Since the ethical concerns about potential sponsors should be considered on a case-by-case basis, it seems like the most relevant aspect of such a policy would be the process by which the sponsors are approved. The ASF Sponsorship policy explicitly avoids having such a process:
Because The Apache Software Foundation Sponsorship Program needs to apply its policies quickly, fairly and equally, without the need for exceptions or board votes, we have tried to maintain this document at the most basic level. However, the Foundation reserves the right to decline any sponsor without providing a reason and at our sole discretion.
For a different example, the Association for Computing Machinery Conference on Fairness, Accountability and Transparency has a sponsorship policy that includes a clear process for Sponsorship Approvals: ACM FAccT - 2022 Sponsorship Policy
In short, to ensure conference sponsorships advance the FAccT principles, everyone on the steering committee for the conference has a say over whether a sponsor is accepted and the policy outlines a clear process for the necessary notifications and decision-making.
Written down and longstanding policy of a venue is not just social pressure. Neither is the property of longstanding policies to come with complicated stories of actual-interpretation (even more so in Germany).
Emphasis on the “not just.” I think it is a mistake to equate the manufactured outrage by some members of the community with the fact that the sponsorship may not have been allowed anyway due to venue policy. Allowing people to manufacture outrage and push organizers around will be destructive for the community in the long term and is unrelated to the civil clause, which is the important part of the specific matter being litigated here.
If the venue policy was the actual reasoning of all the pointless outrage rather than specious concerns of morality, it may have held more water. Instead, it gives more ammunition to the black-hats who can now use this false equivalency to claim that their incitement of outrage was actually due to the civil clause (i.e. gaslighting about intent).
Of course, the other issue is the lack of a policy that would have made none of this a problem to begin with, and I agree that how to fix this is having a sponsorship policy like the ASF’s. That way there’s no question. Of course, careful about how that policy comes to exist, because favoring one group over another will cause the problem to repeat again, but worse.
TL;DR: Nix is successful enough now that we have to deal with a fair number of black hat outrage inciters. It was sort of inevitable and is what happens when you don’t have enough policy to stand as a barrier between them and things that are important.
“Groups grow around founders and rules. You cannot change the rules after the fact unless you are the founder. And to keep out bad actors, you must have the right rules.” - Pieter Hintjens (the late author of ZeroMQ)
This is a political question, not a technical one. I think it’s a huge leap, and very uncharitable to the community members in the conversation, to say that addressing a political question from a political perspective is indicative of intent to manufacture outrage or behave destructively.
I think I see where our viewpoint is different, and it’s where the boundary of “community” stops.
[edit: hit save prematurely]
My perspective is that promoting politics is kind of destructive to the community’s ability to interact with each other. It’s more useful to focus on what brings us together, which is a different set of things and more technical most of the time.
I respect your right to have that belief, but your statement implied not just that the people with beliefs different from yours are acting in a way that is harmful to ‘the community’s ability to interact with each other’, but that those people are ‘manufacturing outrage’—that their goal is the outrage, not the sincere betterment of the community through the lens of their beliefs. I see no evidence whatsoever that anyone is trying to do anything other than contribute to the betterment of the community as they see it, and I want to express my unhappiness with the disrespect you show to these community members by asserting that their intentions are malicious.
This is a political question, not a technical one.
Are we really incapable of evaluating Anduril’s content using criteria that are non-political?
I’d like to express my unhappiness that what’s clearly a technology community has people arguing about things that are off topic, and, furthermore distracting to what’s on topic - namely, technology. If people complained about the quality of Anduril’s potential talks or the fact that they weren’t allowed because of the venue’s policy, maybe that would be more on topic. But, alas.
Note that the post explicitly mentions asking the venue contacts whether this is a clear-cut issue, and receiving an answer «it’s complicated». Three days before the conference, I cannot fault the risk-avoidance reaction. Note that interpretation of the clause has settled under German politics of a decade or two ago, not US politics.
I do see some people that have marginal connection to the actual technical work in the project on both sides. However the majority of arguments (representing, in my opinion, the whole range of arguments) are from people who have made technical contributions and care about the project. That doesn’t make me believe some of the claims make sense (I myself don’t buy that the project of this size does, should, or can have a community unified by values as opposed as people-working-on-it unified by scoped technical interests…), but it does make me believe in good faith.
Impact of fundraising decisions on other fundraising and on non-monetary contributions inherently includes political considerations. At the very least, it includes considerations of political reactions of actors contributing.
Note that the post explicitly mentions asking the venue contacts whether this is a clear-cut issue, and receiving an answer «it’s complicated».
Make no mistake: I’m actually very impressed with how it was handled given all that - namely, in a way to minimally disrupt the conference, keep the talks, and give the community time to “cool off.” It is a demonstration of professionalism from the organizers.
That doesn’t make me believe some of the claims make sense (I myself don’t buy that the project of this size does, should, or can have a community unified by values as opposed as people-working-on-it unified by scoped technical interests…), but it does make me believe in good faith.
I think it probably mostly is. However, the disproportionate reaction of trolls, both pro- and against- Anduril, in the other thread and elsewhere should help us consider how to encourage “cooler heads” in the first place through discussion of the issues, rather than morality trolling, which no one can agree upon.
