NixOS is not dying, please don't spread fear actively

[Aleksanaa]

Yes, NixOS is currently taking a big reputational hit in terms of public perception. And I think that this is not only warranted, but neccessary.

If everyone learns fact checking (and have enough time!) on the internet, I’d agree with this is bringing the problem to public attention, and it’s making the situation towards a better one.

But it is not. People are often full of mindless first impressions, and the more impactful these impressions are, the further they spread. Someone who don’t use NixOS are already telling me something implying, “What happened to your NixOS community? The community is really bad, and you still use it.”

For those who can’t do full fact checking to participate in these discussions (we have more than 1000 people in the Chinese group, but many of them don’t want to go through all of our discussions), the helplessness conveyed by this message does not produce any better effect in the current situation other than to persuade them to quit!

It is warranted, because people who are new to this community or want to be involved should know what they are getting themselves into. That’s just informed consent.

It’s not. People are warranted to access the fact, the shorter and more accurate fact, rather than some sensational news.

Instead of being given a series of “facts” that imply “we’re screwed,” people should be given a short and precise way to understand how the NixOS Organization used to work, what were the problems with the mechanism and people, and how we could improve it.

People are now voting with their feet, and this creates the necessary external pressure to actually do and improve things.

This is not voting in any means. Voting to leave is not voting to decide. People voting to leave cannot again vote to decide. If you want to continue to tear this wound open, we’re going to need to fix this problem in the future at a cost far greater than solving the existing problem.

[nat-418]

Your framing is what is misleading: marsam, to take one example of a contributor who left, was the #4 top contributor to nixpkgs. Marsam made explicit that things are not ok.

[jchw]

I don’t think there’s any disillusion that the contributors that left will leave a hole of sorts, especially marsam. I’d hope that there can be an amicable ending and everyone who is interested in still working on Nixpkgs can come back.

But, as it happens, big-time contributors to projects do often wind up leaving for a wide variety of reasons, and Nixpkgs is graced with not one but a handful of contributors who are contributing on a similar level.

Certainly, there isn’t zero impact, but I do agree the doomsaying is misleading. NixOS will continue to be developed, maintained, released.

[igel]

@Aleksanaa

People are often full of mindless first impressions, and the more impactful these impressions are, the further they spread.

I believe that this statement itself is correct.

Nevertheless, how to make people keeping participating instead of dropping out?

To me this title sounds dramatic, in a way you could call it passive aggressive against others who could make their (valid) criticism public

• which is the worst that can happen for the community and project

From my view onto the situation, I cannot see that those committers mentioned are aware of what happened and taking means and actions, e.g. to find a common way, to take action now when it wasn’t done for years

I’m probably not aware of all messages (of discourse, matrix, …). What I can read is people who try to share the interpretation of the current situation for their personal “believes” and interests

I couldn’t read apologies for things that happened, for behavior shown and ideas how to make sure it will not happen again. It feels like people want to let it escalate (and show their power [by not acting?])
That makes me afraid. Change is not happening if all goes on like it was …

Eelco appears to me to be at the to of the nix-hierarchy and I’m not aware of decisions made against his will or interest. Perhaps a little bit like a king?
How he acts is shaping the behavior of other (committers). Can you get a top committer without Eelco’s support? So, do you have to support Eelco?
I was told Eelco is leading nix/NixOS. Does someone noticed (active) leadership by him in the current situation?

[Aleksanaa]

I’d say nothing if you are really going blind with the rest of this graph. Marsam has done important and mass jobs, it’s true, but the title you want to use still don’t get the fact.

I believe anyone who is not informed will think that this sentence means “the first contributor is gone”, and the contributor has done MAJOR changes, which means the number is also far ahead in terms of share, just like if Felix Yan left Arch Linux. It is a complete negation of the entire project.

If you want to continue playing word games on this title, I have no problem with it, and you can find many more examples of this in journalism.

[Aleksanaa]

How he acts is shaping the behavior of other (committers). Can you get a top committer without Eelco’s support? So, do you have to support Eelco?

It’s nothing about Eelco Dolstra. @domenkozar (sorry for the ping!) decided to give me permission, but he was just making sure that no one else had any objections and judging whether I had enough upvotes (based on the reactions below my application comment, I think I should have enough upvotes).

