IRC channel logs

<krascovict>suggestion to add feather wallet

<krascovict>Is there another channel to suggest new packages for GX?

<mange>Not really. You could send an email to the mailing list mentioning it, but that's no guarantee that it will get packaged. The most reliable way to get it packaged is to package it yourself.

<krascovict>mange, ok, thanks

<meaty>how do I put my guix machine to sleep? All the guides online are for systemd distros

<mange>I don't know your exact setup, but have you tried "loginctl suspend"?

<meaty>i have not but did not know about loginctl, thanks

<civodul>Hello Guix!

<efraim>o/

<futurile>morning both

<apteryx>lilyp: by the way, I have nautilus 46.x ready; will send it for review soon

<apteryx>our new many branches strategy is working well, but causing lots of extra traffic I suppose

<apteryx>ACTION 's reconfigure proceeds to substitute $world

<mccd>Heya, has anyone here managed to get nvm to work with guix? I installed it the regular way but it seems that nvm install stable doesn't work, no bin shows up.

<mccd>Well it seems it shows up, but I cannot execute it?

<mccd>hmm, well I managed to install it through the cli, but I can not actually install any executables with it

<efraim>you should check with ldd (guix shell gcc-toolchain) to see if nvm refers to any system libraries

<mccd>it was easier to just use nix to install this. I wanted a language server haha

<apteryx>managed to fix Totem (it was getting stuck often while seeking)

<janneke>sneek: seen mothacehe?

<sneek>mothacehe was in #guix 8 months ago, saying: doriru: not sure why is that one broken. you can always build it yourself following those instructions: https://guix.gnu.org/manual/en/html_node/Building-the-Installation-Image.html.

<janneke>sneek: botsnack

<sneek>:)

<civodul>it’s been a while

<janneke>civodul: i trivially added a choose-kernel step to the gui-test-program but it somehow failed and i wondered how on to go about and investigate. also, it seems the screenshot names get out of sync after enter-host-name. guess i'll write an email later today

<janneke>hmm, and cross building gnu fibers from i686-linux fails too:

<janneke>./pre-inst-env guix build guile-fibers --target=i586-pc-gnu --system=i686-linux

<janneke>it works without --system=i686-linux...

<janneke>anyway, i'll have to run

<efraim>I'm going to test building it from aarch64

<efraim>can't build gnumach-headers from aarch64

<efraim>well there goes that experiment

<theesm1>good morning guix^^

<futurile>morning theesm1

<attila_lendvai>deleting `/gnu/store/trash' in guix gc takes forever. i'm looking at CPU and iotop, and neither are overly active.

<efraim>Is the store locked while its running or just that terminal?

<civodul>janneke: cross-compilation is only really tested from x86_64-linux

<civodul>i think that’s already enough work :-)

<civodul>re choose-kernel, the installer normally logs to syslog IIRC, which is dump on the console when running these system tests

<apteryx>is someone good at reading flatpak manifests here?

<apteryx>I'm wondering about the deps listed in https://github.com/flathub/org.gnome.Totem/blob/master/org.gnome.Totem.json

<apteryx>which 'modules' are used directly by totem? and which ones are indirect (transitive) dependencies?

<futurile>Q: how do I list the contents of a package that's in the Store? So I know that font-gnu-unifont is installed by guix system - it says it's there with `guix system describe --list-installed=font-gnu-unifont. I don't want to install it in a shell, as I'm trying to understand how guix system uses it

<ekaitz>futurile: if it's in the store, you can guix build it to obtain the path and then list the contents of that?

<nckhexen>Also... that exact command prints a store file name(?) so what'm I missing. Differing definitions of 'list contents'?

<apteryx>find $(guix build x) ?

<nckhexen>Bonus free fun when the build fails :)

<apteryx>if it's "already in the store", chances are the build will pass ;-)

<apteryx>what would happen otherwise though?

<nckhexen>Eh, good point.

<ekaitz>apteryx: unless you guix pulled and you didn't upgrade your packages

<apteryx>find would try the items of the error message? heh

<ekaitz>this happens a lot to me, i pull but I don't update or some weird dependency changed and hehe

<nckhexen>apteryx: It evaluates to nothing, so you get a free reminder of how many million files you have in PWD.

<nckhexen>Nothing bad, just 'fun'.

<apteryx>kind of fun.

