07:00:40 <flypig> #startmeeting Sailfish OS, open source, collaboration -- 18th August 2022
07:00:40 <sailbot> Meeting started Thu Aug 18 07:00:40 2022 UTC. The chair is flypig. Information about MeetBot at http://wiki.debian.org/MeetBot.
07:00:40 <sailbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
07:00:47 <flypig> #info Meeting information and agenda can be found here:
07:00:51 <flypig> #link https://forum.sailfishos.org/t/community-meeting-on-irc-18th-august-2022/12549
07:00:57 <flypig> I am the meeting's chairperson today, and will be doing my best to keep time and order. Please respect the timings and no unauthorised use of the flux capacitor.
07:01:02 <flypig> #topic Brief introduction (5 min). Please prefix your name/handle with #info
07:01:16 <flypig> #info David Llewellyn-Jones - sailor @ Jolla
07:01:32 <thilo[m]> #info thigg - community dev
07:02:15 <flypig> I'm glad you're here thilo[m], for a second I thought I might be on my own!
07:02:18 <thilo[m]> Good $timeOfDay
07:02:23 <flypig> :)
07:02:24 <JacekJagosz[m]> #info Jacek Jagosz - long time SFOS user, maintainer for linux distribution Solus
07:02:28 <Sailor965_> #ingo pherjung - community
07:02:40 <Thaodan> #info Björn Bidar  - Sailor @ Jolla
07:02:58 <sledges> #info Simonas Leleiva - privateer for Jolla, broke speed limit only once (when started off), reading said 88mph
07:03:14 <thilo[m]> :D i didn't want to miss the oppertunity to test hydrogen on my phone ;)
07:03:14 <ViGe> #info Ville Nummela - Sailor @ Jolla
07:04:04 <flypig> thilo[m], not a bad reason at all :)
07:04:05 <thilo[m]> I mean, I maintain this app but mostly use element
07:04:20 <Chris[m]123> #info ahappyhuman ­— community
07:04:25 <sebix[m]> #info sebix - community
07:04:34 <flypig> Oh, we have a good number today, really great!
07:05:24 <flypig> Can someone figure out the score?
07:05:52 <flypig> While we get started with the questions. We actually first have a very quick update on a question from the last meeting.
07:05:55 <flypig> #topic Harbor API progress (in special QtLocation) (2 mins -- asked by fridlmue)
07:06:00 <flypig> #info This is a quick update on fridlmue's topic from the last meeting, since we received a few more details.
07:06:08 <flypig> #info <Jolla> We are working to add as many APIs to Harbour as possible in the 4.5.0 release, but we aren't promising anything in particular.
07:06:15 <flypig> #info <Jolla> It's highly likely that at least Sailfish Secrets will be allowed. We don't yet have any plans we can share with regards to QtLocation, so the previous advice about this stands.
07:06:21 <thilo[m]> 3 vs 3
07:06:32 <flypig> Thanks thilo[m] :) Score draw.
07:07:00 <flypig> That was just an update since there was new info to share, we can go straight on to the first question from JacekJagosz[m].
07:07:13 <flypig> #topic Specify which is the officially supported Android 11 version ID for X10 III (10 minutes -- asked by Jacek Jagosz)
07:07:21 <flypig> #info <Jacek Jagosz> In my opinion the instructions are unclear which version ID users should be on when installing SFOS, and there is a ton of conflicting information.
07:07:26 <flypig> #info <Jacek Jagosz> Install guide mentions just Android 11.
07:07:30 <flypig> #link https://jolla.com/how-to-install-sailfish-x-on-xperia-10-iii-on-windows/
07:07:33 <flypig> #info <Jacek Jagosz> Flashing guide mentions 62.0.A.3.109.
07:07:37 <flypig> #link https://jolla.zendesk.com/hc/en-us/articles/115004283713-Reverting-Xperia-device-to-Android-OS-and-reinstalling-Sailfish-OS
07:07:41 <flypig> #info <Jacek Jagosz> Then users have been recommending updating to the newest possible A11 version - 62.0.A.9.11, first flashing using EMMA and then using OTA.
