08:00:09 #startmeeting Sailfish OS, open source, collaboration -- 11th Jan 2024 08:00:09 Meeting started Thu Jan 11 08:00:09 2024 UTC. The chair is rainemak. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:00:09 Useful Commands: #action #agreed #help #info #idea #link #topic. 08:00:11 Hello! 08:00:20 #info Meeting information and agenda can be found here: 08:00:20 #link https://forum.sailfishos.org/t/community-meeting-on-irc-11th-january-2024/17628 08:00:20 I am the meeting's chairperson today, and will be doing my best to keep time and order. Please respect the timings and bee-hive. 08:00:20 #topic Brief introduction (5 min). Please prefix your name/handle with #info 08:00:39 #info Raine Mäkeläinen, sailor @ Jolla, chairperson today 08:00:45 #info Ville Nummela - sailor@Jolla 08:00:48 Pleased to see many topics. Last late submission was already postponed as we have topics for an hour (schedule wise already too many). If it looks that we're running out time, I may postpone a topic or two from the end. 08:01:10 * LouisDK community member 08:01:15 #info Otto Mäkelä, community 08:01:21 #info nephros, Community 08:01:28 #info communiy member 08:01:33 #info Matias, community 08:02:12 #info sebix - community 08:02:16 #info David Llewellyn-Jones, Community 08:02:38 #info Crabster - lurker 08:03:07 ah, now 08:03:25 * nephros is actually joining from work ( ;) ), so answers *may* be delayed. 08:03:43 good morning. 08:04:45 (Good morning, and in case I don't see ya, good afternoon, good evening, and good night) 08:04:45 Morning poetaster! 08:05:05 #topic XDG-Desktop-Portal specification: interesting for Sailfish OS? (10mins -- asked by nephros) 08:05:13 Morning flypig! 08:05:17 #info This is rather long question. Thus, let’s use original message from the forum 08:05:17 #info Here's the question: 08:05:17 #link https://forum.sailfishos.org/t/community-meeting-on-irc-11th-january-2024/17628/2 08:05:59 #info Certainly, having common APIs is a good idea. More planning is 08:05:59 #info needed and putting down mentioned high-level guidance. We could 08:05:59 #info plan this on a separate forum post. Worth to analyse how much XDG 08:05:59 #info Desktop Portals could eventually help on mobile. 08:06:52 Or forum wiki page maybe. 08:07:59 Big part of this would be in implementing the backend... something like xdg-desktop-portal-sailfishos 08:08:01 Big part of this would be in implementing the backend... something like xdg-desktop-portal-sailfish 08:08:15 So, XDG-Desktop-Portal would help with Sandboxing; porting; and just generally staying consistent with other distros. Does that sound about right? 08:08:28 When I say "porting" I mean porting apps. 08:08:37 It could yes. It's its intention. 08:08:38 Yeap, that's about right. 08:09:13 But its made by the Flatpack guys, so some implementations may bve specific to them. Still, having an API spec is good IMO. 08:09:42 This would grow about from the Flatpak similarly as xdg-dbus-proxy did 08:09:42 Example: Portals specifies a DBus call for taking screenshots. 08:09:51 s/would/could/ 08:10:11 the question of getting 'secure' dbus communication is also adressed by xdg-dbus-proxy? 08:10:33 If there are problems with the spec for non-Flatpack scenarios, that might be something they'd want to knwo about. 08:10:34 I think first step could be identifying where there is overlap of existing functionality and specification. 08:11:07 (and identifying which parts of the spec are not interesting to SFOS). 08:11:40 #info API reference is extensive 08:11:41 #link https://flatpak.github.io/xdg-desktop-portal/docs/api-reference.html 08:12:06 nephros: a sort of mapping between existing dbus interfaces and xdg-desktop-portal interfaces. 08:12:13 #info XDG-Desktop-Portal would help with Sandboxing; porting; and just generally staying consistent with other distros. Does that sound about right? 08:12:16 BTW, newer versions of Mozilla use Portals plus PipeWire for video chatting 08:12:43 #info I think first step could be identifying where there is overlap of existing functionality and specification. 