<nckhexen>s/'fun'/UNIX® fun™/

<apteryx>ACTION thought UNIX and fun were synonyms

<getstate>Would it be useful to extend "update-package-source" to account for packages defined with tree-sitter-grammar? Right now it fails with no "version" field?

<nckhexen>You certainly can't spell 'f— unix' without 'fun'.

<getstate>For context, `guix refresh` fails when I try to refresh tree-sitter packages eg: tree-sitter-vhdk

<nckhexen>It would certainly be useful, the challenge is more doing it in a non-offensive maintainable way. There are several other examples of this pattern.

<futurile>ekaitz, apteryx yes, thanks that gets me further along

<apteryx>ACTION looses hope with totem

<apteryx>it still hangs despite bumping to a newer commit which was supposed to improve things

<apteryx>in GNOME 47 it'll be replaced by 'showtime', or something

<apteryx>so... I'll leave it at that.

<Guest3>Does someone know a working commit of Guix that works with ARM architecture? Latest doesn't build

<apteryx>can freely available but not licensed game *data* be included in a package per the GNU FSDG, since it's not functional data?

<apteryx>hm, reading "License Rules", the spirit would probably make the game assets 'functional', but opinions seem to vary

<apteryx> https://www.gnu.org/distros/free-system-distribution-guidelines.html

<dthompson>given that code is data and data is code, it's certainly a gray area

<dthompson>I would guess that because the data doesn't have a license that makes it non-free by default

<apteryx>well, tE GNU FSDG covers games and games data explicitly as an example, under the 'Non-functional Data'. I guess that clears it up for me.

<apteryx>"For example, some game engines released under the GNU GPL have accompanying game information—a fictional world map, game graphics, and so on—released under such a verbatim-distribution license. This kind of data can be part of a free system distribution, even though its license does not qualify as free, because it is non-functional."

<apteryx>and the sentence before, " It can be included in a free system distribution as long as its license gives you permission to copy and redistribute, both for commercial and non-commercial purposes."

<dthompson>this FSDG is so funny

<dthompson>the*

<dthompson>surprisingly strict in some ways, and relaxed in others

<Guest3> https://ci.guix.gnu.org/eval/1741605/log/raw there arre bunchs of errors on the ci with that as error. is this expected?

<apteryx>it's very tight when it comes to executable code (such as firmware blobs)

<apteryx>and lax when it comes to data that can't be run

<civodul>for game data, i would personally follow the lead of Debian (which is stricter than FSDG)

<civodul>more generally, i think we should “own” our guidelines

<apteryx>I wasn't saying I find the GNU FSDG unreasonable, to be clear

<civodul>Guest3: looks like something broke between 503919f and 5794926: https://ci.guix.gnu.org/jobset/guix?border-high=1734159

<dlowe>When I open-sourced my MUD, I provided a sample world that had enough data to support all the code but wasn't the production data

<apteryx>About Debian being more strict than the GNU FSDG, that's odd, considering they opted to now include non-free firmware in their installers.

<Guest3>civodul: I got a 504 and the website is really slow. I am trying for an ARM build but latest fail. Do I see that https://ci.guix.gnu.org/eval/1733413 is latest to work with ARM?

<dthompson>that's one area where I think the fsdg is overly strict

<apteryx>from a freedom (and security) point of view it makes sense. you don't want untrusted code running on devices that can access the kernel's memory

<Guest3>Evaluation failed means a test didn*t pass?

<dthompson>"nonfree firmware is fine as long as it doesn't get loaded from the cpu" is such a bonkers policy

<apteryx>that's not how it's worded, though, I suppose

<dthompson>that's the effect

<dthompson>it's clearly a compromise to accomodate certain types of hardware

<dthompson>but not enough hardware to make FSDG distros usable for the majority of computer users

<apteryx>"If you trust your chip design, it's not too much of a stretch to also trust firmware that is not easily upgradable."

<dthompson>the FSDG lacks praxis. Debian made the right call.

<dthompson>a free system that no one can use accomplishes nothing.

<dthompson>guix should do the same so we can move stock linux out of nonguix.

<futurile>The main thing that's a bit odd about it is "if you have hardware who's firmware is not updateable" then it's OK, "hardware with firmware that is updateable" is nonfree. In many ways that's user-hostile. Since hardware's become more complex over time and more risky for end-users, it's a bad idea to run non-updatable firmware (and there's not much of it around anyway).