07:07:45 <flypig> #link https://forum.sailfishos.org/t/sony-xperia-iii-android-12/11029/101
07:07:51 <flypig> #info <Jacek Jagosz> You can even flash directly to that version using different tools and firmware provided by someone. Oh, and some people flashed from Android 12...
07:08:02 <flypig> #info <Jacek Jagosz> Could you please state the version number you think is the least problematic, or most well tested? Now you have people using 3-4 different versions, and it definitely makes bug tracking really complicated.
07:08:18 <flypig> That's the question, here's the prepared answer.
07:08:23 <flypig> #info <Jolla> Thanks for your question Jacek. The situation has been changing and we appreciate this can make the most recent advice hard to pin down.
07:08:27 <flypig> #info <Jolla> The Sailfish OS development for the Xperia 10 III was mostly undertaken on Android 11. We have subsequently also tested it on Android 12.
07:08:39 <flypig> #info <Jolla> Our recommendation is to upgrade your phone to the latest version of Android 11 before installing Sailfish OS. However, if you have already upgraded to Android 12, then we don't recommend reflashing back to Android 11 before installing Sailfish OS.
07:08:46 <flypig> #info <Jolla> If you have suggestions for how to make the installation guide clearer then we would welcome the input.
07:08:59 <flypig> Would you like to add anything JacekJagosz[m]?
07:09:35 <JacekJagosz[m]> Nowhere in Jolla's guide it was mentioned that it should be the latest Android 11
07:10:03 <JacekJagosz[m]> Simply Android 11. So it would be great if this one sentence could be added to the official guide
07:10:39 <flypig> Yes, that's a simple change which makes things clearer.
07:10:46 <JacekJagosz[m]> Here nothing about updating is mentioned: write /sys/module/lowmemorykiller/parameters/minfree "15360,19200,23040,26880,34415,43737"
07:11:07 <JacekJagosz[m]> Sorry, wrong paste: https://jolla.com/how-to-install-sailfish-x-on-xperia-10-iii-on-windows/
07:11:45 <JacekJagosz[m]> And here in flashing guide there is a specific release mentioned, but it is not the newest, but 3 releases from the newest: https://jolla.zendesk.com/hc/en-us/articles/115004283713-Reverting-Xperia-device-to-Android-OS-and-reinstalling-Sailfish-OS
07:12:31 <JacekJagosz[m]> I suggest on both sites one sentence could be added, to use OTA to update to newest Android 11 version, that is all I am asking
07:12:52 <flypig> It does state that "anything after the letter A may vary and does not matter in this context"
07:13:08 <Thaodan> I agree.
07:13:39 <Thaodan> But things like this can happen to anyone, sometimes we miss details :)
07:13:52 <JacekJagosz[m]> As a user I updated to the number that was exactly stated, if it didn't matter I thought I can stay on older version
07:14:14 <JacekJagosz[m]> I am just suggesting to make things clearer
07:14:32 <flypig> That's also true. We don't ask users to reflash their devices every time Sony release a new update, so older versions are fine too.
07:15:03 <rainemak> #info Raine Mäkeläinen - Sailor @ Jolla
07:15:10 <flypig> But that's understood JacekJagosz[m], we'll discuss this internally and then perhaps I can message you on the forum for your feedback?
07:15:24 <flypig> We want it to be as clear as possible.
07:15:51 <JacekJagosz[m]> flypig: Thank you a lot
07:15:53 <Thaodan> JacekJagosz[m]: Do you think the sentence above the table is not clear enough?
07:16:10 <ExTechOp> (pardon for late arrival, #info Otto Mäkelä, community)
07:16:47 <Thaodan> In cases were the minor version matters we specify it on the table with an asterisk
07:17:09 <Thaodan> ExTechOp: You have to prefix your message with #info
07:17:41 <JacekJagosz[m]> Thaodan: The sentence is clear, I read it when first installing. I would just advice to also add a sentence where you suggest newest possible version of Android 11
07:17:41 <ExTechOp> Thaodan I wasn't sure would my #info become a part of the current topic or the roll call
07:18:01 <flypig> For the minutes:
07:18:02 <flypig> #info <JacekJagosz[m]> "I suggest on both sites one sentence could be added, to use OTA to update to newest Android 11 version, that is all I am asking"
07:18:29 <flypig> 1 more minute on this topic.
