Re: Guix (and Guile's) promise, and how to (hopefully) get there

[Divya Ranjan <divya@subvertising.org>]

Maxim Cournoyer <maxim.cournoyer@gmail.com> writes:

> Hey,
>
> "Thompson, David" <dthompson2@worcester.edu> writes:
>
>> On Sun, Oct 27, 2024 at 3:13 PM Ekaitz Zarraga <ekaitz@elenq.tech> wrote:
>>>
>>> Many people on this project have tried to change GNU from the inside and
>>> are very critical with the FSF (see the https://gnu.tools/). I think
>>> that's also a good way to do things, changing them from the inside.
>>> Fixing them for all our friends. Honestly, the argument of getting
>>> distance with GNU and the FSF is too simplistic to be taken seriously.
>>
>> Changing GNU/FSF from the inside has been a losing strategy for at
>> least a decade, as a conservative estimate. Nothing has meaningfully
>> changed for the better and the situation continues to deteriorate both
>> socially and infrastructurally. Many have tried to reform GNU, all
>> have failed. Some burn out and never return. Those that remain choose
>> to inhabit the fringes; projects that are historically GNU but in
>> practice are no longer concerned with the project as a whole (Guile
>> and Guix, for example.) We unsubscribe from gnu-prog-discuss and move
>> on. Thinking that GNU can be changed at this point is what is truly
>> too simplistic to be taken seriously. The GNU brand is and has been a
>> net negative for Guix. Juli did a great job describing why in an
>> earlier message. Every conversation about Guix I stumble upon online
>> inevitably derails into a negative discussion about GNU and it's hard
>> to break through the noise to explain that Guix is really cool,
>> actually. It's not priority #1, but we gotta eschew GNU.
>
> I'm late to the party but I thought I'd voice my feeling as I read this,
> catching up slack on the ML.  I find the assertions, or more
> specifically, the level of assertiveness, that GNU is or has been a net
> negative for GNU a pretty simplified world view, at least from my
> perspective.
>
> I believe GNU's largest contribution is to provide a philosophical
> foundation, e.g. articulating the software freedoms.  They've also
> proven dedicated in upholding the same goals they've set forth from the
> beginning; that's not something many organizations can be entrusted
> with.
>
> I know that some contributors (you can count myself as one), got
> interested in Guix *thanks* to its association with GNU.  That you
> assert so confidently that it's been a 'net negative' for the project
> almost feels insulting.
>
> I also do not happen to share your experience with people looking up to
> GNU or the FSF in a bad eye, but I don't take part in hip
> Mozilla/Microsoft/trendy-tech-of-the-moment sponsored events, so who
> knows.

Hello Maxim and others,

I had been mostly observing the opinions shared on this thread, and as someone who’s begun to contribute to Guix in small steps over the last month, I would like to share my perspective on this.

I totally agree with Maxim on the fact that to paint GNU as a “net negative” for Guix, is something that harms the community and not constructive. I myself started using Guix _because_ it was explicitly part of the GNU project, and one of the very few GNU/Linux distributions that aimed to be completely free/libre. Without Guix, someone who wishes to operate a completely free operating system has to rely on alternatives such a Trisquel and Parabola, and in my opinion Guix surpasses them in several ways.

This liberation is a crucial part of any GNU software, and as such it has real impact on people, me and others do testify to this. I had been advised to instead go for NixOS but why would I choose a non-GNU software that doesn’t prioritize user freedom over one that explicitly does so? The technical differences are there, and even in those I believe Guix because of Guile is superior, but beyond that, _philosophically_--as Maxim reminded us of the foundation that GNU/FSF built--it impacts differently.

This is what made me choose GNU Guix and it is one of the primary reasons I overcame obstacles in real-life and other technical challenges to start contributing to the project, I know my contribution remains minimal until now, but I can promise that if Guix continues to walk along the path of GNU, my contributions would only multiply with time.

The overall contribution of GNU/FSF is a separate discussion, I don’t know how people ignore what GNU Emacs has continued to do for over 40 years, or the licenses that literally paved the way for giving a legal foundation to fight proprietary software. This discussion would not only be lengthy, but probably intense and could only marginally touch Guix.

The real question to ask is: does, the foundation and the community, intend to prioritize the values of free/libre software as they work on building it? How much do we value this, is it higher, lower or equal to the technical aspects of Guix?

In my opinion, currently we prioritize it highly or equally, and I think we _should_ continue with that. If the community and the foundation choses to change this to a lower priority, losing the spirit of free/libre software, then it makes sense to leave GNU.

I have used GNU/Linux distributions over the last 5 years, I spent 4 of them exclusively on Arch Linux. My primary machine used to be a 16 year old i3 1st generation, and Guix ran on it as smoothly as any distribution would. I never faced issues with firmware due to linux-libre kernel. This month when I switched to a newer machine, I realized that it’ll be unusable on linux-libre, and I was forced (with much regret) to rely on linux with firmware blobs. Other than this, I have faced no major issues from Guix itself, had I used linux-libre in Arch, I would’ve had the same issue. Thus, I think we might be focusing on the wrong things when we analyze lacunae of Guix.

Just like Emacs has ELPA and MELPA, we also have nongnu alternatives for those who are forced to rely on it, and that might be improved, but why consider that GNU is holding Guix back? I doubt anyone would say GNU has held Emacs back, it hasn’t. If at all, Emacs (despite the recent drama and others in past) has grown to remain one of the best editors with modern features (TreeSitter, multiple LSP clients, etc.)

Similarly, I would suggest the thoughts on improving Guix/Guile to look at aspects of the project where we might not be doing the best job, or if we can take different steps in terms of reachout. But as an user and contributor, please don’t take steps to separate Guix from GNU. It’ll be a considerable loss too both the projects.

Regards,

-- 
Divya Ranjan,
Philosophy, Mathematics, Libre Software.