Nix: The Breaking Point
kilo.bytesize.xyz
kilo.bytesize.xyz
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?
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).
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.
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.
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.
* 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 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.
They're big balls of mud, exacerbated somewhat by nix's tendency to encourage inline expressions and shell scripts. This reintroduces a need to mentally interpret intermediate values, conflicting with easy audits - a big reason for declarative packaging in the first place. I've avoided adopting nix given that I didn't see a clear path towards cleaning up these kinds of packages.
I wonder if this community shakeup could be an opportunity for a more literate and prosaic format for nix. If we're booting some of the long time chooms on account of them being socially regressive, what would you bet that they're also the reason for nix's seemingly "non-negotiable" technical debts? I guess we'll see!
Here's hoping the fork will be a nicer place with cleaner codes.
All the posts that appeared recently on Nix have the same vague, over-politicized language where they are complaining that the person with most potential as a dictator is not one and they want more power to make decision in behalf of nixos.
I feel for Dolstra, pain in the ....
- apparently the current rupture came from a company sponsorship to a Nix Foundation conference, due to the fact this company (Anduril, a heavy NixOS user and contributor) is a mil-tech one, well... I doubt that's the reason. I know many with strong opinion to a point of leaving if some of their principle is touched or even discussed, but that's can't be a community-wide cause, nixpkgs is one of the largest package system of the world, hard to believe that it's community polarize around such event;
- IMVHO the very issue is that Nix{,OS,pkgs} is a FLOSS project but itself it does not reflect a FLOSS one, having some "central" elements (the cache, github infra etc) that are more a commercial structure than a community owned one, and there is a big dispersion in current development: NixOps seems to be abandoned, Disnix is not a complete replacement, Flakes seems to be the future, but they are still experimental end not much documented, home-manager is still a separated things and so on, the hard-to-digest nix language does not help.
Long story short it's not a matter of mil-tech, lack or abuse of authority but the fact that:
- a commercial project have targets, a vision, all party involved do what they can to implement/reach the defined goals
- a community project have a multitude of visions and targets, it change a bit a time, sometimes fast, sometimes slow, with many branches, sometimes competing sometimes cooperating,
Unfortunately, to go the commercial project route we need a hierarchy, contracts and so on, to go the community project root we need a community-matching structure, central points, with people owning them do not works well.
Now, having no official repo, build system etc is hard, some have proposed some tools, but they do not develop much and they do not perform much. I think the point is looking at a more classic FLOSS approach, an entity do the commercial aspects with an official release, while the larger slice of the community create independent hubs, there is not need to be all in the same place as long as the codebase start openly and openly evolve.
Shared binaries from /nix/store in a torrent-style setup might be feasible, but so far there is none. Similarly distributed build for large projects (LibreOffice, Firefox etc) might be possible, but so fare there is none. There are servers dedicated to that, paid for that. Not-so-small companies could pay. Public universities as well, but in the present world public uni do not even mirror anymore FLOSS projects, so... We need a distributed architecture to keep up FLOSS.
Wouldn't the thing be that the reason that they needed to get sponsorship in the first place?
Maybe it's just not possible to run Nix without getting money from people that are involved with murder, in one way or another? I mean there are plenty of people who wants to boycott banks, and even countries for this that and the other?
.. or why don't they just spend their energy on creating a torrent-style setup, instead of complaining about what the guys behind NixOs has done successfully so far?
At some point in time this disappear, maybe when most actors decide to ditch their own operation for someone else "the cloud", and as a result there are no spare resources to support FLOSS. Similarly the FLOSS community became thin, quarrelsome, not really supported etc.
This is why many say the FLOSS ecosystem is dead because there aren't anymore FLOSS model contributors. The current state of FLOSS is mostly big corp show embracing the model because they need pre-existing sw developed and licensed this way.