08:13:02 #info (and identifying which parts of the spec are not interesting to SFOS). 08:13:33 the use case from rinigus puremaps talking via dbus to osm server, for instance? 08:13:55 poetaster, I don't see how this would help with that 08:14:07 poetaster: I don't think there is a Portal spec for that. 08:14:24 he's just using dbus 08:14:40 do we have this one covered? 08:14:45 nephros, he mentions it on the page you linked 08:14:45 Presumably there could be a community-driven effort to implement something like the screenshot interface, to see how it might work? 08:15:12 rainemak, it's currently not sailjail compatible and so puremaps from harbour is gutted. 08:15:22 #info Presumably there could be a community-driven effort to implement something like the screenshot interface, to see how it might work? 08:15:35 rainemak, that's a good ida. 08:15:49 So a SFOS-portal-dbus-proxy? 08:15:49 Let's move on to next topic. 08:15:56 Cool find nephros. 08:16:01 :) 08:16:07 #topic Unlock via dbus (10mins -- asked by piggz) 08:16:07 +1 08:16:21 #info I made a fun little hack which allows a user to auto-unlock the 08:16:21 #info phone if some configured BT device is connected (GitHub - piggz/ 08:16:21 #info harbour-smartunlock 3). Unfortunately it required a nasty hack in 08:16:21 #info nemo-qml-plugin-devicelock to allow unlocking via dbus (GitHub - 08:16:21 #info piggz/nemo-qml-plugin-devicelock at dbus-unlock), I can envisage a 08:16:23 #info setup where the user enters the unlock code into some settings, and 08:16:24 #info a dbus method that accepted the code could do the unlocking, which 08:16:26 #info would be a bit better. This currently works great on the 08:16:28 #info PInephones, but not on my phone which uses sailfish-devicelock-fpd. 08:16:30 #info That closed part does link to /usr/lib64/libnemodevicelock.so.0 and 08:16:30 oh, im here, but have to do school run in a few minutes :) 08:16:32 #info exposes the same dbus method, but replies with an Auth error before 08:16:34 #info hitting the modified code in MceDeviceLockAdaptor::setState() I 08:16:36 #info think this is because it statically links the library, so my 08:16:40 #info override in the dynamic lib doesnt work. Would you be interested in 08:16:42 #info adding some real way to auto-unlock a device? What are your 08:16:44 #info thoughts on this? My use case is to not need to unlock the phone if 08:16:46 #info it know im “close by” … eg, if connected to my watch or my car" 08:16:48 #link https://github.com/piggz/harbour-smartunlock 08:16:50 #link https://github.com/piggz/nemo-qml-plugin-devicelock/tree/dbus-unlock 08:16:52 #info Yes, there some interest to support community with something like 08:16:54 #info this. Your patch touches the same function we have modified in some 08:16:56 #info in-house hacks / demos – so it is one control point that could be 08:16:58 #info used. But it is also very much against the rule of thumb: anyone 08:17:00 #info can lock device, only device lock daemon can unlock it based on 08:17:02 #info something user does. Pure sw unlock easily becomes a backdoor. At 08:17:04 #info minimum such possibility should be limited to privileged processes 08:17:06 #info (our patch allows only root to unlock). Preferably user should 08:17:10 #info still be in control over something like this – say setting allowing 08:17:12 #info unlock over D-Bus , bit similarly to relaxing rules about package 08:17:14 #info installation works. Or limiting such functionality to named client 08:17:16 #info binaries. The smaller the hole the better but nevertheless there 08:17:18 #info will be a hole. Does the possibility for increased convenience for 08:17:20 #info some justify decrease in security for all? 08:17:22 #info 08:17:24 #info Another way might be to implement this as authentication method 08:17:26 #info that device lock daemon would then use for unlocking (when user 08:17:28 #info wakes up the device). But given that daemon is closed source 08:17:30 #info monolith, it would need to be implemented in-house for foreseeable 08:17:32 #info future. And we’d need ability to grade authentication methods 08:17:34 #info somehow to differentiate between “can be used for unlocking device” 08:17:36 #info and “can be used to identify user” and so on. 08:18:36 any comments? 