<joolean>Hello! I'm trying to build an updated version of a package I maintain in guix. I've pulled master, but when I do `guix build [my package]' I get "guix substitute: warning: ACL for archive imports seems to be uninitialized, substitutes may be unavailable" and then later "guix/ui.scm:1255:8: Wrong type to apply: #<unknown-type (0x24f . 0x7faf767a2e48) @ 0x7faf76d672e8>"
<buenouanq>Corin: I am a normal casual computer user (i.e. not a software developer, Guix or otherwise) and have been using GuixSD as my main and only OS for the last year or so.
<buenouanq>Have had fewer problems than I ever did on SystemDebian.
<Corin>I notice there are a lot less pre-compiled packages than distros like Debian though. Only around 12%.
<buenouanq>Guix is about software freedom - A big part of that is not blindly trusting builds and actually verifying them yourself.
<buenouanq>It's all automated though, you really won't even notice.
<thomassgn>I've set up my .config/guix/latest link to point to a git repo to hack on packages and services, but now evrytime I call a guix command from bash it compiles a lot of guile files to my cache. Can I either: compile them like guix pull does (preferred); or set some variable to make it be compiled only once or some such?
<thomassgn>the git repo is a clone of savannah's guix repo
***Piece_Maker is now known as Acou_Bass
<lfam>thomassgn: You have a couple options. You could just go to the source directory and run `make`. You could use `git archive` to make a tarball that you pass to `guix pull --url`.
<brendyn>amz3: amz3` Did you end up managing to build GNUnet from git?
<baconicsynergy>brendarn: I need to reinstall guix :) I think I'll spin up a vm pretty soon
<brendarn>How can I find out why a paticular package is being built? I run `guix build serf --check' and then emacs-minimal gets download. Why? I run `guix graph serf --type=bag-with-origins|grep emacs
<buenouanq>ng0: I hear you're attempting to bless us with a gnunet service?
<ng0>well since ebing clear since the end of 2015 about what needs to be done and no one picked it up and faded into forgotten grounds so that I in parallel came up with ideas almost similar to what the past GSoC discussions were about, yep. it just takes a bit longer, I hope that the last two bugs get solved and we can release a new gnunet version.
<ng0>actually I'm working on the service since september but ran into walls.
<ng0>ah, don't take it as criticism though. I was just a bit surprised when I discovered the past work in december when I talked about this with lynX
<buenouanq>I see gnunet as the much needed future of the internet.
<buenouanq>I see guix as the much needed future of package management and meta os design.
<ng0>I see myself in the position of someone who integrates gnunet into systems and constructs a system around it. I'm still learning scheme, so when it comes down to the advanced parts I might or might not require either a long break to learn or some mentor.
<buenouanq>Having them work nicely together will just be fantastic.
<ng0>at least I hope I will be able to achieve what I planed without much help
<ng0>sneek: later tell snape: with the two font licenses applied 0ad works
<ng0>I think I know enough to ask the right questions though. The theory of one part, the distribution part, is planned through, twice. What comes afterwards like distributed longterm storage etc are projects entirely up to my project and not Guix specific. would be cool to end up in Guix eventually though depending on how it will be integrated if not in it's own application.
<ng0>sneek: later tell snape: btw, 0ad is pretty huge, I don't remember the total size after it is build but I hope it's small enough to not be as big as libreoffice.
<ng0>buenouanq: But I think I'm doing okay, just sometimes I have too high expectation for myself in contrast to my view on other peoples work and behavior. Having some problems 2 years after discovering a language, coming up with a plan for a system and doing too much to sit down for theory and fun often, that's okay.
<buenouanq>Well, what exactly are you trying to put together then? I was imagining just a normal service like we currently have for postgres or whatever - What you're hinting at sounds much more grand.
<ng0>yes, a normal service. but also I expect it to scale to something similar such as our dovecot service. I only had problems testing it to proof wether it's broken or wth is going on as I know my way around in gnunet configuration mostly. and furthermore it's a base requirement for the live USB system I work on
<ng0>I'm doing some more changes to the project after initial feedback I got, so in about 1.5 - 3 weeks the website should look different and have less implicit content, bugtracker existing (either at ii or at notabug), etc
<buenouanq>So it'll create the gnunet user and deal with all permissions launch at start up etc, then any other user can access gnunet through it, right?
<ng0>that's the way gnunet is intended to be set up, yes
<buenouanq>so I won't have to manually do that or run gnunet arm -s or anything anymore
<ng0>also solutions to be documented for the maintainer question, etc. "no maintainer per se" just reflects the state that I was reading into lots of approaches how to solve the role and also shows that I like to take breaks to be prepared for most possible scenarios. It's not just me, but it's still small.