OS? 09:09:20 There is a reason we haven't talked about this in public, and that is because we wanted a more solid base and agreement before we speak about this publicly. We unfortunately never reached that, but it is not too late. 09:09:21 given that it 1. ever comes out and 2. they manage to stay close to the Jolla tablet 09:09:52 M-schmittlauch: do they have a community-level port for youyota tablet? 09:10:10 M-schmittlauch: It should not be that hard, as we already have the firmware and the OS available for Jolla Tablet, this should not take huge resources from us. But I'm not sure if that will ever be the case. 09:10:26 I know this is very frustrating for everyone involved, but I think that this is still an important development device for the community 09:10:28 given jolla never did this for any aigo tablets, and i think expplicitly banned their MAC addresses... 09:11:18 it is unfortunate to see the tablet project fail for a second time though 09:11:48 It's just that Youyota's communication is a trainwreck, and if there are good possibilities for the Sailfish X way one could tell them to release it without an OS if it was the only possibility to get that project through 09:11:59 They haven't annouced any faliure yet right? 09:12:07 (btw, sorry for the late topic suggestion) 09:12:18 M-schmittlauch: No worries :) 09:12:20 *Jaymzz officiall no 09:12:54 No, but neither any further developments since some months. They basically turned completely silent 09:13:11 alright in that case I won't bring all the hopes down as a "second failure". There still is hope. I'll try to dig into it (unofficially, i.e not as a Jolla employee) 09:13:12 which is never a good sign 09:13:41 its cursed 09:13:50 I agree. That's why I'm going to do some digging and will get back at you probably here if I find anything out 09:13:54 But e.g. chenliangchen who helped them with communication to the community, also got locked out and heard nothing for a while. 09:14:18 (for the record: chen isn't reponsible for any of this) 09:14:25 General section already started or not? 09:14:39 Yes, unfortunate events took place with communications side of things :( 09:14:47 m4g0g: not yet 09:14:49 Jaymzz: thanks, even an official cancellation would be better than the project just hanging in the void eternally) 09:15:07 M-schmittlauch: Sure :) so, shall we move on? 09:15:15 fine for me 09:15:26 #topic General discussion (15 min) 09:15:30 #info John888: more on your topic: if you can build Sony AOSP for H3113 and flash it to your H3123 (as only modem baseband differs) successfully, then our image will most probably be working for you too 09:15:34 #link https://developer.sony.com/develop/open-devices/guides/aosp-build-instructions/build-aosp-nougat-8-1-oreo-4-4/#tutorial-step-4 09:15:45 sledges was ready with ctrl-V :D 09:15:48 Some questions. sledges any updates regarding the daemons situation (so we can have Who Maps) 09:15:53 Hello guys. I have some questions. First one: Any info about new API in SFOS-3? 09:16:20 also this bug https://together.jolla.com/question/97109/email-imap-idle-doesnt-work-with-both-connections-active/?answer=175138#post-id-175138 there was some work done by damien and there is some more info on the comments 09:16:34 can someone create a bug in the internal bug tracker 09:16:41 im wondering what to do about my topic cause im left with more questions and no answers. Bring it up on next meeting? 09:16:43 m4g0g: that was answered in couple of previous meetings; at the bottom of page: https://jolla.com/sailfish3/ 09:16:44 #topic : Are any/all apps such as: Opera Mobile, Telegram, WhatsApp, WeChat, Skype, email client (at best Aquamail), Authy (or other 2FA app) and OpenVPN fully working with the current build such as aosp_h4113-eng mentioned above? 09:17:24 Mister_Magister: create a step by step guide on how to replicate and start a tjc topic 09:17:37 Just one thing that might be something Jolla could take a look at now that it has the license to develop alien dalvik further is some sort of bridge to be able to expose wifi and bt in aliendalvik as some apps like camera remote apps or fitness tracker apps or smartwatch apps need this. So they need to have direct wifi support to change the wifi network for example. 09:17:41 that would take too long xD 09:17:52 What about supporting uri handlers? 