Can someone summarize, to someone who hasn't partaken in any official Nix channels, what the catalyst of this was, if any? Or at least what sorts of actions/events have caused this discontent?
The longer boiling tensions that have led to this point:
- Flakes. They're largely promoted as the solution to composability but don't handle all use cases, and this has caused tensions with those that prefer more flexible but less ergonomic tools
- The decision making process that led to flakes being a thing. There was a community process established to replace the previous de-facto BDFL process with a community driven process, and then the previous de-facto BDFL merged flakes in and started promoting them without following said community process. Later attempts to engage with the community process have had the issues where flakes now have a lot of genuine community support, but also some resistance from core contributors due to either the genuine feature gaps, or the fact that they're now presented as a fait accompli.
- Moderation. There was a significant contributor who was up to some hateful conduct in (mostly) non-Nix community spaces that was banned across all project venues. There were groups unhappy that they were banned, and groups unhappy that they were not banned quickly enough. The project tried to navigate this with an attempt at middle grounding their moderation policies, which annoyed both groups.
- Corporate influence from Determinate Systems: Newer initiatives from the former BDFL have mostly been announced through the company they founded Determinate Systems, and include stuff like a centralised repository in the form of Flake Hub. You can either see that as "proving it in the wild before bringing it to the community process" or "bypassing the community process to make it a fait accompli", but either way the ex-defacto-BDFL seems to see the community process as too slow and is end-running it with his new company.
- Corporate influence from other parties: DetSys is getting the most hate right now due to employing the ex-defacto-BDFL and still having enough leadership positions to be seen as responsible, but other groups like Tweag or Cachix have also been seen as attempting to commercialise parts of the ecosystem at time.
* September 2023: The "Nix Community Survey 2023" is looking for gender data, and the mods don't like that most contributors are men.
* November 2023: The moderation team tries to institute a Code of Conduct https://github.com/NixOS/rfcs/pull/114 ... and they get their way
* November 2023: Some are not happy about it: https://discourse.nixos.org/t/moderation-team-accountability... -- the moderators talk about their "authority" and of course lock and hide the thread. It's "disruptive" and "off-topic", you see.
* This sort of activity continues -- moderators consolidating and increasing their power, citing how they need the power to control "concern trolls" and such -- and now in April 2024, we get https://save-nix-together.org/
The "anonymous contributors" want to drive out the NixOS founder entirely, so that _they_ are in charge. They want "to hold people accountable for bad behaviour at all levels" and lament having "responsibility without authority" - in other words, they want power, power, power. They want power over everyone. Their justification is that they believe they have the moral high ground, and they deserve to lord it over everyone else.
Hold onto that hard power, Eelco, and tell this lot to fork the project. Let's see how they enjoy moderating noxious.org instead of nixos.org
The only conclusion I can draw, as to why an author would choose to write only in generalities, is that any concrete facts or statements would be weaker under scrutiny, or less well received, than the uneasy benefit of the doubt that we extend someone making vague, unfalsifiable claims.
I don't want news I read about groups I'm a part of to be vague to the point of confusion either.
I should be clear: my original thoughts when reading it were exactly the same as yours.
Right so the author of the post in question, is in favour of Nix taking sponsorship from Andruil, or against it?
They are partially writing an opinion piece, but not about their opinion about working with the military. They are writing about Nix leadership vs Nix community.
https://www.theregister.com/2023/09/08/nixcon_drops_anduril_...
There was also a longer discussion on HN 7 months ago: https://news.ycombinator.com/item?id=37418351
The reason that Anduril was dropped as a sponsor for the 2023 EU conference was that the venue wouldn't allow the association. This time the NA venue did not have that policy, so the sponsorship went through.
I mean they even have their own independent build system?