Personally I have nothing against mil-tech sponsorship, I'm deeply against the absence of a FLOSS model. The Ops-less state of things, where no one own it's infra, have iron and so on, is a terrible vulnerability waiting catastrophic detonation and Nix/Guix success and failures are in the same line: most people get convinced that's normal to have a system based on paravirtualisation, they see nothing strange in downloading pre-built docker images for anything, run k8s on a single homeserver and so on.
We are loosing knowledge trying mimicking GAFAM model witch is archaic and anti-users, and it's built especially for that, to lock in people, these days even mentally, since so many see no vulnerabilities in using for instance GitHub not as a mere repo host but for PR/CI and so on, even if we have witnessed bans, massive geo-bans and so on. We have a big slice of people unable to visualize the dangerous IT mess we are in.
Nix community in storm is just a symptom.
If one were to fork the project, the sensible way to do that would be to build a nixpkg-compatible alternative system that doesn't have some of the design flaws (such as a global non-discoverable configuration namespace and an underdocumented custom functional programming language...).
I don't think anyone's gotten together the hours to do that. But I'd certainly love to build an immutable, repeatable system using a Python or Kotlin config myself.
So given those assumptions, why don't they fork it?
Also when you say that the value in nix lies in the packages what exactly do you mean by that? Do you mean the totality of all the prebuilt and checksummed packages? If you where to fork-fork, and not rebuild nix, wouldn't you still be able to use them? Apart from what is built afterwards?
I think the people who think the nix language is the valuable part are trying to save it, not fork it. But that's a generalization.
A fork would be able to use nixpkgs even if it didn't use nix-the-language itself, with some careful planning for migration.
It's a bit like long form subtweeting.
Remember the Rust polemic where the entire Rust moderation team resigned? I was following the debate at the time and could barely understand what the actual problem was... only after reading A LOT of arguments in multiple websites I was able to kind of understand what was going on (here I am being vague myself, ha! If you know what that was all about, please comment, I don't feel like I can do it properly).
What is this all about? Is it a particular person, or organization, being a troublemaker?? Is it one event? Many?
Do you think your project's internal politics are some kind of "state secret" or what?? And by the way, why do Governments have so many secrets anyway!?!?! You've got to be up to no good to want to keep everything bloody hidden!
For goodness' sake, just spit it out!! We can handle the truth.
Over the weekend I’ve been working with NixOS foundation members and other community members to address what you’re describing, and I’m happy to say that the foundation is very willing to delegate governance to the community!
This kind of thing takes time to erect, but I’m doing my best to push this swiftly while trying to limit burnout. Stay tuned for a statement from the NixOS foundation this week!
"The ad-hoc structure of maintainers here better represent the goals of Nix users and their methods of success. Yet, these contributors are beginning to flee."
It also makes the claim above, is there any truth to that? Is there evidence of contributors fleeing, and if so where? What is stopping them from forking the project?
Suggest clicking through since the URLs are not nearly as clear as the content:
https://www.anduril.com/article/altius-700m-live-fire-test/
ALTIUS-700M brings unmatched payload capacity to the loitering munitions market, combining multi-domain launch, proven collaborative autonomy, and category-leading range with a larger, penetrating warhead to kill or disable armored targets and infrastructure.
https://www.anduril.com/article/anduril-s-lattice-a-trusted-...
U.S. Customs and Border Patrol, for example, uses Lattice for land and maritime border security to autonomously identify, track, and detect people
https://www.anduril.com/article/army-selects-anduril-and-pal...
TITAN is a vehicle-mounted expeditionary ground station that will accelerate and simplify the Army’s ability to access and process massive volumes of Space, High Altitude, Aerial, and Terrestrial sensors to provide actionable targeting information for enhanced mission command and long range precision fires. TITAN will reduce sensor-to-shooter timelines by enhancing the automation of target recognition and geolocation from multiple sensors [...]
"and long range precision fires"
What happens when a substantial amount of contributors come from China (which is common given how many developers are available in China)? Are they going to start restricting who can contribute? Or have strict vetoing of non-US nationals?
) And incompetent or lazy actors :)