<ieure>futurile, No such thing as non-updateable firmware, just a matter of how difficult it is to update.

<dthompson>the fact that this also applies to CPU microcode was the last straw for me

<Googulator>"can access the kernel's memory" is indeed a good standard to use - "firmware" that's really just a binary blob you're supposed to load into kernelspace is dangerous, actual firmware that merely resides in volatile rather than nonvolatile device memory, not so much

<futurile>and if you've ever worked with a hardware vendor on an SoC, you just know their code is going to be riddled with problems. It's the nature of the industry.

<apteryx>futurile: we're talking about firmware companinies would be "game" to bake and not touch anymore

<futurile>ieure: sure "for certain values of" not achieveable by end-users.

<apteryx>which essentially makes it hardware

<ieure>The line between hardware and software is very blurry these days.

<apteryx>the solution will eventually come when the hardware will be freed itself (riscv is improving things on this front I hear)

<futurile>agreed - one of my coworkers once said something like "hardware is just slow moving software these days"

<Guest3>doubt that risc v will change things, like why should industry allow it

<dthompson>we need to make *today's* computers work

<Googulator>The core problem is, modern lithography is too expensive to produce consumer silicon at a profit

<civodul>regardless of one’s position on this issue, i think the first step we should take is to have our own guidelines with a process to amend it collectively

<Googulator>So higher-priced "pro" silicon is used to subsidize consumer-grade stuff

<dthompson>civodul: could a first step be to introduce "guix distro guidelines" v0 that is just the FSDG?

<Guest3>civodul: own as of guix or gnu?

<Googulator>But the only economical way to prevent the diversion of consumer silicon to professional use (which would woefully undermine such a scheme) is through DRM-like "locks" in the firmware/microcode

<dthompson>and from there amendments can be made collectively

<Googulator>So, no chance of seeing reasonably performant, reasonably priced silicon that's truly free

<civodul>dthompson: yes, definitely

<apteryx>civodul: if that would be to allow people debating ad-eternam on the mailing list why firmware should or should not be included in our installer, I think I prefer the status quo.

<civodul>actually that’d prolly be the 2nd step, first one being: how do we define what goes in there, etc.

<civodul>(non trivial!)

<apteryx>proprietary firmware, that is

<podiki>hmm QA has some issue processing the comparison to master for mesa-updates https://qa.guix.gnu.org/branch/mesa-updates

<civodul>apteryx: it’s definitely tricky, which is why we’d need a solid process in place

<dthompson>civodul: hmm, well that sounds like something less actionable

<civodul>that sounds like governance

<Guest3>but isnt the acual problem that fsdg not having a clear defined rule for this problem? so just fsdg needs fixed? why reinven the wheel

<civodul>we can’t just go on without proper rules

<civodul>it just doesn’t scale

<civodul>(unless we want a bunch of BDFLs that is)

<dthompson>I think someone needs to use their influence to get a process in place ;)

<dthompson>then make a v0 guidelines doc that is just the fsdg because that is the status quo

<dthompson>Guest3: the fsdg is unfixable

<dthompson>you can guess as to why

<Guest3>actually no clue

<dthompson>ah okay :) in short, rms

<ieure>What is "fsdg?" DFSG, Debian Free Software Guidelines?

<dthompson>ieure: gnu free software distribution guidelines

<ieure>Thanks.

<dthompson>it's so confusing the dfsg and fsdg use the same letters

<dthompson>that the*

<futurile>ieure: no, there's a GNU version. Debian was first and then revised theirs later.

<Guest3> https://www.gnu.org/distros/free-system-distribution-guidelines.html

<dthompson>"free system" right

<Googulator>time to introduce SFDG too

<Googulator>and then FGSFDS

<dthompson>yeah how many anagrams can we get

<futurile>ACTION laughs

<civodul>we need more people on board to define governance

<dthompson>Guix Somewhat Free Distribution Guidelines ;)

<civodul>next step is probably the RFC process: https://issues.guix.gnu.org/66844

<Guest3>dthompson: okay, so why is he a problem? Is he some diicator or what. I thought gnu is a organisation

<dthompson>Guest3: he's the gnu bdfl

<Guest3>dthompson: what I mean is, why one person can stop development if many say it is good to go

<dthompson>"chief gnuissance" as he prefers to be called

<dthompson>Guest3: it's a good question :)

<dthompson>why indeed

<Guest3>now I understand why so many people hate him

<futurile>civodul: I did do some research on how people bootstrapped 'governance', RFC->(bootstraps process)->Governance model+discussion -> Governance seems the route

<dthompson>too much to get into here

<apteryx>dthompson: you seem to have some grudge against the person; I don't think it's a good place to spit it out though.

