08:00:06 <ViGe> #startmeeting Sailfish OS, open source, collaboration -- 16th February 2023
08:00:06 <sailbot> Meeting started Thu Feb 16 08:00:06 2023 UTC. The chair is ViGe. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:06 <sailbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
08:00:11 <ViGe> #info Meeting information and agenda can be found here:
08:00:15 <ViGe> #link https://forum.sailfishos.org/t/community-meeting-on-irc-16th-february-2023/14326/
08:00:22 <ViGe> Hello everyone! As was mentioned in the end of previous meeting, I will be the chairperson today, and I'll do my best to keep time and order.
08:00:31 <ViGe> This is my first time as the chair, so please forgive me if I make any mistakes.
08:00:39 <ViGe> We again have quite a lot of questions, so please respect the timings!
08:00:47 <ViGe> #topic Brief introduction (5 mins). Please prefix your name/handle with #info
08:00:56 <ViGe> #info Ville Nummela - Sailor @ Jolla, chairperson today
08:01:11 <lolek> #info lolek - Just a regular user
08:01:17 <pherjung[m]> #info pherjung, community
08:01:18 <lolek> btw hi ViGe
08:01:32 <lolek> and generally hello everybody ;)
08:01:45 <b100dian> #info Vlad G., community
08:02:36 <ViGe> hi lolek, nice to see you here, you had quite a few questions :)
08:03:21 <lolek> I'm trying to do my best :D
08:05:04 <ViGe> Not too crowded here. 1-3 for the community if no one else shows up.
08:05:13 <Crabster> #info Crabster - Just a regular lurker
08:05:32 <ViGe> 1-4 it is, clear win for the community
08:05:41 <ViGe> #topic Decouple encryption password from phone password (5 mins -- asked by lolek)
08:05:46 <ViGe> #info <lolek> ok we finally have encryption that allow us to use alphanumeric values which is step forward in good direction.
08:05:49 <ViGe> #info <lolek> Now the only problem is that the same pw is needed twice to start up the phone.
08:05:53 <ViGe> #info <lolek> Imho this shouldn’t be required and some decoupling is required.
08:05:57 <ViGe> #info <lolek> I’d like to get information what’s exactly needed for this and if community can help with this?
08:06:02 <ViGe> Here's the prepared reply:
08:06:07 <ViGe> #info <Jolla> This is a broad topic carrying many details in multiple layers. Your humble opinion is about right.
08:06:11 <ViGe> #info <Jolla> Thank you for your feedback and input. Let’s see what we can do.
08:06:37 <ExTechOp> Sorry for being late, can I still submit an #info ?
08:07:08 <ViGe> ExTechOp: go ahead, it's not like it's the first time someone shows up late :)
08:07:37 <ExTechOp> #info Otto Mäkelä, community
08:07:43 <rainemak> #info rainemak - Sailor @ Jolla
08:08:43 <lolek> hmm ok so the answer is very fogy for me and tbh it says not much. I do understand that there are many things that this is related but in the question I asked if there's something the community can do to help with this effort or it's only something that Jolla must handle on their own?
08:10:08 <ViGe> lolek: My understanding of the matter is that currently there isn't really much that the community could do.
08:10:31 <ViGe> let's move on, many topics to go through today
08:10:34 <lolek> ok, so this is entirely on the Holla side.
08:10:38 <lolek> *Jolla
08:10:40 <ViGe> #topic Bootloader relock (5 mins -- asked by lolek)
08:10:46 <ViGe> #info <lolek> This is something that came out on the previous meeting.
08:10:50 <ViGe> #info <lolek> Thaodan said: if the bootloader supports user signatures you can relock it with your own key to accept only kernels e.g. from Jolla but Sony phones don’t support that so far.
08:10:55 <ViGe> #info <lolek> My question is: What Jolla can do to change this, and how community can help achieve this?
08:11:00 <ViGe> Here's the prepared reply:
08:11:05 <ViGe> #info <Jolla> There is nothing Jolla can do about this, as this is something that Sony has to provide.
08:11:41 <lolek> ok, so this is task for community to make a pressure on Sony then. Maybe Jolla could participate in that?
08:11:41 <ViGe> that was a short reply, and probably not what you wanted to hear, but it is what it is :/
08:11:53 <lolek> no worries ;)
08:12:05 <lolek> this topics are actually small ones and quick ones
08:12:09 <lolek> *these
08:12:44 <Thaodan> #info Björn Bidar - Sailor @ Jolla
08:12:46 <lolek> I would imagine that community can only "spam" sony with mails, but there has to be also someone else
08:12:54 <Thaodan> sorry a little to late
08:13:03 <lolek> Thaodan: good you're here, what you think about this? ^^ btw hi
08:13:06 <ExTechOp> Sony does not have a certificate signing system, unlike eg. some Linux distributions?
