I don’t really know what the original post said, because it is hidden, but for a few months now, maybe a year, I’ve seen a constant stream of random drama in the Nix community. I’ve mostly watched anything happening from the sidelines, because I never feel like dealing with any of it, but it is pretty annoying to me, because I honestly feel like it undermines Nix(OS).
For me, the practical impact is what’s most frustrating. A lot of energy spent on fighting just seems to paralyze the project’s governance. Progress slows down, and in some cases, it feels like things were getting worse, not just standing still. The fact that forks started popping up, and the community itself started to fracture was sad.
I can understand the impulse to try and improve the world, even if it’s just in the small space you have some influence over, like the Nix community. But I genuinely believe this isn’t the right venue for fighting some of these larger, systemic battles, like trying to control which corporations can or can’t use Nix, for example. These are problems far bigger than any one software project can solve. Nix should be a software project first and foremost. In my honest opinion, the focus should be on making it the best it can possibly be. That’s the tangible thing we can realistically achieve, but we can’t fix the world with it.
The ironic thing in all of this is that I’m a member of a left-wing political party, and I probably hold more radical views than many of the people in these discussions. But that’s exactly why I know this isn’t the place for it. For this project, I would choose pragmatism every time.
Approximately nobody is fighting over this example question.
The two recurrent sources of drama in the community for the past couple of years are the following:
Whether a corporation is trying to exert undue influence on the culture and/or technical direction of Nix, and what measures should be taken to prevent this.
Whether a ‘Nazi purge’ or ‘woke purge’ is happening and/or should be happening in Nix.
Fixing the world is not really on the agenda. Both classes of question are about getting our own house in order. There are simply conflicting visions for what ‘in order’ means, and a lot of hot-blooded takes muddying the waters.
To be honest, this doesn’t feel correct to me at all — I don’t know anyone who thinks Anduril is trying to participate in the community in a way that is distinct from how other companies participate. E.g. ignoring who is doing thing these thing, sponsoring conferences and paying people to do various open source projects are perfectly normal behavior we see from other companies doing Nix, without controversy.
(If anyone disagrees, I would like to hear it.)
The controversy is because Anduril is Anduril. Why this is bad, in turn, I would split into two camps:
Some people think Anduril is bad purely because it is a weapons manufacturer. They don’t want weapons manufacturers around, whether they are Anduril, Lockheed Martin, Rheinmetall, Fabrique Nationale Herstal, or the Aviation Industry Corporation of China.
Other people think Anduril is bad because it is a Thiel company, and, separately from the actions of Anduril itself, it is an objective fact that the “Thiel verse” as a whole is very active in (far right) politics. (For example, is J. D. Vance a public figure of the same provenance with his current views in a world in which Peter Thiel never existed? I think not.)
This again feels like a bit of a motte and bailey. Yes, no one has explicitly said “it is the job of the Nix community to fix the world’s problems”. But from the perspective of “things are political, like it or not” — which is definitely common around here — both allowing Anduril and denying Anduril are both political actions, and both will have an effect on the broader world.
From that perspective, it is not “mission creep” to ban Anduril, but simply “doing our part” given larger conflicts, and where a neutral action does not, in fact, exist. In this worldview, neutrality is a mirage of an alternative, since a neutral action does not in fact exist, and “Nix is orthogonal to all this” is either naivety from those that don’t want to think about the consequence of their actions, or worse, a bad-faith argument from those in secret outright support of Anduril/war/Thiel/etc.
All of the above is supposed to be not by own views on this subject, but an fair and faithful analysis of the arguments against Anduril.
(My own views I discussed recently on Mastodon; You, dear reader, can find them there if you so want to.)
I will say that I am generally blissfully ignorant of the political views of my fellow SC members who voted for the recent moderation changes. We 4 are also, if I recall correctly, split evenly 2 vs 2 on the “Anduril vote” many months ago. (That is, the vote on whether to allow Anduril job postings on discourse.) I take this as good evidence that the moderation votes were not at all part of some secret pro-Anduril or pro-right-wing agenda.
I can say with confidence that none of us 4 want anyone banned or otherwise disciplined for their political views. But we all were concerned that some people were not getting disciplined because of their political views. We don’t want political views to be grounds for either extra harshness or extra leniency.
(To be extra clear, this was an outcome we were concerned about. Whether that outcome was the explicit and conscious intent of the moderators or not, I care little about. The whole point of the concept of “bias” is that it is not conscious and not intentional.)
I would thus say with confidence that no “Nazi Purge” or “Woke Purge” is on the agenda of anyone currently part of governance.