<dthompson>a grudge isn't accurate framing.

<futurile>this was a good read - https://peps.python.org/pep-8002/

<dthompson>I don't think it's particularly unusual to have a negative opinion of how rms runs gnu/fsf these days

<futurile>it reviews some of the models and discusses upsides/downsides

<dthompson>civodul: I'm not involved in guix day-to-day stuff but I definitely support introducing some RFC process to start making progress on important governance issues

<Guest3>idea

<Guest3>check if ci already build it and prevent from building again locally

<Guest3>wastged hours of cpu time for building something that isn't going to work

<Guest3>does that make sense?

<civodul>futurile: yes; note that we also have teams, and now they’re even properly defined

<apteryx>Guest3: yeah, it's been suggested in the past

<apteryx>nobody has come up with the code yet

<dthompson>seems reasonable, though you'd want a way to build anyway in case it's just flaky

<Guest3>just ask user for confirmation

<apteryx>like "this package appears to have failed building on the Guix CI -- really continue?" or something

<Guest3>yeah

<Guest3>sounds good

<dthompson>but defaulting to not building seems fine

<dthompson>yeah and a flag for non-interactive use

<Guest3>now can someone please write that down, so information is not lost

<apteryx>conveniently logged

<Guest3>finding is another question

<apteryx>or you could create some [feature] tagged email to submit to bug-guix

<Guest3>wrote it down on paper. if in some years no one did it, i will do. currently no time because of university and not skilled enough (probably, still novice learned ofguile)

<podiki>i had put in a proposal for an open software foundation (if I remember name?) sustainability grant, and establishing governance for guix was one of the major tasks i proposed

<podiki>sadly didn't get the money :( or else i'd be living the dream of working fulltime on guix!

<Guest3>plain image kernel not supported - rebuild with... I just ran guix system image on an arm vm and want to create another vm, since host is arm cpu. x86 works out of the box. does arm require special treatment?

<Rutherther>podiki why is governance important?

<podiki>well for all the reasons we just saw above :) we don't have a real process for big changes in guix

<podiki>and we are growing (I think) so we could use some structure, much as I'm not much of a structure person myself

<podiki>we run pretty well so far I think, but we do have plenty of issues where I think some more structure will keep things moving along

<ekaitz>and we are too dependent in guile, that has the same problem but the lack of people it's way worse

<podiki>combine with probably needing more people/machine power for builds

<podiki>(for guix)

<podiki>maybe I'll search again for some funding opportunities, would really love to be a guix "employee" :)

<ekaitz>podiki: you are located in the US?

<podiki>yes. if you know where money is here, i will buy a shovel

<ekaitz>i've been paid for my work in guix for some years... but being in EU makes it easier for me

<podiki>or, err, metaphorically

<ekaitz>podiki: the NSF has money, that's for sure

<podiki>NLS grant for you?

<podiki>yeah i have a tab with some nsf funding i need to look through

<ekaitz>I had nlnet grants, yes

<ekaitz>several

<podiki>nlnet, right

<podiki>do we know anyone that has gotten grants or funding in the US for Guix work?

<ekaitz>but depending on the part of the US you live, this amount of money will keep you starving

<ekaitz>yes

<ekaitz>let me search

<podiki>thanks! feel free to PM or email me if that is better

<podiki>Guix with a few (more?) full time people would be a huge boost I think

<ekaitz>i'll write you an email

<ekaitz>podiki: if i could remember your name...

<ekaitz>(i think i remember it, but not how to write it)

<podiki> https://git.savannah.gnu.org/cgit/guix.git/commit/?h=mesa-updates&id=bacb3d62c7288da4d7ff4e0f6162ef199cca87d8 for instance :)

<podiki>thank you!

<ekaitz>thanks!

<ekaitz>damn, I was close

<podiki>close is still impressive!