08:13:28 <lolek> well no idea, I'd see this like exception maybe only for Jolla?
08:13:43 <lolek> so Jolla would have their keys so they could sign off their official builds?
08:13:55 <lolek> I think that would be something for a starter
08:14:07 <lolek> is this doable?
08:14:34 <Thaodan> If I read correctly this independent of Jolla one pw you enter is a pw for luks disk encryption while other is for pam (the user).
08:14:59 <lolek> Thaodan: :D we're already on the bootloader relock ;)
08:15:18 <Thaodan> lolek: while that's not for Jolla to change
08:15:42 <ViGe> We should already move to the next topic
08:15:49 <Thaodan> Sony has to implement the bootloader extension to use custom bootloader keys, it is not just about relocking.
08:16:04 <ViGe> #topic Providing a better mechanism in the OS for installing additional authentication methods and letting users choose what they prefer. (10 mins -- asked by lolek)
08:16:05 <lolek> ok so we need to spam then
08:16:11 <ViGe> #info <lolek> this is a continuation of: https://forum.sailfishos.org/t/community-meeting-on-irc-31st-march-2022/10714/4
08:16:13 <Thaodan> No
08:16:17 <ViGe> #info <lolek> Also I asked there if Jolla can’t just license BB picture password since BB is gone.
08:16:21 <ViGe> #info <lolek> And for this part I’d like to also get an answer and I’m willing to wait for this one as it would probably require contacting with BB.
08:16:22 <Thaodan> just be polite
08:16:28 <ViGe> Here's the prepared reply:
08:16:32 <ViGe> #info <Jolla> Additional authentication methods could go quite nicely hand in hand with the first “Decouple encryption password from phone password”.
08:16:38 <ViGe> #info <Jolla> There are no immediate plans to implement better mechanisms for installing additional authentication methods.
08:16:46 <Thaodan> also it's not sony open devices change to make.
08:17:39 <lolek> Thaodan: ok, so whom?
08:17:58 <Thaodan> Sony the device manufacturer.
08:18:03 <lolek> regarding picture pw... I do understand that there's no reply regarding relicensing from BB?
08:19:25 <ViGe> lolek: The answer is above. There are no plans to implement additional methods of any kind.
08:19:44 <lolek> that's not good ;(
08:20:20 <lolek> well ok, carry on
08:20:29 <Thaodan> (I'm just explaining the decinical differences in case you want to push for this change your self and to understand.
08:20:31 <Thaodan> )
08:20:49 <lolek> Thaodan: yes, thank you!
08:22:31 <ViGe> Ok, well if there's nothing else to discuss about additional authentication methods, then perhaps we should move on then
08:22:59 <Thaodan> yes please
08:23:13 <lolek> yes, please do
08:23:27 <ViGe> the next question was asked by piggz, unfortunately I don't see him or rinigus here
08:23:43 <ViGe> but let's go through the question and prepared answer anyway
08:23:48 <ViGe> #topic Enabling VoLTE on community ports (20 mins -- asked by piggz)
08:23:54 <ViGe> #info <piggz> With 4.5 SFOS released, I wonder what is the state with VoLTE support. Sounds like Jolla’s implementation is closed source. Any plans to release it? Earlier, we were told that it is too early and no definite answer could be given. Maybe it is the time?
08:23:59 <ViGe> #info <piggz> I believe unofficially ported devices could benefit from looking into Jolla’s approach. Right now, I personally, even have no idea where to start digging. And, in principle, time used for duplicating efforts is not very efficiently spent.
08:24:03 <ViGe> #info <piggz> In addition, does keeping VoLTE closed give Jolla an advantage over competitors? most other mobile linux distros are using ModemManager, so this seems unlikely, opening up VoLTE would be a great boost to sailfish ports.
08:24:07 <ViGe> #info <piggz> If its not possible to open the code, could the compiled packages atleast be put into the adaptation-common repos so that porters can experiment? If there are other device-specific hacks required to enable volte, could these be documented to help porters understand what needs to be written? Would any sony specific packages also be useful on other sony ports?
08:24:13 <ViGe> Here's the prepared reply:
08:24:16 <ViGe> #info <Jolla> As we communicated earlier, there are device-specific parts of the implementation that we may not be able to share publicly.
08:24:21 <ViGe> #info <Jolla> Thank you for your input. We’ll consider and think what we could do to help on the VoLTE front.