As I said in this topic. Eelco is leading the development of Nix, at least is one of the Nix team which is leading the development of Nix. However, Eelco hasn’t leading the development of NixOS and Nixpkgs in any ways, since some years ago (If you don’t have enough experience with this yet, Nixpkgs is the package set for NixOS, but it also holds the NixOS module system, so NixOS is the manifestation of Nixpkgs).

I couldn’t read apologies for things that happened, for behavior shown and ideas how to make sure it will not happen again. It feels like people want to let it escalate (and show their power [by not acting?])
That makes me afraid. Change is not happening if all goes on like it was …

Not no apologies, regarding the Anduril donation controversy, people from the foundation team have made some responses, but many people in the community do not feel that their response is enough and cannot prevent us from falling into similar controversies again.

Regarding the maintenance status of Nix, before anyone points it out, the Nix team has indeed discovered many problems with participation and current stability. But again, the solutions they proposed were not seen as community-oriented. Of course some people didn’t read it either. In addition, according to others, some of their actions are hurting the feelings of some other contributors.

Therefore, the community urgently needs a well written regulation to enforce a healthy management model, to eliminate people’s worries about something very bad happening again, and to resolve some of the current management chaos.

Does someone noticed (active) leadership by him in the current situation?

People think he should be responsible for adding Anduril sponsor without discussion. He is believed to be responsible for joining flake without consideration and then failing (or delaying) to stabilize it over the years. People think that he blocked some nice features, such as the addition of “meson build Nix”.

This is the main thing I saw. There are others but you may need to read very lengthy discussions and articles to gather. Sorry.

[tomberek]

Aleksanaa, thank you for reminding us about what matters and encouraging peolple to be optimistic. I agree.

Your original post should stand, as-is. I keep seeing everyone drawn into more debates and argument. I fell for this myself. For that, I appologize to the community. To others, please stay positive.

[igel]

Isn’t being positive the foundation of human life? The question here could be more, if you are allowed to be (positively) critical? Isn’t almost all criticism positive because it shows engagement and a will for change?

If you control nix, you control NixOS, no?

On community in Nix

• pleasant words, but trustworthy?

e.g.

“Fourth, the NixOS Foundation in no way controls or governs the Nix community, which has, since its inception, demonstrated its ability to self-govern very well.”

vs.

“I remain committed to creating a community where …”
“join the Determinate Systems community”

What is this community? It’s not the “nix community”, or is it?
If someone can “creating” and shaping the community, he controls what’s going on in a strong way, no?

As I understand, he argues in a formal way of some facts and positions. Didn’t he shape “everything” since the beginning in his way (with people who support him strongly), so that he still has the power without the formal positions, no? Does e.g. Eelco tries to understand the criticism that exists?
Not the words he is using but his statement sounds escalating to me.

[kampka]

I don’t think the question really is whether or not NixOS / nixpkgs will survive as a project.
There is clearly enough technical dependence and interest in the project to keep it alive for a long while.

The question really is: what will the community look like when all the dust has settled?

My feeling is that the crux of the matter lies in the fact that we are spending our time debating how to solve “the problem(s)” without recognizing that we cannot even agree on what those problems really are.

So I guess I’ll add my two cents to the issue as I see it.

Maintainers quitting on the project: Yes, it sucks, but it’s also something that just happens.
People quit projects for all kinds of reasons, often intently personal to them, and any project with any plans for longevity needs to be able to cope with it. I’m not saying we should take the resignation of maintainers or any contributors in protest as something to just brush of, by all means.
Their concerns should be heard, we should thank then for all their hard work and wish them all the best without malice. Furthermore, we should strive to make this a community they may eventually return to, if we can. It is, however, an issue that we can fix. And must fix. But in the end, fixing it is a technicality.

The foundation and it’s leadership being disconnected from the community falls into the same category.
It’s an issue for sure, and one that keeps on giving while it exists, but in the end it boils down to an organizational matter. One that can be fixed, and relatively easily as well.

A moderation team that is chronically overworked, being too lax for some while to draconian for the other, all the while loosing the trust of at least parts of the community, is also a technicality that can probably be fixed through relatively simple means.

So in my opinion, these are all just issues, roadblocks that can be fixed along the way.

They will not cause the death of NixOS by a long shot. Call it implementation details if you will.
So whats the problem then if there is any?