/* just passing by, didn't read the rest */ Well no, it’s 2025 and by now it’s just funny to not acknowledge that, all other questions aside, drama (that can be contrasted to) is part of the Аnduril brand. Like, maybe they didn’t even mean it in 2023, but they certainly embraced and developed the image after seeing our reactions, reactions of reddit and hackernews, et c., inspiring their hiring strategy from Discourse posts to the “don’t work at” video.
The teasing branding is definitely real, but I think that is a distinct from their formal interactions with the community that I was discussing.
For example, just because they made that video, doesn’t mean if they sponsor an event would they insist that they be allowed to play video on loop at their booth.
i think this characterization is kind of fair, tho i’m inclined to maybe add a few:
the MIC distinguishes itself from plain commercial arms manufacturers in that they have political clout they could use to further steer politics toward armed conflict for their private gain, as alluded to in Eisenhower’s farewell address.
the incorporation of AI into drones to make for autonomous weapons marks an escalation in the arms race, steadily increasing the amount of damage that may be inflict on one’s enemies (going by recent conflicts, increasingly civilians) by those who possess these (a concern aggravated by the political trends mentioned by @Ericson2314).
the US has a history of foreign interventions, while they have recently been alienating their existing diplomatic relations, while actively supporting what may be considered a pariah state.
I can only speak for myself, but all my recent noises were to that effect. I honestly wish tech projects could be truly neutral — as in, not just a veneer for “maintaining status quo,” but actually leaving all this magical thinking crap outside the door and work together on the code. Yes, things are bad if you look beyond your screen, but they should also be fixed beyond your screen. I don’t imagine all that many people on the “against” side share that position, but yet here I am. And yes, as such there were aspects of moderation and other things that I didn’t like, but running an institutional “Woke” purge in NixOS like DHH et al. is not the solution here, just dragging things way over the other side of the fence. I’m not saying it is already happening, but may happen and as such, I’d like the NixOS governance to effectively guarded against such nefarious schemes.
On the surface level, sure. But tacit participation of the founder of the aforementioned thread does raise a bit of a red flag — it didn’t cost much to say “meh, not interested, we got our own thing going and have drones to ship.” And given their favourable opinion towards the current administration it’s not unreasonable to think they might agree, even if they have not stated that out loud. So in so far as their participation in the community via donations or such I was never that much against, as long as line is drawn on political endorsement and there are mechanisms in place that would guard against them getting any more influence, should they want to.
I’m more on this side. Even if in ideal world we would not have to prosecute wars, that’s not the one we live in and as such, weapons are necessary and you have to work on their R&D to maintain the edge. But being in agreement with people like thiel, trump and vance is them consciously agreeing to work towards making the world worse. We can’t stop them from using OSS code and I don’t even think we should refuse purely philanthropic donations with no strings attached either or cut them out of the community on principle — even if I personally think that world would be better off if companies and people like that disappeared off the face of the Earth — but we should have good safeguards that will not let them inject more of their inane PoV of how things should be run into the community, possibly even ultimately subverting it. Which is why I think it would be safer if, as a temporary measure, we didn’t allow people associated with that company (or other potential hostile actor) run until safeguards are in place. After that, I’d personally let them have their fun in the sandbox within those constraints.
So all in all, at least for me it’s not about improving the world as a whole — we can do that IRL — but about insulating the project from outside politics and hostile takeover to the extent feasible. Yes, yes, I know everything is to some degree political, but IMO we as a project should strive to only act on that if external politics directly threatens the well-being of the project — for example I wouldn’t be for any project-sponsored diversity outreach initiative, but push back against the dumb anti–„Woke” crusade currently en vogue from encroaching on the project to push people like that out. And similarly, we should ensure that the project has guardrails that make *nduril and the like taking over as close to impossible as we can, but other than that I don’t really feel we should cut them out from much (I guess giving them free advertising could be out of question if we collectively decided that’s against our values).
It’s (to me) not drama for drama’s sake (I didn’t even really participate in much of previous dramas), but a drama to end all dramas — famous last words, eh — because IFF we can set up a safeguard system we can all trust will ensure self-sovereignty of the project, then it will reduce the need for going up in arms over each vaguely troubling thing.
Insulation from politics gets a bit hard once a “politically controversial company” (Anduril in this case) offers sponsorship and you have to either accept or explicitly refuse. That’s how the worst situations actually started, at least from what I noticed and remember.
Moderation is also a major pain point in this respect. When discussion does become political and too heated, moderation is very likely to get under political pressure – accused either of inaction or of siding with someone, etc.