08:25:18 <lolek> "we may not be able to share publicly." that dreaded NDAs... :/
08:25:42 <b100dian> even closed source, a list of chipsets supported that would allow porters to enable VoLTE on those specific SoC would be useful
08:26:08 <lolek> b100dian I think there's probably something more.
08:26:56 <b100dian> well, the closed packages are not available in a public OBS repo for one
08:30:10 <ViGe> as neither piggz or rinigus are here, I suggest we move on
08:31:23 <ViGe> unless Thaodan wants to add something, as you clearly know more about the topic than the rest of us? :)
08:31:38 <Thaodan> I think it depends on what we are talking about ringus device is similar to Jolla ports than lets say piggz.
08:32:48 <Thaodan> Similar to my (personal) ports but a little older
08:33:03 <b100dian> rinigus indeed has sony tama devices, would the fact that it is Sony help more than just being a specific Qcom chip?
08:33:42 <Thaodan> b100dian: Somewhat yes as they use the same Qualcomm blobs.
08:34:35 <b100dian> so is there a possibility to add the existing packages to an 'unsupported' repo so porters can test?
08:34:37 <Thaodan> blob versions between vendors might be different and vendor specific quirks that would require testing from the porter could be different too.
08:39:17 <ViGe> Let's move on
08:40:00 <ViGe> #topic Qt Upgrade (30 mins -- asked by piggz)
08:40:06 <ViGe> #info <piggz> The time has come again to ask about the potential for a Qt Upgrade in Sailfiish.
08:40:10 <ViGe> #info <piggz> Qt 5.6 was released almost 7 years ago, and security support ended in 2019
08:40:14 <ViGe> #info <piggz> Newer Qt versions not only add useful feature, but bring improved compiler support and support for newer c++ features.
08:40:18 <ViGe> Here's the prepared reply:
08:40:22 <ViGe> #info <Jolla> Qt Upgrade is not on our plans for the near future.
08:41:23 <ViGe> That was all of it - I somehow feel we are not going to spend the whole 30 minutes discussing this
08:41:54 <ViGe> especially since neither piggz or rinigus are here
08:42:09 <ExTechOp> I assume upgrading Qt versions is a major chore which easily brings up new bugs?
08:42:11 <b100dian[m]> oh, that's not good news
08:45:23 <ViGe> ExTechOp: I believe you are right
08:45:46 <Thaodan> I don't think anyone can add anything that hasn't been said already, while I think anyone would like the feature we can't progress it by talking more about it here right now.
08:46:03 <lolek> Well from my point of view I think the bigger problem is licensing
08:46:20 <lolek> anyway I'm fine to continue
08:46:41 <ViGe> Thaodan: yup, this topic has been covered in these meetings a few times already
08:47:26 <ViGe> Unless anyone objects let's move on
08:47:43 <ViGe> #topic ability to buy/enable Android AppSupport for community ports (10 mins -- asked by vlagged[m]
08:47:48 <ViGe> #info <vlagged[m]> Compared to the alternatives (anbox of old or waydroid now) Jolla’s AAS is specifically more integrated with SailfishOS and makes “selling” the idea to switch to Sailfish more easy for anyone.
08:47:53 <ViGe> #info <vlagged[m]> Adding the ability to pick any device that has a community port can help with people’s different hardware needs and expectations while still giving a great experience for Android apps. This would also mean that the work for the tight integration with Sailifish OS pays off even more.
08:47:59 <ViGe> #info <vlagged[m]> Since this is not something that any user could make an assessment of, I would suggest having a similar step to the one that is already in place for enabling Jolla Store access, which is that the porter confirms that the device is working with AAS. What is enabled then is the ability to buy the license that would activate the download for the user that bought it.
08:48:05 <ViGe> #info <vlagged[m]> For the porter to be able to make this assessment a minimum requirement document would be needed, specifying the e.g. min kernel version, kernel params etc. Waydroid could be used as a similar in spirit “proof of work” for the porter.
08:48:09 <ViGe> #info <vlagged[m]> Since adding this support may or may not succeed, it would probably mean that the porter would have to have access to give it a try (assuming ‘proof of work’) without costs other than having a set-by-Jolla minimal feature parity (e.g. audio should work, you name it) for the port.
08:48:14 <ViGe> #info <vlagged[m]> To be able to add this kind of access a possibility would be, when adding the store support for a ported device, to request a “porting enabling” license for the specific device the porter uses for testing so that he can also use the store to install Android AppSupport and debug it.
08:48:19 <ViGe> #info <vlagged[m]> For debugging, the same level of support as of today for community ports should be enough. If the porting fails, it would be nothing (but time) lost.