The problem is that in the wake of all of these issues, people are loosing their civility.
People are shouting at one another, calling each other names. Emotionally overloaded words like
“right-wing”, “leftist”, “woke” are being used, people are asked to kindly “just fuck off”.
In less well moderated spaces, people are talking about factions, warning their respective listeners
of the “networks” and “conspiracies” that exist in this community.
Then of course the trolls are showing up, fired up by the fury of their respective echo chamber,
raveling in the joy of seeing us argue amongst ourselves instead of actually doing the work.

Recent internet history has shown us that we need close ranks and do it quickly.
Once we give this poison a chance to take hold in our community, we will never get rid of it again.
If we fail, NixOS might not be dead or dying. It’s community will also not be dying.

But it will also not be dying in the sense that Twitter is not dying, or that Reddit is not dying.

For me personally, that will not be enough. And from where I am standing, we are not doing a great job at the moment.

[Aleksanaa]

If you control nix, you control NixOS, no?

This is not the point, either good or bad way. I’ve explained it again and again, explained this problem from various aspects, to letting you know that this is untrue.

And his article is indeed misleading, but it does not mean that these remarks are really true. If you were in the audience for his business, you might think “that’s true!” and continue to give DetSys your faith, but you’re not, so why take this article seriously?

[Aleksanaa]

Thank you for saving my cyber saliva. Yes, you pointed out some real issues.

[7c6f434c]

@igel

If you control nix, you control NixOS, no?

Nix team sometimes has discussions to the tune of «what could we do so that Nixpkgs considers our stable release trustworthy» (Nixpkgs has long known better than to use the latest stable release of Nix as nix-stable, intentional support for old versions makes new features less relevant and bug risk more relevant). It’s complicated in all directions.

@nat-418

Your framing is what is misleading: marsam, to take one example of a contributor who left, was the #4 top contributor to nixpkgs.

Nixpkgs is used to such losses. Even more so if you include people dropping to not-exactly-zero.

@igel

I couldn’t read apologies for things that happened, for behavior shown and ideas how to make sure it will not happen again.

I am not sure Foundation can make sure anything doesn’t happen again, if after saying «this is the interim policy which does nothing, please propose a real one» they got no proposals in response, and in the second round… actually, still nothing, because a procedure for vetoing is not a policy that can be interpreted ex ante (I do not say that a policy cannot leave space for judgement calls, but if it has nothing but judgement calls, it is a procedure but not a policy). But this time Foundation at least has a chance to apply something.

Do I think we have efficient and fair decision processes? No. Have we ever had them? No. Was it even worse before? Yes.

@igel

e.g.

“Fourth, the NixOS Foundation in no way controls or governs the Nix community, which has, since its inception, demonstrated its ability to self-govern very well.”

vs.

“I remain committed to creating a community where …”
“join the Determinate Systems community”

What is this community? It’s not the “nix community”, or is it?
If someone can “creating” and shaping the community, he controls what’s going on in a strong way, no?

We have two references to communities. One is called «Nix community», the other «DetSys community». Things are said about each, and these things make no sense if they are stated about the same community. These are two different communities. Yes, both have interests related to Nix, Nixpkgs, etc. Yes, they intersect.

But why is it popular to declare that obviously they must coincide?

@AngryAnt

Theoretically, given the above structure, how doable would it be to jettison the current foundation, construct a new one, and perhaps shift to something like the rust reimplementation of nix?

What is really tied to the Foundation is the binary cache, and, to a degree, Hydra. A new foundation would need to convince sponsors that it is better to donate build capacity for Hydra replacement to them, which will be hindered by reluctance to abandon the old infra because of the cache.

Providing NixOS images with non-CppNix implementation (as a secondary option) is a lot of work but hopefully with limited coordination issues. Making them the default ones… well, that’s an RFC with two-digit number of months to settle the details, after the images apprently work fine.

Rebranding is obviously also a coordination issue, although lesser than with a buildfarm, as a renamed fork can have a compatibility promise and so this can play out over time. Long-term forking of community events with grudging collaboration on Nixpkgs as a shared resource… I do not find the energy to be sure if it would be better or worse, at least it would force us to be honest that there are many communities around Nix*, with complicated intersections and interactions, and no sizeable project is obliged to fully coincide with a single community out of them.

@Aleksanaa

Regarding the issue of “representation” or “endorsement”, if we think who should not represent us, we should actively solve this problem, instead of directing the problem to the remaining majority of “represented” people. It’s no use.

If we think who should not represent «us», we should also come to at least of statement of who is «us»… I have seen some divergence in opinions of who would be represented.