08:18:47 just reading.... 08:19:22 smart-unlock is a feature on android ... perhaps you could consider adding such a thing? 08:19:38 im quite happy to not have to make my own hack 08:19:55 just for the "connected device" use-case 08:20:56 let's see how this develops 08:21:05 It would be nice if the user could select how "hard" security they want/need on their devices, instead of having it hardwired by the operating system. 08:21:31 So the obvious downside as mentioned is potential privacy violations; either allowing criminals or the state-level to get access? 08:21:45 ExTechOp, your suggestion would mitigate that. 08:21:58 flypig: but if the uses opts to not-lock if their watch is in vicinityl then thats fine 08:22:19 #info It would be nice if the user could select how "hard" security they want/need on their devices, instead of having it hardwired by the operating system. 08:22:45 piggz, yes, understood. I guess I'd just personally not want it to weaken security protections when it's not used. 08:22:57 sure 08:23:05 people pay with cards at stores with nothing more than a swipe these days. not secure at all. 08:24:01 i wouldnt want to weaken security for all, but just add some convenience for some with little downside 08:24:01 I don't think that's a same analogy 08:24:31 I'm suggesting that pairing nfc, bluetooth are MORE secure. 08:24:33 that's all. 08:24:37 which kind of device are you using at the moment to unlock? 08:25:00 time flies... 08:25:14 nephros: my use case is basically ... if my watch is connected, which is on my person, then allow unlock automatically 08:25:32 let's continue this one the general discussion if we have time left. ok 08:25:34 piggz, ble? 08:25:48 happy enough for this to be considered, have to go to school now :) thx rainemak 08:25:48 Maybe it can be constrained to BT MAC + must be a HID device + must have a button + button press unlocks? 08:25:48 ... on the general 08:25:58 thanks piggz 08:26:04 #topic Inclusion of FolderListModel in harbour (5mins -- asked by poetaster) 08:26:09 #info I recently, somewhat belatedly, noticed that I had disqualified 08:26:09 #info a number of packages I already have in harbour by adding 08:26:09 #info FolderListModel. (the labs one). Is it possible to allow this? 08:26:09 #info Or am I missing an equally convenient mechanism. I have my 08:26:09 #info ‘own’ python based version but that’s a hack. 08:26:17 #info There isn’t really any good reason for not allowing 08:26:17 #info Qt.labs.folderlistmodel in Harbour. In fact, we were already 08:26:17 #info planning to allow it in 4.5.0 release, but unfortunately the change 08:26:17 #info just missed the window. It will be allowed when the next Sailfish 08:26:18 #info OS release comes out. 08:26:38 Oh, cool! Yeah! 08:26:45 Great news :) 08:26:51 this is should be clear and straightforward :-) 08:27:00 s/is// 08:27:19 short topic. not like me. 08:27:31 Is there a significant difference between the labs one, and the one from Silica (or is it Nemo)? 08:27:32 let's move on 08:27:43 #topic AAS System Webview (5mins -- asked by Matias) 08:27:55 #info SFOS AAS comes with Android System Webview version 83 which is 3.5 08:27:55 #info years old by now. This causes certain banking apps and potentially 08:27:55 #info other Android applications that rely on a more recent version of 08:27:55 #info webview to stop working and it is also a bit of a security risk. 08:27:55 #info Installing and activating a different webview implementation 08:27:55 #info requires superuser rights (infeasible) or signature spoofing 08:27:57 #info (which nobody seems to have tried in the context of webview yet). 08:27:59 #info With the reorganization over I was hoping someone could shed some 08:28:01 #info light on this. 08:28:03 #info 08:28:05 #info So, is there an update for AAS/Android System Webview somewhere 08:28:09 #info down the pipeline? 