07:19:13 <Thaodan> Ja\
07:19:25 <flypig> JacekJagosz[m], are we okay to move on to the next question?
07:19:42 <Thaodan> JacekJagosz[m]: As said it doesn't matter usually and if it matters it's stated.
07:20:56 <JacekJagosz[m]> Yes, all done from my side
07:21:01 <flypig> Thansk JacekJagosz[m].
07:21:13 <flypig> #topic Is there a general cause to the various different mobile-data bugs with 10 III devices? (10m -- asked by PeegeeTips)
07:21:22 <flypig> #info <PeegeeTips> There are several bugs on the forum that look like they might be related.
07:21:29 <flypig> #info <PeegeeTips> 1. Xperia 10 iii can't enable mobile data on Talkmobile (UK):
07:21:34 <flypig> #link https://forum.sailfishos.org/t/xperia-10-iii-cant-enable-mobile-data-on-talkmobile-use-the-same-settings-used-on-my-xa2-plus/12485
07:21:38 <flypig> #info <PeegeeTips> 2. No IPv4 mobile data connection with MNO Telekom Deutschland possible:
07:21:44 <flypig> #link https://forum.sailfishos.org/t/ipv4-mobile-data-connection-not-possible/11917
07:21:48 <flypig> #info <PeegeeTips> 3. Mobile data (IPv6) connection used even though a WLAN connection (IPv4) is active:
07:21:53 <flypig> #link https://forum.sailfishos.org/t/use-of-mobile-data-ipv6-although-a-wlan-connection-ipv4-is-active/11918
07:22:02 <flypig> Here is our prepared answer:
07:22:08 <flypig> #info <Jolla> There have been a number of changes to the networking stack recently that could be related to the first two. Some users were experiencing mobile data connections when the data access point protocol was set to Dual:
07:22:14 <flypig> #link https://forum.sailfishos.org/t/4-2-0-19-xperia-10ii-no-internet-connection-on-mobile-data-when-leaving-wifi-if-data-access-point-protocol-is-set-to-dual-default/6697
07:22:19 <flypig> #info <Jolla> This was carrier specific, with a fix being rolled out in Sailfish OS 4.4.0.64:
07:22:30 <flypig> #link https://github.com/sailfishos/connman/pull/8
07:22:35 <flypig> #info <Jolla> For issue 3, a fix was introduced to work around the fact that VoLTE can create new IPv6 interfaces with potentially broken routing data (this is carrier dependent). This fix was rolled out in Sailfish OS 4.4.0.68:
07:22:44 <flypig> #link https://github.com/sailfishos/connman/pull/29
07:22:50 <flypig> #info <Jolla> Unfortunately it’s impossible to say whether the three are related without further investigation. We already had an internal issue lodged for issue 3 which has been looked into, and we also now have internal issues for the other two as well.
07:23:01 <flypig> #info <Jolla> The discussion in the forum is already quite comprehensive and very helpful, but as always any further detail you can add there is also appreciated.
07:23:22 <flypig> Is PeegeeTips here today? Any other comments welcome.
07:24:37 <ExTechOp> I believe this is kinda related to the long-standing issue of SFOS being quite bad at switching networks (mobile/wifi/mobile), specially for Android clients
07:26:56 <flypig> Yeah. Over the years we've squashed many bugs related to this (both for Android App Support and SFOS generally), but issues do arise.
07:28:37 <sledges> suggestion to use Ofono Logger when mobile data cannot be enabled at all (regardless of wlan state etc) and send logs (app has disclaimer that some information may be private, such as phone numbers, but emailing it straight to monich has been revoked due to in-app sharing broken)
07:30:41 <flypig> sledges, just to clarify, what should users do with the logs then?
07:31:32 <flypig> And would you add a comment on this to the minutes please sledges?
07:32:16 <flypig> 2 more mins on this topic.
07:33:25 <sledges> ensure there are no phone numbers in the logs (simply do not make/receive phonecalls during log grab); I believe sending them to jolla customer care is now the best bet, where data protection is ensured
07:33:42 <flypig> Great, thank you sledges.