[chrism]

To the extent raising a hand here matters, I will commit volunteer to maintain some nonzero number of packages in nixpkgs, at least ones I use personally, but not under any regime that kicks out its release manager a month before release for reasons unrelated to competence. Wherever that regime isn’t is where I will volunteer to fill in gaps.

[ashkitten]

@chrism

kicks out its release manager a month before release for reasons unrelated to competence

I am concerned with the implication that there can be NO valid reason to disallow someone from participating in the community aside from incompetence.

[paperdigits]

@ashkitten

disallow someone from participating in the community aside from incompetence

I don’t understand. If someone is not competent (I assume technically here, but that is unclear), then shouldn’t we work to make them competent?

@chrism

regime that kicks out its release manager a month before release

That regime is already here and now.

If you find Jon’s reddit post, you’ll see he admits he is not innocent, he accepts responsibility, but if you continue to read the replies in that same thread, you’ll see he hasn’t actually learned anything nor reformed his behavior. If he says he is guilty, then what is left to discuss?

[Aleksanaa]

We’ve got enough of Jon Ringer vs Delroth (vs moderation team) here. Can we put aside the controversy and explore strategies that actually help solve the problem, such as fairer and consistent rules? If we don’t, are we going to have to stage a weekly wrestling match (and then kick the loser out of the community)?

[nim65s]

NixOS is not dying

I think this is more about nix community dying. I’m talking about the first of the two from:

@7c6f434c

We have two references to communities. One is called «Nix community», the other «DetSys community».

Many of its members already left, because to them this community is dead. As a matter of fact, it is not: see, I am still posting a comment here, so, yay, life.

But I think they can not see any future for it. https://github.com/KFearsoff/nix-drama-explained seems to be a pretty good explanation about that. The current Jon vs delroth case doesn’t matter a lot: what matter is they think they tried everything to find ways to deal with the whole class of similar cases, and everything failed.

I would be super happy if anybody could find a solution, maybe one that nobody saw previously.

But, to me, On community in Nix is sparking the opposite of confidence: it looks like eelco is actively not looking for such solution.

[Aleksanaa]

Many of its members already left

Now I’m very curious about it. As we all know, not everyone who leave will immediately remove their maintainer handle, and not everyone will say “I’m leaving” with a loud voice. So apart from some members we can see in a few ways (20 as I have seen at this moment: Issues · NixOS/nixpkgs · GitHub), we may need to obtain GitHub’s contribution data, matrix’s message frequency, and discourse’s background statistics to determine whether our diversity has indeed declined in an obvious way.

I am in an almost unclear state on this matter. I don’t think that many people were actually involved in the quarrel. There should be many people who are aware of the current situation. There may be many people who are still unclear about the causes and consequences. But in fact, I don’t know how much impact this has.

For me, I still need to speak. I am currently maintaining 64 packages. I’m checking repology and zh.fail almost everyday so I can say I’m really “maintaining” them. I don’t care if this number increases five times. It may increase five times in one or two years under normal circumstances. I don’t think this conflicts with “having a life” (I have met some famous Arch Linux contributors a couple of times, and after careful observation, I really don’t think they are the kind of person that everyone is guessing to be, but their number of commits is almost comparable to r-ryantm) (If you know who I’m talking about, don’t tell them!)

[paperdigits]

The positivity and optimism this post is trying to bring is very welcome at this juncture. Thank you @Aleksanaa

[ibizaman]

Thank you for the post. I fully agree with the sentiment. Everyone I talk to about nix is really enthusiastic with how the project works technically and how well it solves a bunch of problems we have. I am too!

Now, I do not like to voice my opinion on a topic I cannot have an opinion on. On the contrary, I do not mind giving my opinion if I have the necessary background. I wonder how many reading this feel the same way. Anyway, in this instance, I clearly do not have the background necessary to voice a meaningful opinion but I would like to.

Is there some place explaining the facts behind all this? A post detailing what happened and that was verified by the community? Currently it seems everything is scattered all over the place. Is this link the current most accurate post GitHub - KFearsoff/nix-drama-explained: This is a repository that aims to concisely explain the issues in Nix community

EDIT: clarifying a bit. I meant I’m willing to voice my opinion on RFCs, do the work to help the community, etc. As long as I feel I have a good and fair understanding of the issue. I didn’t mean just « raising my opinion » for the sake of it.