08:28:11 #info 08:28:13 #info Links/references: 08:28:15 #info 08:28:17 #info 1. Rabobank app stopped working 08:28:19 #link https://forum.sailfishos.org/t/rabobank-app-stopped-working/2898 08:28:21 #info 2. Old Android WebView (74.x) --> update needed, ideas? 08:28:23 #link https://forum.sailfishos.org/t/old-android-webview-74-x-update-needed-ideas/13465 08:28:25 #info 3. Make updating Android (System)Webview possible 2 08:28:27 #link https://forum.sailfishos.org/t/make-updating-android-system-webview-possible/15116 08:28:29 #info 4. GeckoView instead of Android System WebView · Issue #167 · mozilla/geckoview · GitHub 1 08:28:32 #link https://github.com/mozilla/geckoview/issues/167#issuecomment-1031423692 08:28:34 #info We are using Webview from current Android version and it is updated 08:28:36 #info always when we update AppSupport Android base version. We have 08:28:40 #info AppSupport update for next Android version in pipeline but not yet 08:28:42 #info sure when can we release that out. 08:28:44 sorry, nephros I just changed the topic... please raise your question again on the general discussion. 08:30:08 I see. Would it be possible to whitelist other webviews in the system image to allow the user to choose other ones if they so desire? Just an idea. 08:30:51 #info Would it be possible to whitelist other webviews in the system image to allow the user to choose other ones if they so desire? Just an idea. 08:31:37 Matias, I'll ask around 08:32:03 Thank you! It's greatly appreciated. 08:32:05 I doubt that I get answer for this meeting 08:32:21 let's speed up a bit 08:32:23 #topic Mobile data in Italy with Xperia 10 III not working (4.5.0.24) (5mins -- asked by Valorsoguerriero97) 08:32:33 #info Mobile data not working under WINDTRE (ITALY) on 10 08:32:33 #info III. Same APN setting on my 10 works ok. Not here. I 08:32:33 #info can only call and receive sms… not mobile data. What 08:32:33 #info about this bug? Can you give me an estimated time for 08:32:33 #info resolution? 08:32:45 #info You replied to the forum already with a solution. Good that this got sorted out. 08:32:46 #info Summary: Ask the operator to disable VoLTE on your contract/line. 08:32:46 #link https://forum.sailfishos.org/t/xperia-10-iii-no-mobile-data/17711/19 08:33:23 there operator specific issues are tricky 08:34:31 moving forward 08:34:39 #topic Jolla community dinner @ FOSDEM '24 (10mins -- asked by Louis) 08:34:44 #info There has been a Jolla community dinner after FOSDEM on Saturday in 08:34:45 #info the previous years at non-COVID times. I personally miss the 08:34:45 #info Moroccan restaurant La Kasbah (the host at 2019 and 2020) as I 08:34:45 #info found it quite suited? Anyone from Jolla up for the planning or 08:34:45 #info should it be done by time community. Also if anyone have better 08:34:45 #info suggestions I’m open for it. 08:35:12 #info We’re thinking that this would be more like Community event this 08:35:13 #info time. That said, it’s under planning. We’re still checking suitable 08:35:13 #info place, but it should be such that it encourages to discuss and serves 08:35:13 #info food. Central location it’ll be and likely scheduled to Saturday 08:35:13 #info 7:00 PM giving everybody enough time to move from FOSDEM venue. 08:35:13 #info 08:35:15 #info There’s also a forum topic about the dinner but no post yet 08:35:17 #info regarding this Community event. 08:35:19 #info 08:35:21 #link https://forum.sailfishos.org/t/any-plans-regarding-fosdem-2024-dinner/17632/5 08:35:50 I'm already looking forward to it! 08:36:19 flypig, good to hear. Just want to know if anyone from Jolla wants to plan it. 08:36:40 I could also step in if needed 08:36:57 LouisDK, we're planning it already 08:37:26 rainemak: Nice. Do you have a resturant in mind? 08:38:17 LouisDK, not yet selected. 08:38:56 We'll post to the forum when there's info to share 08:39:01 If I recall correctly, last year was Café Georgette, before that it was La Kasbah like you said LouisDK. 