Hi all,

Oleg pointed me to the excellent article by Konrad about "conviviality":
https://blog.khinsen.net/posts/2026/07/06/conviviality.html

It explains so many things, for example why I got drawn to Guix, and why most of you despise genAI. It does a better job at describing the risks of genAI than the GCD does; and it is more convincing than Konrad's own message about the GCD (perhaps because the article is decoupled from any specific technology).

But more interestingly, the article also clarifies why I felt the need for using genAI in Guix in the first place: the conviviality is already largely lost.

To paraphrase one of Konrad's eye-opening phrases: "I package what the software allows me to package, rather than what I actually want to package."

I spend most of my Guix-time 'fixing' packages that I have never heard about, because that churn is necessary to refresh/add the packages I do care about. And we fail to keep up (despite the extremely hard work by Oleg and others who contribute so much).

It should be a warning sign that me and others *already* feel the need to use genAI just to be able to use Guix; we went off-track somewhere.

And Konrad taught me that those other packages weren't even 'broken' to begin with. I feel ashamed for having been a fan of the Python wall of shame. But then, why can those (old) packages prevent me from including the (new) packages I do want to use?

In fact, we are doing *worse* than other ecosystems. Half of Konrad's article is about the poor state of the Python world. Yet adding packages to PyPI or conda feels far more convivial than adding them to Guix!

Technically, I'm not limited at all; I could put everything I want in my own channel. But Guix can be so much more than that. And we even regularly break 3th party channels, forcing them to keep up or perish. Arguably we are becoming the problem we set out to solve.

The article also allowed me to articulate why "guix refresh"-bots would not be the solution. The need for those bots is the problem; and that need would only be reinforced by making those bots. (We should still make these bots though, because we want to, not because we need to.)

How exactly we should revive conviviality w.r.t. what we package would be a collective exploration. If any operating system can do it, it is us, because all the groundwork is there to keep old packages around.

Hugo


  • Reviving Convivia... Development of GNU Guix and the GNU System distribution.
    • Re: Reviving... Simon Tournier
      • Re: Revi... Development of GNU Guix and the GNU System distribution.
        • Re: ... Andreas Enge
          • ... Development of GNU Guix and the GNU System distribution.
    • Re: Reviving... Andreas Enge
      • Re: Revi... Development of GNU Guix and the GNU System distribution.
        • Re: ... Andreas Enge
          • ... Development of GNU Guix and the GNU System distribution.

Reply via email to