Hey #Guix

Today, the channel guix-science reached its 300th Pull Request! 🎉

Feel free to join the party. 🥳
Join the fun, join Guix! 🤩

https://codeberg.org/guix-science/guix-science

We with Joseph made a short, but important guide about Load Paths for Guile developers.

https://trop.in/blog/how-to-set-up-guile-load-path

#guile #guix

We with Joseph made a short, but important guide about Load Paths for Guile developers.

https://trop.in/blog/how-to-set-up-guile-load-path

#guile #guix

Hey #Guix

Today, the channel guix-science reached its 300th Pull Request! 🎉

Feel free to join the party. 🥳
Join the fun, join Guix! 🤩

https://codeberg.org/guix-science/guix-science

https://codeberg.org/hako/Testament/issues/20
2025-10-29 Guix Moe build farm head node migration

Head node of the build farm has been migrated to a new machine nuporta.
CPU: Intel Xeon E3-1275v5 (4C8T, Q4 2015) -> AMD Ryzen 7 3700X (8C16T, Q3 2019)
RAM: 64GB DDR4 ECC
Disks: 2x 4TB HDD -> 2x 8TB HDD
Config: https://codeberg.org/hako/Testament/src/branch/trunk/config/nuporta.org

Specs of nuporta is more suitable for build tasks and long term operation. It can also be upgraded on RAM and SSD in the future.

I forgot to migrate Guix signing key and the old key is no longer available. A new signing key is used additionally for new substitutes, see Usage for setup.

(public-key (ecc (curve Ed25519) (q #552F670D5005D7EB6ACF05284A1066E52156B51D75DE3EBD3030CD046675D543#)))

nuporta is located in Finland, replacing the original cache-fi mirror. Since there's no machine in Germany any more, cache-de mirror has been made as an alias to cache-fi and cache-cdn is placed on the new head node.

#guix #guix-moe

The whole reason I got into #guix was to have something that actually solves the problem Conda claims to solve. With Conda, I've never had a project that, from beginning to end: (a) ran on multiple machines, (b) had a consistent environment where I could run my experiments, and (c) stayed like that.

I still need to convince my co-authors to switch to #guix (or even #nix), but I have hope we'll get there.

https://codeberg.org/hako/Testament/issues/20
2025-10-29 Guix Moe build farm head node migration

Head node of the build farm has been migrated to a new machine nuporta.
CPU: Intel Xeon E3-1275v5 (4C8T, Q4 2015) -> AMD Ryzen 7 3700X (8C16T, Q3 2019)
RAM: 64GB DDR4 ECC
Disks: 2x 4TB HDD -> 2x 8TB HDD
Config: https://codeberg.org/hako/Testament/src/branch/trunk/config/nuporta.org

Specs of nuporta is more suitable for build tasks and long term operation. It can also be upgraded on RAM and SSD in the future.

I forgot to migrate Guix signing key and the old key is no longer available. A new signing key is used additionally for new substitutes, see Usage for setup.

(public-key (ecc (curve Ed25519) (q #552F670D5005D7EB6ACF05284A1066E52156B51D75DE3EBD3030CD046675D543#)))

nuporta is located in Finland, replacing the original cache-fi mirror. Since there's no machine in Germany any more, cache-de mirror has been made as an alias to cache-fi and cache-cdn is placed on the new head node.

#guix #guix-moe

« Conda ≠ PyPI: Why Conda Is More Than a Package Manager »

From a scientific practitioner perspective (= verify, reuse, rebuild 2-5 years later), the key of package managers is: « Not just binaries: FHS-like user space inside environments ». Yeah! 🤩

About Conda 🤔 2 design choices seems limitations:

• Solver-driven consistency. Conda uses SAT-based dependency solvers.

• Building for the future. This ensures that binaries remain forward compatible.

🤔 An under-the-hood assumption seems wrong here. The SAT-based solver won’t remain consistent 2-5 years later.

Conda builds for the immediate future (2-6 months) which isn’t the timescale for scientific projects.