08:39:22 Quite different experiences :) 08:39:24 flypig: Yes. It's sww what rainemak comes up with. 08:40:12 Yes, indeed :) 08:40:49 Do you want to continue still or is it ok to move forward? 08:40:55 I shall hold the question until general, but I'm interested to know if anyone else here will be at FOSDEM. 08:41:13 rainemak: Just continue. 08:41:22 Trying save some time for general discussion 08:41:27 #topic Jollas future plans regarding Alien Dalvik (10mins -- asked by Louis) 08:41:30 #info What is the situation regarding Jolla and Alien Dalvik. Does Jolla 08:41:30 #info have “enough rights” to expand the development and offering of 08:41:31 #info Alien Dalvik to eg. community ports? (Eg. sell it under an “as-is” 08:41:31 #info agreement without any official support from Jolla) 08:41:42 #info Let’s correct the name first, it’s not Alien Dalvik but Android 08:41:42 #info AppSupport. Not quite sure what you trying to ask with “enough 08:41:42 #info rights” but yes, we have all rights to the AppSupport. This has 08:41:42 #info been discussed on several occasions in the past, see links below. 08:41:42 #info Proposed as-is is likely not a good idea with current Sailfish X 08:41:43 #info pricing model. 08:41:45 #info 08:41:47 #info It's worth noting that Android AppSupport for community ports is 08:41:49 #info more than just providing the relevant packages. No need to repeat 08:41:51 #info answers again, please check “Ability to buy/enable Android 08:41:53 #info AppSupport for community ports” topic from 2023-02-16 log and 08:41:55 #info “Standalone AppSupport license for ports” topic from the 2022-06-09 08:41:57 #info log. 08:41:59 #info 08:42:01 #info If we were to do this, it wouldn't be in the immediate future. 08:42:03 #info 08:42:05 #link https://forum.sailfishos.org/t/community-meeting-on-irc-16th-february-2023/14326 08:42:07 #link https://forum.sailfishos.org/t/community-meeting-on-irc-9th-june-2022/11704 08:42:11 #link https://irclogs.sailfishos.org/meetings/sailfishos-meeting/2023/sailfishos-meeting.2023-02-16-08.00.log.html 08:42:14 #link https://irclogs.sailfishos.org/meetings/sailfishos-meeting/2022/sailfishos-meeting.2022-06-09-07.01.log.html 08:42:36 rainemak: Answer accepted. Just move forward 08:42:48 :-) 08:42:48 #topic Mic issues under AOSP and SailfishOS (Xperia 10 ii + iii) (5mins -- asked by Louis) 08:42:55 #info 3rd party caller my be able to hear their own voice due to a bug in 08:42:55 #info AOSP. Any updates on this from Jolla? Echo during calls on Xperia 08:42:55 #info 10 II using Loudspeaker 1 and [LENA][11][4.19] Echo cancellation is 08:42:55 #info not enabled for voice calls · Issue #771 · 08:42:55 #info sonyxperiadev/bug_tracker · GitHub 08:42:57 #link https://forum.sailfishos.org/t/echo-during-calls-on-xperia-10-ii-using-loudspeaker/10990 08:42:58 #link https://github.com/sonyxperiadev/bug_tracker/issues/771 08:43:00 #info Not really updates. Officially we’re stating Android 11 as required 08:43:02 #info stock Android version and clearing Android 12 as ok. That said, 08:43:04 #info without guarantees you could try to update your Sony Xperia 10 III 08:43:06 #info to Android 13 before flashing Sailfish OS and check if it’s better 08:43:10 #info – please note that Android 11 binary blobs are still needed from 08:43:12 #info Sony’s web site. Bear in mind that if you upgrade stock Android 08:43:14 #info version to 13 before flashing Sailfish OS and there are other kind 08:43:16 #info of issues and you want to downgrade, you need to flash Android 12 08:43:18 #info with Windows and use Emma. 08:43:20 #link https://docs.sailfishos.org/Support/Supported_Devices/ 08:43:22 #link https://docs.sailfishos.org/Support/Help_Articles/Managing_Sailfish_OS/Reinstalling_Sailfish_OS/#reverting-xperia-back-to-android-os 08:44:37 moving on... 08:44:39 rainemak: The issues does still presists with Xperia 10 iii flashed with Android 12 and 13 prior to Sailfish. 08:45:05 Just move ahead 08:45:09 LouisDK, that's what we expected 08:45:18 #topic Open PR discussion (5 mins -- asked by Jolla) 08:45:18 #info Any open PRs to discuss? 