07:34:24 <sledges> #info for mobile data issues please grab logs via Ofono Logger:
07:34:26 <sledges> #link https://openrepos.net/content/slava/ofono-logger
07:34:58 <flypig> Perfect, thank you.
07:34:58 <sledges> #info ensure no phonecalls made/received during the session, then provide the tarball to jolla customer care
07:35:24 <sledges> sadly logger is no longer in Jolla Store, i bet it's sandboxing/in-app sharing issues etc
07:36:11 <Thaodan> #info Ofono Logger is also available in chum.
07:36:15 <Thaodan> #link https://build.merproject.org/package/show/sailfishos:chum/openrepos-logger-ofono
07:37:15 <flypig> Okay, thanks for the helpful info. It's a shame PeegeeTips isn't here to comment, but hopefully this will be helpful to them.
07:37:24 <flypig> We hit time, so next question.
07:37:33 <flypig> #topic Open PR discussion (10 mins -- asked by jolla)
07:38:08 <flypig> #info <Mohjive> Open PR URL: https://github.com/mer-hybris/hybris-initrd/pull/27
07:38:33 <flypig> We had this submitted. Is Mohjive here to give some context?
07:39:20 <flypig> Thaodan, sledges, you are the main protagonists on this PR. Could you give us some info on the situation with it please?
07:40:17 <sledges> technical debt, we would need to update repos+re-test of all devices for any breakages
07:40:42 <sledges> if there is capacity to do integration, this change is more than welcome
07:41:06 <Thaodan> I agree with sledges however it should not be needed anymore for recent devices having a recovery partition.
07:41:26 <flypig> It's Xperia 10 specific?
07:41:48 <Thaodan> All Sony devices starting with Xperia 1/Xperia 10 II have a recovery partition again, however we didn't test that far enough so far.
07:42:02 <Thaodan> I did some tests but didn't come far enough.
07:42:37 <Thaodan> It is still work to work on embedded recovery for those devices that don't support a recovery partition.
07:42:49 <flypig> Is there any way the community could help with that testing?
07:42:56 <dr_Gogeta86_web> hi
07:43:05 <flypig> Welcome dr_Gogeta86_web.
07:43:10 <dr_Gogeta86_web> sorry to be a bit late in the party
07:43:14 <dr_Gogeta86_web> *at the
07:43:29 <flypig> Could you #info yourself for the minutes please dr_Gogeta86_web?
07:43:29 <Thaodan> Initrd could use some rework to use systemd earlier or allow for luks on system partion.
07:43:47 <dr_Gogeta86_web> #info dr_gogeta86 retired porter ... avid SFOS user
07:43:53 <flypig> Thanks!
07:44:33 <Thaodan> Flypig: If a community member is a porter or wants to delve into it they can test it by building droid-hal and droid-hal-img-boot for their device with the PR applied.
07:45:03 <flypig> Thanks Thaodan. I think that could be something for the minutes too, please.
07:45:19 <Thaodan> #info Flypig: If a community member is a porter or wants to delve into it they can test it by building droid-hal and droid-hal-img-boot for their device with the PR applied.
07:45:41 <mohjive> hello, see that I joined right on time (or just a little bit late) :)
07:45:59 <Thaodan> #info To enable embedded recovery this patch has to applied (or recreated) for the specific devices: https://github.com/Thaodan/droid-hal-img-boot-sony-nile/commit/f217376f663cbc82b501972b7197b24b0453a457
07:46:01 <flypig> Ah, good that you're here mohjive (if you could #info yourself for the minutes, that'd be great too).
07:46:19 <Thaodan> #info For the Xperia XA2 the patch can simply be rebased.
07:47:00 <Thaodan> #Info I'll write a summary under the hybris-initrd PR on how this PR can be used, what has to be done.
07:47:14 <flypig> Thanks Thaodan, I think that'd be helpful.
07:47:32 <mohjive> #info Mohjive - long time Sailfish user, Linux nerd
07:47:39 <flypig> Thanks mohjive!
07:47:49 <flypig> We have a couple more mins on open PRs. mohjive would you like to add anything? Or if anyone has other PRs they'd like to raise?
07:48:36 <dr_Gogeta86> I've finally acquired a Sony 10ii ... but is one of them with broken/corrupted TA partition ... which result in missing audio ...