Scientific practitioner: STOP Conda! 🤒
Switch to #Nix or #Guix. 😍
The future will thank you. 😘

https://conda.org/blog/conda-is-not-pypi

The whole reason I got into #guix was to have something that actually solves the problem Conda claims to solve. With Conda, I've never had a project that, from beginning to end: (a) ran on multiple machines, (b) had a consistent environment where I could run my experiments, and (c) stayed like that.

I still need to convince my co-authors to switch to #guix (or even #nix), but I have hope we'll get there.

« Conda ≠ PyPI: Why Conda Is More Than a Package Manager »

From a scientific practitioner perspective (= verify, reuse, rebuild 2-5 years later), the key of package managers is: « Not just binaries: FHS-like user space inside environments ». Yeah! 🤩

About Conda 🤔 2 design choices seems limitations:

• Solver-driven consistency. Conda uses SAT-based dependency solvers.

• Building for the future. This ensures that binaries remain forward compatible.

🤔 An under-the-hood assumption seems wrong here. The SAT-based solver won’t remain consistent 2-5 years later.

Conda builds for the immediate future (2-6 months) which isn’t the timescale for scientific projects.

Scientific practitioner: STOP Conda! 🤒
Switch to #Nix or #Guix. 😍
The future will thank you. 😘

https://conda.org/blog/conda-is-not-pypi

Dear @zimoun , thank you for such a rich meeting and sharing your experience with #ReproducibleResearch and #guix for this #tem25 thesis . :)

#orgmode Babel (blocks of code calling anything integrated with the text and images) are great for #LiteraryProgramming and reprod. research.

Also, thank you for your mention of #ggplot2 that @ansate nailed too a little later. :)

Thanks also to @bzg too. 🙏

Re #PeerGovernance, I'd be delighted to share my experience with the guix community anytime.

Today, @edumerco motivated me to give a deeper look to #Org mode and #Emacs #Lisp for processing data as a reproducible computational notebook. It reminds me this great MOOC [1]. 🤩

And today I learn more about #Sociocracy thanks @edumerco! Well, the concept of #Guix teams needs more love. 😍

Bah the kind of day when you feel part of something. 🥳

Thanks @bzg for the connection. 😁

1: https://www.fun-mooc.fr/en/courses/reproducible-research-methodological-principles-transparent-scie

Today, @edumerco motivated me to give a deeper look to #Org mode and #Emacs #Lisp for processing data as a reproducible computational notebook. It reminds me this great MOOC [1]. 🤩

And today I learn more about #Sociocracy thanks @edumerco! Well, the concept of #Guix teams needs more love. 😍

Bah the kind of day when you feel part of something. 🥳

Thanks @bzg for the connection. 😁

1: https://www.fun-mooc.fr/en/courses/reproducible-research-methodological-principles-transparent-scie

Using Guix for Reproducible Research—a guide in 4 steps 👇
https://guix.gnu.org/cookbook/en/html_node/Reproducible-Research.html

Cc: @swheritage

#ReproducibleResearch #OpenScience #Guix

Fixed three packages in GNU Guix:
156d81666b2 gnu: tracy-wayland: Fix build.
bedb96ad2e3 gnu: trueprint: Fix build.
ae7386f007c gnu: tlf: Fix build.

The common motive of those errors is that Guix now uses GCC 15 by default, while some packages use deprecated C/C++ features that are now marked in GCC as errors. A quick-fix for such errors is to use older GCC version (say, 13) in "native-inputs".

This can be a "low-hanging fruit" for those who want to make the first-time contribution to #Guix.

Fixed three packages in GNU Guix:
156d81666b2 gnu: tracy-wayland: Fix build.
bedb96ad2e3 gnu: trueprint: Fix build.
ae7386f007c gnu: tlf: Fix build.

The common motive of those errors is that Guix now uses GCC 15 by default, while some packages use deprecated C/C++ features that are now marked in GCC as errors. A quick-fix for such errors is to use older GCC version (say, 13) in "native-inputs".

This can be a "low-hanging fruit" for those who want to make the first-time contribution to #Guix.

A second blog post about the technology behind the bordeaux build farm for #guix is online. Where we learn how to talk to the data service and the build coordinator from #guile scripts. https://enge.math.u-bordeaux.fr/blog/tbfg-2.html