08:48:22 <ViGe> #info <vlagged[m]> If it succeeds, there could be a bunch of new customers for SailfishOS as a platform and for the AppSupport product for each port. And we get more friends to move over here:)
08:48:26 <ViGe> Here's the prepared reply:
08:48:31 <ViGe> #info <Jolla> Thank you for your feedback and thinking. As earlier stated, we have been looking at the possibility of stand-alone AppSupport for community ports
08:48:40 <ViGe> #link https://forum.sailfishos.org/t/community-meeting-on-irc-9th-june-2022/11704/20
08:48:47 <ViGe> #info <Jolla> It's worth reminding that AppSupport for community ports is more than just providing the relevant packages.
08:48:54 <ViGe> #info <Jolla> We'd also need processes for verifying that everything works, ensuring the right apps are made available in the Store, ensuring that ports are kept up-to-date and also implementing technical changes to the repository backend.
08:50:40 <ViGe> vlagged[m] is not here, but b100dian is and can substitute?
08:50:45 <b100dian> Yes
08:51:19 <b100dian> Thanks for your respons, I understand this is something that is being worked on
08:51:54 <b100dian> While backend changes may take a while, what is the plan to enable debugging of AppSupport on those ported devices? Would that still require the backend changes?
08:52:49 <ViGe> I'm not sure what you refer to with debugging?
08:53:25 <b100dian> Well to be able to say that AppSupport works on a device, you need to be able to install to look at the logs etc
08:53:39 <piggz[m]> Hi, managed to connect via sailtrix !
08:54:01 <ViGe> piggz[m]: welcome! Unfortunately you missed your own topics :(
08:54:22 <piggz[m]> Yeah, ill read the log latet :)
08:54:37 <Thaodan> You can add something in the general discussion maybe
08:54:46 <piggz[m]> I dont suppose there was much good news :D
08:55:44 <ViGe> b100dian: right. That is indeed a good question, and I think at this point it's fair to say that suggestions are welcome
08:56:23 <b100dian> Maybe something similar to the closed beta program? I only know there is one, not the details
08:59:08 <ViGe> maybe
08:59:34 <ViGe> ok, time to move on to the next topic, again one from lolek :)
08:59:45 <ViGe> #topic: Moving away from GitHub (10 mins -- asked by lolek)
08:59:55 <ViGe> #info <lolek> Since Microsoft took over GitHub they started to use OSS projects to train their CoPilot product on others code and is making money out of it.
08:59:58 <ViGe> #info <lolek> The Software Freedom Conservancy organization calls OSS projects to move away from GitHub.
09:00:01 <ViGe> #info <lolek> I’d like to know if Jolla is aware of this and are there any plans in the future to move away from that platform to somewhere else?
09:00:04 <ViGe> #link https://en.deepweb.net/blog/open-source-developers-urged-to-move-away-from-github-after-copilot-launch
09:00:07 <ViGe> Here's the prepared reply:
09:00:10 <ViGe> #info <Jolla> Jolla is aware of this, however we don’t have plans for moving away from GitHub.
09:00:32 <lolek> well this is very very bad info
09:00:48 <lolek> so I vote for making plans for it :d
09:01:11 <Thaodan> Part of the issue is that there nothing really that exactly compares to it.
09:01:58 <Thaodan> Customers requirements can always change a decision on the long run for any company in my opinion.
09:04:05 <b100dian> And maybe codeberg makes up in perf/stability in the next years
09:06:51 <lolek> Thaodan: well host this on your own? I think gitlab allows this. Jolla has servers if I'm not wrong so thats hould be doable: git.jolla.com ?
09:07:29 <ViGe> lolek: We used to do that in not too distant history, there's a reason why we moved to github
09:08:14 <lolek> well right now it means you've stumbled into mondern vendor lock in which is cloud lock in
09:08:56 <b100dian> I think this is a case of commoditizing services, maybe git hosting is cheaper to source out than to do it yourself, and yes, may be a honeypot lock in
09:09:41 <ExTechOp> This lock-in of course has been the Microsoft strategy already for quite some time.
09:10:06 <Thaodan> It depends on what costs more Github isn't enterily free either. Convenience vs. long term costs.
09:10:32 <lolek> b100dian well i do undretsand this but right now it seems that you're paying for something and what's worse the company that devlier this to you not even charge you for this but also make you a worker for them
09:11:00 <ViGe> 10 minutes are up, but we saved quite a bit of time on previous topics so I think we can continue a few minutes on this one
09:12:14 <b100dian> It may get to the point where people would refuse create github accounts. So in that case their possible contributions would stop before they even get a chance.