07:50:05 <dr_Gogeta86> the strange thing works with stock android ( despite the Calibration on TA is 0x00 0x00 0x00 ) ... but on sailfish broke pulseaudio and droid driver ... happened few times so far on also on forum ... but nobody investigated further and reflashed on android
07:50:14 <mohjive> @flypig, As you know I stumbled on broken recovery mode, and I was hoping that getting this PR in would stabilize the development
07:51:04 <mohjive> it was in this context I found the PR
07:51:51 <Chris[m]123> I was wondering if anyone had tested https://github.com/sailfishos/qtbase/pull/11
07:52:09 <Thaodan> mohjive: Broken how? We have to separate issues, one is the need to flash recovery and the other is usb-network not working, both are unrelated!
07:52:27 <flypig> #info The context of mohjive raising this PR:
07:52:29 <flypig> #link https://forum.sailfishos.org/t/xperia-10-ii-unable-to-reach-recovery-mode/11822
07:52:38 <mohjive> @Thaodan I cannot boot the recovery image after flashing it
07:54:41 <flypig> Is that your PR Chris[m]123?
07:55:27 <Chris[m]123> Yes, it is indeed. While nothing seems to crash, it seems to misbehave when apps are launched from the home screen with the app booster
07:55:28 <flypig> #info https://github.com/sailfishos/qtbase/pull/11
07:55:57 <flypig> I'll add another 5 minutes to this topic for discussion.
07:55:58 <Thaodan> mohjive: the hybris-initrd PR is unrelated to your issue.
07:56:29 <dr_Gogeta86> flypig, can I continue to explain the problem ?
07:56:57 <flypig> dr_Gogeta86, sorry, I should have mentioned, but would you mind pausing until the General discussion?
07:57:23 <dr_Gogeta86> ok
07:57:39 <flypig> Thanks... not too long to wait I think.
07:57:49 <mohjive> Thaodan: ok, that's unfortunate (for me)
07:58:24 <flypig> Chris[m]123, I'd need to look through our bug tracker, but I think probably we've not had a chance to test it.
07:59:39 <Chris[m]123> So for me the pull request at least doesn't crash and my small test suite that I wrote works, but only if apps are launched from the command line and not through the home screen or with the booster. I added that as a comment to the PR a few minutes ago, because it feels like the PR shouldn't be merged in this state.
08:00:28 <flypig> That's useful info, thank you Chris[m]123. I did a cursory check and didn't find anything on our bug tracker, so I'll create an issue for it.
08:01:36 <Thaodan> #info I added an explanation on how the embedded recovery for hybris-initrd  should be used here:
08:01:36 <Chris[m]123> I'm also wondering if other app developers have encountered the issue with the SSL deadlock that I was encountering, as that was the motivation for creating this PR.
08:01:38 <Thaodan> #link https://github.com/mer-hybris/hybris-initrd/pull/27#issuecomment-1219153929
08:02:02 <flypig> #info Chris[m]123 has tested the qtnetwork PR and not found issues with it, so it should be in a good state to merge.
08:02:14 <Chris[m]123> Wait, that's wrong
08:02:21 <flypig> #undo
08:02:21 <sailbot> Removing item from minutes: <MeetBot.items.Info object at 0x7f23b9d976d8>
08:02:45 <Chris[m]123> It has issues with the booster, but it works when launching apps without the booster
08:03:04 <flypig> Ah, I misunderstood. Would you formulate something for the minutes please?
08:03:29 <Chris[m]123> When launching through the booster, it stops reporting any SSL errors which is problematic.
08:03:52 <Chris[m]123> I'll try formulating something for the minutes, please give me a minute ;)
08:04:01 <flypig> Sure, no worries. Thank you.
08:06:23 <Chris[m]123> Chris[m]123 has tested the qtnetwork PR and has found an issue with it in combination with the booster, so it shouldn't be merged yet. He would like if someone could help understanding why it doesn't work with the booster, as he doesn't know how to debug this issue with the booster.
08:06:26 <Chris[m]123> Something like that?
08:06:43 <Chris[m]123> But prefixed with #info?
08:07:07 <flypig> Yes, perfect. Please prefix it.