I think there may be a lot of collective unconscious memory of Oracle and Sun here, or at least of Sun and OpenSolaris. I’d encourage looking up the “OpenSolaris Missile Crisis” involving Sun playing puppeteer with OpenSolaris:
The community was deflated; OGB composition became more and more Sun dominated, and OpenSolaris was in the doldrums for the next three years.
(In a way, we have an actual “missile crisis” going on…)
I think the CoI rules at least offer a partial answer here, assuming it’s for actors with similar incentives to pack the governance board from one company, though there are ways to break this guarantee with collusion still.
My intuition is that transparency is important here… all major teams’ meetings should ideally be publicly legible. If there’s too much protocol switching into private minutes, people can ask why. If people are worried about a team being taken over (even by cross-membership on another team), they at least have notes that can show the temperature of that team in public.
In a low trust game, a lack of information is met by “assume the worst.” While there are still ways the approach can be gamed, at least it would have to be gamed in writing, and added transparency is better for everyone so that trust improves over time.
Maybe this is a reason why these governance issues are more systemic. And accountability as a means of building trust is probably a good idea anyway when there’s a periodic turnover of people who have to relearn how to work with each other in a healthy way.
Regardless, this is a “bigger than any one of us” kind of problem, but hey there are some thoughts.
For sure, but at least for me the solution is ensuring that it can’t threaten the project self-sovereignty nor use the project to advertise values we agreed do not represent us. Of course that’s not going to satisfy everyone, but I think it’s at least the bare minimum we should have, that we currently don’t.
Yeah, that’s why I think it’s not enough, not every conflict of interest stems from having the same employer. Different employers may decide to collude together to install people amenable to their goals in the SC and it doesn’t guard against that for example. Or as someone pointed out, from the same project team — which is less of a problem, because at least the project is then being biased towards project-relevant goals, but still might be an issue if it causes other aspect of the project being neglected by comparison.
I believe we do have good mechanisms for the sponsorships now. (we had none when this first happened) SC represents the community roughly proportionately, as defined by elections. And then SC decides whether to accept. Of course some people may still be unhappy, because not everyone has the same opinion, but I don’t think we can improve this too much anymore.
@jaen so on a personal level I actually agree with much of what you are saying. I just feel:
we do in fact have those safeguards, like how SC members must work for different employers. That and more meeting minutes transparency, as @numinit says, I think will do the job.
while many more professional people at Anduril would love for them and us to have a better relationship, Palmer doesn’t have much day-to-day responsibilities and is more an unrestrained shit-poster. On some libidinal level, I’m sure he gets great joy out of our community trying to kick out Anduril, and his corresponding ability in threads like that with DHH to successfully push our buttons.
I am not saying these people (like DHH, Palmer Lucky) would not be a threat if they did have unrestrained power over the project. E.g. I would expect to say the same performative trashing of any CoC that has happened with DHH’s projects. (Side note, all 4 of us that voted for the moderation changes love having a CoC.) But seeing that they emphatically don’t have that power, I think the main effect of our anti-Anduril controversies is:
Free marketing to them
The trolls are well-fed
free propaganda of a “look, the left is also illiberal!” variety (when the objective fact is that all rich country democratic backsliding has happened under right-wing populists, not left-wing populists, so far).
I think nothing would cramp Palmer Lucky’s posting more than use allowing Anduril to participate, with strong safeguards and requirements for bland, apolitical behavior within Nix community spaces, just as is applied to every other company participating in Nix. I think this would make us look really good and mature, put the ball back in their court, and stop giving them a bunch of free press.
OP of this thread called @nyanbinary a “nonperson” on github, I don’t think they’re engaging in good faith and it may be better to lock this thread.
edit: there is some conversation that seems of benefit going on here far removed from what OP might be doing, idk if you can split out part of a thread?
And then they’re going to be all like hurr durr they be excluding me for no reason herp derp woe is me, SMH ~~’ At least go to 4chungus if you want to shitpost or something.
Not enough for inter-employer collusion, though. Which is why I would rather see them strengthened to cover any loopholes like that.
Not yet anyway — but if we don’t take that seriously and work to build more resilient safeguards, that might change.
That’s possible, I suppose. Maybe it’s just how the drama de joure was handled didn’t give me confidence that’s where we are headed and combined with the twitter thread raised to a level of red flag raised half-mast that was enough for me to finally speak up. If ya’ll there in SC really aim to do that, then it makes me feel slightly better if no full re-election happens (though I don’t see any harm in re-affirming your mandate if that’s really what you plan to do either).