Impact of fundraising decisions on other fundraising and on non-monetary contributions inherently includes political considerations. At the very least, it includes considerations of political reactions of actors contributing.
Something that would better be clarified with a sponsor policy rather than ad-hoc.
It is true that the common denominator in the community is primarily technical, not political. This makes it easier to focus on the technological and more difficult to discuss the ethical and political. However, just because discussions are difficult does not mean they are not worth having.
Technical choices always have political and ethical contexts and consequences. A community culture or policies that define such considerations as off topic is implicitly taking a specific political/ethical position on these questions. To use the recent example, to say that there should not be discussion of the ethical implications of accepting corporate sponsorship from a defense contractor is not substantially different from taking the position that the sponsorship should be accepted.
Part of what makes these conversations difficult is that participants in the community have direct interests in the decisions being made because of their material and political ties to the kinds of institutions being considered for ethical evaluation. More open acknowledgement of these interests doesn’t necessarily make the conversation any less difficult but it could at least allows for a clearer understanding of where different perspectives are coming from.
For a potential sponsorship policy, it could be worthwhile to specify that those involved in the decision-making process should acknowledge conflicts of interest arising from their involvement in a sponsorship decision: either directly through an interest in the potential sponsor, or indirectly through an interest in other company that could be impacted by a similar sponsorship decision in the future.
A more practical issue to resolve is perhaps how much we need sponsors and how much can we safely depend on them.
I think all this noise around scaling-up the user base is insane. My perspective is limited but it seems that NixOS was successful due to limited labor and compute rather in spite of it. Let’s not get in bed with some psychopaths so they can buy us stuff we don’t need.
TU Darmstadt and CCC should be excluded from consideration of future events. Telling sponsors they’re unwelcome days before an event is unacceptable. Offering to partially provide facilities for a subset of the event is a complete non-solution when it comes just days prior to the event, especially when the target of their issue is a sponsor. IANAL but I don’t see how a defense contractor talking about how they use a package manager could possibly be construed as being involved in weapon creation, I don’t think that argument holds water.
My read is that the parties at Darmstadt and at CCC found themselves in the position of being able to make a political statement against the military industrial complex, and made one, and unfortunately, Nix unwillingly was forced to come along for the ride.
Parties that put the nix community in the position of being forced to take a side on politics are parties that nix should never work with again. Pulling the rug at the last minute (and on a corporate sponsor) after they’ve already booked airfare and prepared slide decks and so on and so forth is totally just operating in bad faith.
Edit: to be clear, I don’t like the military industrial complex either. But usurping the voice of the nix community for your own views is worthy of burning bridges.
Maybe you should find a venue and organize NixCon next time? And probably find a different streaming platform and video recording team since you won’t use c3voc.
Easy to complain about the volunteer organizers’ mistake (not having checked the sponsors against the host’s policies) when you’re not doing anything yourself.
great quote from a master software engineer. thanks for putting it in the spotlight.
Good point. Regardless of the sponsor, last minute policy enforcements as a host should be unacceptable, and it should be unacceptable to the NixCon team for future venues.
I think people are overblowing the scope of the issue:
How can you avoid “last minute enforcements” when announcements of sponsorship are themselves “last minute announcements”?
The conference is organized on a best effort basis. If details and funding cannot be closed months beforehand, “last minute” problems like this are not completely avoidable.
The Anduril sponsorship was on the website for quite some time. Only when posted on chaos.social a few days before the conference did the outrage start.
That Anduril was sponsoring was known for more than “a few days”.
Nixpkgs-related builds are a lot of compute compared to other Open Source (Free/Libre-positioning-leaning or not) desktop/server OS distributions, and yeah, the already-unreasonable compute spent being insufficient for some workflows is a low-key issue more or less all the time.
I’m a bit confused on how this is seen as a last minute enforcement by the host. The policy of TU Darmstadt is longstanding. They just clarified that it indeed runs counter to their policies, they did not create a new one. That C3VOC volunteers are giving, well, volunteer service, is also well known and as I read the statements, they were quite willing to work with the organisers to find an agreeable solution that works for them. That CCC having a long ongoing debate about their level of support for military/defense is also not quite news.
It is also the organisers role to check the policies of the venue where they are. The TU acted the moment they were informed and asked. They are not responsible for checking the events they provide rooms for around the clock.
I think this gets a little bit blown out of proportion, as it was an obvious accident that the organisers of NixCon owned and will probably take better care of next time. Things like that happen all the time. Venue rules, particularly of universities, can be pretty weird. My favourite two I was presented with somewhere in a 10 pager:
Under no circumstances are organisers, speakers, and attendees allowed to make negative statements about the host company.
There’s a ban on circus equipment and bouncy castles (I want to know about the bouncy castle incident that probably triggered that rule)?
If you figure something like that out 3 days before the conf, something has to go.
If you take a couple of minutes to do some research based on the links which have been posted earlier in this thread, you can find the following (among other things):
Is the contracting entity a military or military-related institution, or a company or an arms-related area of business within a more broadly-based company in the field of defence technology?
Anyways, this isn’t your call to make. If TU says this doesn’t fly with their policies, it doesn’t matter whether you agree with their reasoning or not.