08:07:14 <Thaodan> Chris[m]123: Yes but you can also use I
08:07:14 <Chris[m]123> #info Chris[m]123 has tested the qtnetwork PR and has found an issue with it in combination with the booster, so it shouldn't be merged yet. He would like if someone could help understanding why it doesn't work with the booster, as he doesn't know how to debug this issue with the booster.
08:07:21 <dr_gogeta86> cargo build
08:07:21 <flypig> Chris[m]123, is your PR a backport from Qt 5.12? We have to be careful with licences.
08:07:24 <dr_gogeta86> sorry
08:07:28 <flypig> :D
08:07:37 <Thaodan> Your name is prefixed in front of info
08:07:53 <Chris[m]123> Ah, that makes sense. I'll try it again
08:08:00 <Chris[m]123> #undo
08:11:00 <flypig> Were you planning to re-add your comment for the logs Chris[m]123?
08:11:38 <Chris[m]123> flypig: It is indeed a backport. I have, as recommended by I believe dcaliste, read through the PR and made some notes on what the issue was for myself and an overview of how it was solved. I used that information for writing it from scratch. I didn't copy code as far as I was aware of.
08:11:44 <Chris[m]123> #info I have tested the qtnetwork PR and have found an issue with it in combination with the booster, so it shouldn't be merged yet. I would like if someone could help understanding why it doesn't work with the booster, as I don't know how to debug this issue with the booster.
08:11:55 <flypig> Great, thank you :)
08:12:44 <flypig> We should move on as we're over time on the PRs. Thanks for the good discussion, let's move to General Discussion and have some more.
08:12:51 <flypig> #topic General discussion (20 min)
08:13:06 <flypig> dr_gogeta86, do feel free to continue.
08:13:32 <Chris[m]123> Sorry for interrupting you :)
08:13:45 <dr_gogeta86> ok  ... back on track I've finally acquired a Sony 10ii ... (was a refurbished one )
08:15:19 <dr_gogeta86> works fine with stock android ... but with SFOS no audio ... the culprit is the TA partition corruped  ( 07-08 07:12:31.348  7736  7736 I audio_hw_sony_cirrus_playback: get_ta_array: Read TA unit 4702 (size=0) values: 0x0 0x0 0x0 0x0
08:15:19 <dr_gogeta86> ) ... many user got the same issue ... but then flashed back to android
08:16:29 <dr_gogeta86> also documented here https://forum.sailfishos.org/t/4-1-0-24-4-2-0-21-xperia-10-ii-xq-au51-single-sim-pulseaudio-dies-with-coredump-no-sounds-in-device/8268/2
08:16:50 <flypig> Congrats on your new phone dr_gogeta86
08:16:52 <Thaodan> dr_gogeta86: please link the forum post first next time
08:16:55 <Thaodan> thanks
08:17:36 <Thaodan> We have looked into this but didn't have any results so far.
08:19:00 <dr_gogeta86> I think the only solution is to flash values on TA partition
08:19:28 <flypig> So you were able to reflash Android, then SFOS again, to fix the issue?
08:19:34 <dr_gogeta86> is a kind of surgery ... but this is how the CAF open driver works ... apparently Stock driver ... aren't affected
08:19:41 <Thaodan> We have some ideas but nothing concrete, we don't exactly know how far the values form the TA values are used.
08:19:41 <dr_gogeta86> same issue reflashing
08:20:01 <Thaodan> I suggest to ask the question in the next meeting to we can prepare an answer for you.
08:20:24 <dr_gogeta86> I hope to fix first :-D
08:20:36 <thilo[m]> I am currently working on building a sfos application with a (compiled to native with graalvm) java backend. Is anyone interested in the topic?
08:22:01 <flypig> I'm fascinated to hear thilo[m] how the integration from Java works (and what your app is)
08:22:10 <Thaodan> IMHO: Java on SailfishOS is cursed, we are lucky for not having it. But is is interesting.
08:22:15 <Thaodan> *it is.
08:22:49 <flypig> dr_gogeta86, will you come back to us at the next meeting with an update?