08:45:45 I had none prepared for this... 08:46:42 Without Damien's update I'm lost :( 08:46:50 Let's skip open PRs for today... .-) 08:46:57 flypig, yeap 08:47:02 #info Bug reports by the community bug coordination team (5 min -- asked by Jolla) 08:47:07 #info The Community Bug Coordination Team have done a superb job once again this fortnight. 08:47:07 #info As a result of their work, we now have: 08:47:07 #info 6 high quality bug reports now recorded internally and tagged as "tracked". 08:47:07 #info 1 bug report missing “tracked” tag. 08:47:09 #info 3 marked as duplicates and closed. 08:47:36 Let's move on to the general discussion. 08:47:36 #topic General discussion (8 min) 08:48:00 Kudos to the bug team again! 08:48:07 flypig, you had some general topics... 08:48:21 flypig, Yeap! Once again 08:48:24 Oh, well, I'm curious to know who else here will be at FOSDEM. 08:48:33 I'll be 08:48:41 I will for sure. 08:48:52 The last time, I asked about SFOS future plans, has there been progress that could be made public? 08:49:50 I'd love to visit FOSDEM sometime, but probably not this year :( 08:49:51 ExTechOp, not really, you can trust that we share those when ready 08:50:21 I can't make FOSDEM either. Family, winter school break collide. 08:50:59 I'm looking forward to seeing you there LouisDK, rainemak :) Sorry to hear that though poetaster (I think you mentioned it previously, but it's still a shame). 08:52:07 flypig, ah, just checked. I'm wrong. 3-4 fossdem, 5-9 school free. hmmm. depends on what we plan for that week. 08:52:35 Well, if things change and you can make it, it will be wonderful to meet up :D 08:52:51 yes, I'd really like that! I never get out :) 08:52:52 Matias, got some input regarding the AAS WebView 08:53:08 poetaster: :D 08:53:44 #link https://chromium.googlesource.com/chromium/src/+/master/android_webview/docs/aosp-system-integration.md#Why-are-there-security-restrictions-on-which-apps-can-be-used-as-a-WebView-implementation 08:54:21 #info "Why are there security restrictions on which apps can be used as a WebView implementation?" ^ 08:55:14 let's schedule next meeting next... 08:55:18 #topic Next meeting time and date (5 mins) 08:55:21 So on a standard Android device, does that mean it's not possible to upgrade the Webview? 08:55:24 nephros, earlier you asked: "Is there a significant difference between the labs one, and the one from Silica (or is it Nemo)?" What is the labs one? 08:55:45 Sorry, I spilled over; you can ignore me. 08:56:07 https://doc.qt.io/qt-5/qml-qt-labs-folderlistmodel-folderlistmodel.html 08:56:13 It's possible to update the default Google system webview on regular Android devices, where it is whitelisted in the system image and available in the play store, I think. 08:56:31 Thank you rainemak! 08:56:41 Matias, yeap it could be possible via Play Store 08:56:48 Got it, thank you Matias! 08:56:59 Proposing Thursday 1st February at 08:00am UTC due to vacations so right before FOSDEM. 08:57:06 Thank you too for the extra info rainemak! 08:57:09 There is Nemo.FileManager::FileModel 08:57:13 flypig: Yeah, the one which rainemak linked is the labs one. There isn't any in Silica and we don't use the one from Nemo. 08:57:32 ok understood. 08:57:41 2024-02-01 works4me 08:57:46 +1 08:58:17 Thanks ViGe. 08:58:44 nephros, it's a matter of choice and both could be available for Harbour use 08:58:53 flypig, my better half says I should just go to fossdem :) 08:59:18 Yay! You have a very wise better half :) 08:59:26 :) 08:59:45 #info Next meeting will be held on Thursday 1st February 2023 at 08:00am UTC: 2024-02-01T0800Z 09:00:03 Thank you everyone for the meeting! 09:00:11 ok, so I wasn't aware of the Nemo FolderListModel ... research. 09:00:14 thanks all! 09:00:17 Thank you rainemak, all! 09:00:19 thanks! See ya around! 09:00:25 Thank you all! 09:00:35 ahoi! 09:00:43 First time, I was quite nervous :) 09:00:47 I think now we peeked number of topics... one less would be better, let's improve a bit. 09:01:06 Nice work Matias, did great ;) 09:01:10 Thank you, everybody! 09:01:20 Matias, don't be you did well! Welcome 09:01:30 #endmeeting