09:12:20 <lolek> so making this simpler words, you're paying github to be able to work for them :D
09:12:38 <lolek> yep... this is why I'm not doing any pull requests and so on.
09:12:49 <lolek> and I don't have account there
09:13:39 <b100dian> I think this is exactly a "long term costs" example
09:14:15 <lolek> so maybe Jolla could rethink creating new place on forum where we could just upload our patches instead of GH? I know, there's a problem what if the patch has a bug and so on. Well then it would be on the Jolla side and everyone submitting the patch would need to agree to give away all the ownership rights to that.
09:14:24 <lolek> I'm fine with this.
09:15:11 <lolek> in the end you can alwasy do signed-off and that's it.. the author info will stay and maybe could stay but the code ownership will belong to Jolla or be transfered to proper license
09:16:09 <ViGe> I think it's now time to move on
09:16:33 <ViGe> somewhat related topic
09:16:36 <ViGe> #topic Open PR discussion (5 mins -- asked by jolla)
09:16:41 <ViGe> #info <jolla> Any open PRs to discuss?
09:21:45 <ViGe> looks like the answer is "no"
09:21:51 <ViGe> #topic Untracked bug reports (5 mins -- asked by pherjung)
09:21:57 <ViGe> #info <pherjung> Untracked bug reports... (see the forum)
09:22:03 <ViGe> #info <Jolla> Thank you again for all of your work in checking and collating bug reports.
09:22:07 <ViGe> #info <Jolla> Here are the results:
09:22:11 <ViGe> #info <Jolla> - 6 high quality bug reports now recorded internally and tagged as "tracked".
09:22:14 <ViGe> #info <Jolla> - 1 duplicate bug report closed
09:24:12 <pherjung[m]> Is the process still correct? Or can I increase the amount of bugs report?
09:24:46 <ViGe> Please don't. I think the number of bugs is about right.
09:26:43 <ViGe> #topic General discussion (10 min)
09:28:08 <ExTechOp> About SFOS 4.5.0.16 (Struven ketju): I've found that it does work better with wifi/mobile network switching, and it somehow also feels a bit more responsive than earlier releases. Good work!
09:28:35 <ViGe> ExTechOp: That's good to hear, thanks!
09:28:38 <rainemak> ExTechOp, thank you!
09:30:27 <rinigus> Re volte: missed discussion, but its expected at this time. I'll try to formulate questions in forum and lets see if we can get help that way
09:31:10 <rinigus> Re qt: thanks for straightforward reply
09:31:26 <ViGe> rinigus: sounds good, thanks
09:34:23 <ViGe> Ok, time to start wrapping up
09:34:55 <ViGe> Thanks everyone who joined, I hope you don't think I made a huge mess
09:35:01 <ExTechOp> It however feels like the bug/feature of losing alarm tones (phone only does vibration) after a few days of uptime is still there.
09:35:01 <ExTechOp> It's a bit problematic to debug, since it only randomly turns up with a lengthy delay. Am I the only one keeping the phone on without rebooting for days and days at a time?
09:35:56 <rainemak> Thank you ViGe !
09:36:07 <ViGe> ExTechOp: You most definitely are not the only one, but unfortunately I believe most of us reboot quite often as we are using our devices for development
09:36:31 <ViGe> #topic Next meeting time and date (5 min)
09:36:41 <ViGe> Proposing Thursday 9th March, 8:00 UTC
09:36:51 <ViGe> That would be three weeks from now
09:37:02 <ExTechOp> Due to school holidays?
09:37:09 <ViGe> yes
09:38:34 <lolek> ExTechOp: ViGe: I do agree that maybe it's placebo but the WiFi searching, switching is more responsive on Xperia 10.
09:39:46 <pherjung[m]> ExTechOp: Indeed I just had this bug yesterday
09:40:14 <lolek> ExTechOp: my current uptime is 6 days I don't reboot the device expect I forgot to charge it :D
09:40:20 <ViGe> I don't hear any objections, so...
09:40:24 <ViGe> #info Next meeting will be held on Thursday 9th March, 8:00 UTC
09:41:09 <ExTechOp> I think there was a thread about the sound issue on the forum, could someone in the know add instructions on what kind of information would be useful for developers when this happens?
09:41:14 <lolek> ExTechOp: the only prob is that I have constantly sounds turned off and only vibration :d
09:41:27 <ViGe> So, thanks again everyone, see you again in three weeks!
09:41:38 <ExTechOp> Thanks everyone!
09:41:42 <ViGe> ExTechOp: If there isn't, please file a bug report in the forum
09:41:54 <ViGe> #endmeeting