08:23:02 <thilo[m]> its going to be a little game adapted from the android app open flood. Its growing here: https://github.com/thigg/sailing-the-flood-to-java
08:23:43 <flypig> Are you calling Qt methods from Java?
08:24:04 <flypig> Or do you essentially have a canvas?
08:24:59 <thilo[m]> grallvm is a project that compiles java applications to a minimal native image with a minimal runtime enviroment. In my case, I compile a shared library (.so) and talk to the api of the native lib by putting everything into flatbuffers. The interface to the shared lib is connected to the c++ part of the app
08:25:17 <thilo[m]> the c++ part and the qml part are a normal qt app
08:25:57 <thilo[m]> it might be possible to do the qt part in java as well, but thats going to be a bit complicated, and I am not sure graal can compile that without additional configuration
08:26:36 <thilo[m]> currently it looks like the overhead is way smaller than python. The resulting binaries are around 12 mb and that should also be the memory footprint of the thing
08:26:36 <flypig> Flatbuffers being essentially cross-language serialised structures?
08:27:01 <ExTechOp> Would anyone be interested in making a mosh port into Chum?
08:27:17 <thilo[m]> yes. I started with json, but I went a bit wild on performance. I wanted to see whats possible and the overhead of flatbuffers and io performance is way higher
08:28:20 <thilo[m]> s/overhead/serialized size/
08:29:10 <flypig> It sounds like an amazing approach thilo[m]. Would you be up for sharing your experiences in the newsletter?
08:29:52 <flypig> ExTechOp, do you have some feel for how much work would be involved?
08:30:22 <thilo[m]> i'd love to
08:30:34 <thilo[m]> I think I can also publish the game by that time
08:31:13 <flypig> thilo[m], ah! Superb! Then we should discuss it further :)
08:31:23 <thilo[m]> yes, let me know
08:31:47 <flypig> Is the app already compilable and runnable?
08:31:58 <ExTechOp> Unfortunately no, I just know udp-based mosh would be very nice for connections where connectivity may be an issue. I use it all the time on my other devices. There is a port, but it's kinda old.
08:31:58 <thilo[m]> yes
08:32:13 <thilo[m]> I'll put the build on github
08:32:44 <flypig> thilo[m], great, I'll have to try building it (I may need your guidance on that too...).
08:32:46 <thilo[m]> flypig: actually there is already a build on github.
08:33:01 <thilo[m]> yes, building is a bit complicated, but I can help you
08:33:08 <flypig> ExTechOp, I agree, it's an obvious good choice for a mobile device.
08:33:42 <flypig> thilo[m], that doesn't surprise me, but thanks, I'll seek your advice :)
08:33:50 <dr_gogeta86> @flypig, I hope ...
08:34:23 <flypig> dr_gogeta86, please do, even if it's to let us know you're still looking into it :)
08:35:30 <flypig> Okay, we have hit time I'm afraid.
08:35:51 <flypig> 2 mins for any final comments or additions to the minutes.
08:36:19 <thilo[m]> additional comment: thanks for chairing today, flypig ;)
08:36:30 <flypig> Thanks thilo[m] :)
08:36:35 <flypig> And I'll take the opportunity to thank the Bug Coordination Team for their amazing work again this fortnight:
08:36:53 <flypig> #info As a result of the Bug Coordination Team's excellent work, we now have:
08:36:55 <flypig> #info  - 6 high quality bug reports now recorded internally and tagged as "tracked".
08:37:01 <flypig> #info  - 1 bug reports tagged as "fixed".
08:37:05 <flypig> #info  - 2 flagged as duplicates.
08:37:43 <thilo[m]> we should send pherjung a cake ;)
08:37:53 <flypig> Seconded :)
08:37:56 <flypig> Alright, time to wrap up.
08:37:59 <flypig> #topic Next meeting time and date (5 min)
08:38:03 <flypig> Proposing Thursday 1st September at 07:00am UTC
08:38:17 <flypig> Any objections?
08:39:23 <flypig> Okay, let's set it at that then!
08:39:24 <flypig> #info Next meeting will be held on Thursday 1st September 2022 at 07:00am UTC: 2022-09-01T0700Z
08:39:40 <flypig> Thanks for some lovely discussion as always everyone!
08:39:45 <flypig> #endmeeting