09:18:12 @sledges, Thanks but I don't currently own any of the phones, I came here essentially to find one that I need to buy and is at least somewhat working 09:18:35 also any updates on the keyboard agreement jolla was working on? 09:18:38 Mister_Magister: do more investigation, or leave it to jolla 09:18:41 John888: then I'll get back to you next meeting, if you can't find anyone else to test that for you:) 09:19:00 r0kk3rz: if someone tell me what can i do i would do some more investigation 09:19:23 Thanks a lot sledges - I'm afraid you are the only volunteer :) 09:19:44 John888: I'll ask Sony AOSP team themselves 09:20:13 sledges: can you also ask them for xz2 compact aosp support 09:20:34 if you don't ask, you don't get, hehe:) 09:20:37 Third one: sharing: sfos->android(there is plugin in openrepos)/android->sfos. And the last question: any info about changing harbour restrictions: share plugins, accounts, etc? 09:20:41 That's great, thank you also if you could ask them to test Mycelium app ? 09:21:22 what about outsourcing ports? 09:21:27 m4g0g, I'd also like to know about relaxing harbour restrictions. Particularly multiple executables. 09:22:11 as there are tons of good ports jolla could let someone port phone (do initial work) and then continue with working port and make it official 09:22:33 we need compact hardware 09:22:42 the phones are idiotically big 09:22:42 related to previous topic of wayland update, after simply updating wayland only sdl apps have issues (not sure if because of also updating sdl), when updating qtwayland lipstick crashes (I haven't figured out why, but I had to patch lipstick to support the new version) 09:22:46 Btw is there any reason why Sailsfish OS is currently sold only to EU? Is it political or some other reason? 09:23:32 John888: +1 to question. Can't buy legally in Belarus/Russia 09:23:34 John888: and RU, and bolivia? wherever Jala is based 09:23:48 John888: No political reasons, we are expanding Sailfish X sales outside Europe. No ETA yet but should be in this Q 09:24:01 nice 09:24:16 Jaymzz: give us smaller phones 09:24:28 Mister_Magister: I think one issue during the meeting was that test setup and results were a bit unstructured. So if you can somehow put it together in a structured manner that could be useful 09:24:49 ApBBB: you didnt like the last one someone did... 09:24:49 so if I'm outside the supported regions it's simply refused to sell or now warranty or how does it work ? 09:25:06 ApBBB: I'd be the first in line for a Sony compact. Especially the X compact. Such a fantastic reminder of my old N9s. But unfortunately that is not on the roadmap yet. 09:25:17 M-schmittlauch: i think it was clear enaugh… 09:25:30 *enough :P 09:25:34 ApBBB: no updates on daemons, we've narrowed down last time to allow only socket-activated daemons in harbour, but still need to deal with potential risks arising, possibly by increased vetting of the apps 09:25:39 leszek: xd 09:25:49 Mister_Magister: just write it down in a list what to do to reproduce 09:25:55 Say I'm from Russia, and I ordered the Sailfish OS, does it mean I breached the license? Will it be voided or what happens? 09:25:59 Mister_Magister: needs moar spreadsheets, some kind of base line, a test, and results 09:26:03 and put on tjc 09:26:36 Jaymzz: i'd be first in line for an xz2c 09:26:45 there is also some weird 30% cpu usage on x force when no process uses cpu xD 09:26:54 ApBBB: long way to go for XZ stuff. Keep your fingers crossed 09:26:57 yep xz2compact looks nice 09:27:08 5 minutes left on general discussion peeps 09:27:13 Jaymzz: the problem is my J1 is getting old :/ 09:27:24 Jaymzz: https://together.jolla.com/question/97109/email-imap-idle-doesnt-work-with-both-connections-active/?answer=175138#post-id-175138 can you have someone look into this 09:28:02 martonmiklos: could you please follow up my comment, i.e. was I talking sense there :) i feel similar to try to introduce the standard lithuanian kbd layout on mobiles, that no-one would use, but it's so convenient once gotten used to :D https://together.jolla.com/question/179903/zero-position-in-the-hungarian-layout/?comment=181955#comment-181955 09:28:23 leszek: it has its faults but i can live with it. its small enough. 09:28:34 ApBBB: Sure. 09:29:14 Jaymzz: its an old bug and on something that should work flawlessly. thanks. 09:29:22