<divya>Does anyone know how to mount NTFS file systems in guix? The typical setup in `operating-system' doesn't actually work. So I have to do it separately by typing out the command each time with sudo. Can I make a system service for it that runs every time I login?

<divya>And on that note, what's the best way to create a system service that requires sudo. I also wanted to add kmonad as a service, but having it as a home service didn't work since it required sudo to access my keyboard.

<ekaitz>divya: https://git.elenq.tech/guix-configuration/tree/desktop.scm#n155

<ekaitz>that works

<Guest3>how can i add substitutes on a foreign distribution?

<divya>ekaitz: That just adds the ntfs-3g package, I already have that. I want to know how to have my configuration mount ntfs disks

<ekaitz>divya: once you have that you should be able to mount things

<ekaitz>divya: check udisks-service-type

<divya>I can mount, but I need to do that by manually putting the command.

<divya>Let me check that.

<dariqq>Guest3: youd need to authorise the key (guix archive --authorise) and then add the url to the invocation of the daemon (--substitute-urls argument) in the systemd unit (or however else you use to start the daemon)

<dariqq>s/however/whatever

<divya>ekaitz: Can you hint a bit on how to use udisks-service-type, do I add that as a service in system configuration?

<ekaitz>divya: i think so, but i'm not sure

<divya>At least some documentation? I am not seeing anything exactly on it.

<ekaitz>divya: it comes as is because you should use it as any other service i think

<ekaitz>but i hate configuring things so i might be the worst person for this task

<theesm1>re the firmware/fdsg discussion earlier on here: the secondary processor exception gnu has is pretty funny; iirc purism utilizes that on the librem 5 by basically using a secondary cpu as a service processor to load all the binary blobs before handing over to the main CPU again which is odd but compliant to ryf/gnu standards etc.

<divya>ekaitz: Yeah, but how do I add the specific disk and mountpoint to the service.

<ekaitz>divya: look a little bit below: there's some configuration

<ekaitz>which doesn't say anything

<ekaitz>great

<ekaitz>idk :(

<divya>Yeah I was looking hard haha

<divya>No worries, let me see what I can find, maybe ask in guix-devel or something.

<divya>Since I'm with guix, I guess it is not at all advised to just manually edit /etc/fstab

<ekaitz>divya: better help-guix

<divya>ah, right

<ekaitz>(or both)

<ekaitz>podiki: mail

<podiki>ekaitz: received, thank you!

<divya>Last time I tried fixing this, I had seen this: https://lists.gnu.org/archive/html/help-guix/2021-10/msg00038.html

<divya>Not sure it's been fixed. I hate being on ntfs, but I have sensitive data on it and can't change clean it to change filesystems.

<ArneBab>Is anyone already working on updating icecat to FF 128?

<ArneBab>(new ESR)

<Guest3>dariqq: thanks

<futurile>dthompson: I feel there's something wrong with this sentence from spritely - "our CTO @dthompson co-hosts the annual Lisp Game Jam twice a year" - did you get confused about annual, or perennial?

<futurile>dthompson: is it that one where people from the USA get confused by us Brits saying "fortnightly"

<dthompson>futurile: the game jam is twice per year

<dthompson>does that clear it up? I'm a little confused

<podiki>"annual" = once per year...?

<podiki>although maybe that is not what it strictly means?

<dthompson>lol okay yeah I see it

<dthompson>my brain kept seeing "biannual"

<ieure>banannual

<podiki>silly brains

<dthompson>I know it's a biannual jam so I kept reading it that way...

<podiki>i often tell my students to read out loud for this reason, often helps catch the brain fixing things from what is written :)

<dthompson>😅

<podiki>as long as we aren't talking about biweekly, still a great mystery of the timing world

<ieure>podiki, Or how many days are in the week (if you're a bodybuildingforum member).

<podiki>hahaha that one is great

<podiki>was used in a book I used for a class, about every day math

<ieure>Truly one of the all-time greats.

<podiki>to be fair, counting and intervals is always tricky :)

<ieure>Oh dang, the original forum is defunct!

<dthompson>ieure: oh that's a classic

<ieure>Hope someone saved a copy of that thread.