There are also concerns with regard to civil liberties such as spying on their customers out of self-interest and on behalf of the US government, cooperating with authoritarian regimes such as China or their participation in Trusted Computing. Also note that while nowadays, just like many other tech companies, Microsoft might get criticized as being 'woke' by people with a certain political outlook, in the late 2000s, they were throwing people off xbox live for mentioning being gay in their profiles (or just having the surname 'Gaywood', for that matter).
It's truly a mirror of the culture war you see in other parts of society where people declare things they like value neutral and things they don't like a war on culture.
Didn't Andruil get sent of after a while anyway?
Why don't the do-gooders just fork Nix and do EthicalNix or whatever, and lead by example instead of trying to take over the existing structure, wouldn't that be more in line with open source ethics, and you know common courtesy? Maybe drop github and Microsoft due to ethical principles while they're at it?
At any point in time there are always multiple entities claiming to have a superior morality. There is also always an entity with absolute power. So both points are moot and this question is moot as well.
> Didn't Andruil get sent of after a while anyway?
I would assume people like a project that is lead in a way that the community doesn't have to make ad-hoc interventions to undo decisions they think are bad.
And this is where the power dynamics come into play. There is leadership that assumes it has the power to take certain sponsorship. So the community does not want that entity to have absolute power, it actually wants that entity not to have power. So now there is a power struggle for absolute power. The community would like to get absolute power so it can just make decisions about military sponsorships beforehand. The current power holders want to keep absolute power and take whatever sponsorships they want.
I don't think open source ethics prescribe you have to fork a project instead of taking it over. Common courtesy from the current power holders would be step down, or common courtesy doesn't actually mean anything "you know".
It's up to whomever will be in power to decide where the line of ethics is drawn. I certainly hope they would consult a broader audience to see if they line up with the community. If you think github and Microsoft should be dropped as well I would implore to make that case.
It's interesting that you think there are other unethical organizations involved with Nix while at the same time arguing that the current leadership should remain. If I think something is unethical I argue against it instead of for it. In case I find myself on the unethical side of an issue I try to see if there is a higher ethical standard that I hold that trumps that one instead of arguing that the unethical thing is good.
There is no such thing as "community" when talking about power. Certain people will have power. Maybe wider community agrees with them. If there is an established way to transfer this power (and install a different governance structure), then they can use it. If there is no such way, they may ask edolstra to step down, but surely he is not obliged to do so.
Historically, forking was a way for open source communities to diverge when they had incompatible values. There are plenty of examples of successful forks, preferred by community.
A teeny little bit of me thinks, that people who are not happy with current leadership do not want to fork the project because they are not sure that wider community cares enough to follow them.
Recently I heard that there was some drama about politics, and that might serve as an excuse for someone to step in and take over the organization, I don't really feel like giving these people attention however, so I haven't really looked into it all that deeply.
The first paragraph in this very post is just pure rhetorics without substance.
I suppose the point of the post is that whoever is behind kilo.bytesize.xyz should become the benevolent dictator for life of the project, due to pleasant sounding platitudes, or something?
The onus should be on the article to convince people that something is happening and what it is. Where there's smoke, there may just be someone blowing smoke.
As an outsider myself, it certainly looks like something is going on as more than 20 Nixpkg maintainers left in a week: https://github.com/NixOS/nixpkgs/issues?q=label%3A%228.has%3... It's fine to ask if someone can explain what is going on, it's a waste of everyone's time to say you have no clue and you haven't read anything and it's probably nonsense.
He could have for instance have used that link to verify, and I wouldn't have been thought that it was weird that he said that people were fleeing, without providing proof.
Since this post likely is geared towards engineers and scientifically minded people, and he is in fact calling for people to resign, I suppose so that he or she can take over, is sources and data really to much to ask for?
If your calling for people to resign so that you can take over, which seems quite audacious to begin with, at least you should be able to clearly explain, where the people have failed, so that they can defend themselves.
The purpose is to warn people so that they not fall for what sounds like really nice things, but that in effect just is nothing but empty words.
Which I think is fairly evident if you read my _whole_ comment.