<dthompson>mega64 does a theater-style reading of it

<podiki>not to get too off topic, but came across this series of reading famous letters, clearly these internet threads are the future of that

<dthompson>would love to see some old gnu mailing list threads acted out

<futurile>dthompson: sorry, probably was a bit unclear - anyway, made me chuckle :-)

<dthompson>no it was clear! I was just blind to the typo

<nutcase>ekaitz: can you provide more information about your nlnet funding? It sounds interesting and I just saw https://nlnet.nl/fediversity/ where they even mention NixOS. Furthermore, the FAQ tells that people outside of the EU can also apply. Maybe this is an option for podiki ?

<ekaitz>nutcase: i'm not funded by nlnet anymore, but have been for a couple of years. Not only me, I also funded some other people's work with the projects I proposed.

<ekaitz>nutcase: and yes, you can apply if you are from the US

<ekaitz>and podiki Guix is a good fit for that call

<ekaitz>the problem with NlNet (if it has any) is it's project oriented. You can't just say: i will maintain this project

<singpolyma>Always gotta be building something new to qualify

<ekaitz>nutcase & podiki : I talked about NlNet when I talked for Guix London. https://youtu.be/Cj7DyiRqWBk

<[>does building new bugs count?

<nutcase>ekaitz: thank you for the link. I'll watch the video later that week (hopefully).

<ekaitz>nutcase: don't waste your time with the presentation if you don't want to, we talk about nlnet in the questions part

<divya>What is the go-to way for having a custom pipewire configuration? I used to have a very specific pipewire configuration. Maybe first tell me how to have a simple working configuration, because I don't think even that works for now.

<shcv>has anyone written code for fetching package definitions from a url? It seems like a lighter-weight way to add a package to a system than subscribing to a channel

<jaft_r>divya: Should just be "(service home-pipewire-service-type)" in the list under "services" in your Guix Home (for a simple configuration; simple, here, being the out-of-the-box config.).

<divya>jaft_r: I do have: (service home-pipewire-service-type

<divya> (home-pipewire-configuration

<divya> (wireplumber wireplumber-minimal)))

<divya>But when I open pavucontrol, I see that pulseaudio isn't working

<divya>I do listen audio when played, but something like telegram-desktop doesn't work

<jaft_r>Gotcha; remind me: you're not using %desktop-services, right?

<divya>Not in home configuration

<jaft_r>But in your system config.?

<divya>Oh, yes I do have it in system config.

<jaft_r>Gotcha; O. K. Did you remove (or modify) pulseaudio-service-type, any?

<divya>I had tried (service pulseaudio-service-type) in system services before, but don't think it did anything.

<jaft_r>I think it's included in %desktop-services (I only reference it in my config. under "modify-services"); if you haven't removed it – any –, then I'm afraid I'm not sure why it wouldn't be present. I'm using a pretty unmodified setup and it's just pulseaudio-service-type, from %desktop-services and home-pipewire-service-type in my home config. for things to work, for me.

<divya>hmmm, any documentation for looking into this? I need a pretty tinkered setup that uses pipewire, pulseaudio and jack.

<jaft_r>Granted, I'm not specifying wireplumber-minimal (just using the default) so /maybe/ that's it but it's beyond what I'm familiar with, I'm afraid. I wouldn't expect so but I don't know how wireplumber works. PipeWire service documentation's at https://guix.gnu.org/manual/devel/en/html_node/Sound-Home-Services.html#index-PipeWire_002c-home-service

<divya>Thanks.

<nutcase>if anyone is familiar with rust / cargo-build-system: I have a beginner's question: "anyhow" source can't be loaded during a package build. What am I doing wrong? https://paste.debian.net/1333108/

<divya>So now I have pipewire working, and pulseaudio also getting handled, but I can't get Jack to work.

<divya>Even when I run an application with pw-jack, it dosen't seem to work. For example, pw-jack carla doesn't connect to jack.

<rekado>divya: pw-jack works fine for me. I use pipewire with Guix Home.

<rekado>no special configuration; just (service home-pipewire-service-type)

<rekado>I tested with ardour

<divya>Can you check if pw-jack carla works?

<divya>I'll check with ardour, thanks for the recommendation.

<divya>Running pw-jack and then selecting Jack/Pipewire in the audio system option of Ardour doesn't work.

<divya>In the backtrace, I see failed to load module "atk-bridge".

<dthompson>pw-jack has worked fine for me as well.

<dthompson>I also use ardour

<podiki>ekaitz and nutcase thanks for the info!

IRC channel logs

2024-10-22.log