Add sponsorship policy. [WIP]

[djacu]

Fixes #110.

Formally submitting this as I think it is a sufficiently good place to start.

• Taken from @piegamesde's and @tomberek's input in the related issue.
• Modified to improve spelling and grammar.
• Wasn't sure how to parse the "and should be a home page or content related" phrase so I removed it.
• Changed line numbers so it formats in Markdown nicely.

[piegamesde]

This sponsorship policy formalizes the status quo of our current procedures fairly well. Or in other words, I fail how to see how it would help preventing any situations like the one we found ourselves in for this NixCon NA.

I mean, if your goal is "having a sponsorship policy at all", then sure, it's fine, go ahead. But if your goal is to actually do something to improve the situation by the means of creating a sponsorship policy and to prevent bad situation in the future, then that goal will not be reached.

To be blunt: To me, this text is almost useless and completely disregards most of the discussion points we have had on this topic. And I am increasingly bitter about the time and energy I lost on debating this since clearly nobody is listening.

[paperdigits]

To me, this text is almost useless and completely disregards most of the discussion points we have had on this topic. And I am increasingly bitter about the time and energy I lost on debating this since clearly nobody is listening.

Please submit your own sponsorship policy such that we can formally read and understand exactly what is desired; please be take action, rather than have a reaction. That's the only way we'll get anywhere.

[Janik-Haag]

Please submit your own sponsorship policy such that we can formally read and understand exactly what is desired

they did ... https://github.com/NixOS/foundation/issues/110#issuecomment-2006874249

[paperdigits]

That's certainly a draft, where is the more concrete, flushed out text?

[piegamesde]

@paperdigits would you prefer ignoring a finalized version of my text over ignoring my drafts? Should I paste what I wrote into a pretty PR for you to take it seriously?

[paperdigits]

I find your patronizing tone very off putting, @piegamesde, I'd ask that we maintain a civil tone and stay on topic.

While your draft is good, the bits such as "3.4. If want to rule out certain kinds of companies out of principle, put that here" and "5.3. Insert the rest of https://discourse.nixos.org/t/nixcon-na-2024-is-getting-sponsored-by-anduril-what-to-do-about-it/41258/17. This could be fleshed out in here or somewhere else, should there be a better place to write these down" compose quite a bit of the difficult bits that are being discussed. Without those bits filled in, it's more difficult to reach a final consensus.

It doesn't need to be a PR per se, but something that doesn't have "fill me in" parts would be really helpful.

[zimbatm]

Thanks @djacu

This sponsorship policy formalizes the status quo of our current procedures fairly well. Or in other words, I fail how to see how it would help preventing any situations like the one we found ourselves in for this NixCon NA.

@paperdigits, please take some time to read your comments and consider how your feedback could be applied to yourself. Constructive feedback would try to address Piegame's core concern instead of redirecting.

Both proposals also have a similar hole where the composition of the selection committee is left open for interpretation.

3. Objections should be specific to the sponsor and cite relevant reasons why the sponsorship would impede the NixOS mission.

@piegamesde under that rule, I believe we could discard some sponsors on the grounds that they are too controversial. The amount of focus we lost on that topic is already considerable, and it is good grounds for future discards.

That said, the current policy requires 2 objections, and it's unclear if that would only count as a single objection. Or if the same objection can come from two members. What did you have in mind @djacu ?

[piegamesde]

@paperdigits apologies for my tone.

Of the two placeholder points you remark, I have basically given up on both of them, so consider them removed. At this point, I think it is easier for me to focus on adding the points I care most about back into this proposal instead of polishing my own draft further.

Especially about my point 5. "Rejected company applications", I can live without what I drafted in 5.3, although not specifying it here will mean that the same task will simply be delegated to the event organizers. However, I'd like to see 5.2 "Employees of rejected companies must be welcomed as community members in the community and at events, as long as they drop all public affiliation with their employer within these spaces." incorporated into this proposal in some way or another.

[djacu]

That said, the current policy requires 2 objections, and it's unclear if that would only count as a single objection. Or if the same objection can come from two members. What did you have in mind @djacu ?

Well that was taken from the issue. But my thoughts were that two separate members would need to bring up the objection. Similar to a motion on a council where one person brings up a motion and a second votes to carry the motion forward. Except here it would be simply two votes.

[djacu]

However, I'd like to see 5.2 "Employees of rejected companies must be welcomed as community members in the community and at events, as long as they drop all public affiliation with their employer within these spaces." incorporated into this proposal in some way or another.

I am good with the first half of 5.2 but the second half is a bit ambiguous. To what degree would they have to drop all public affiliation?

• Not having their company on their badge (if that is a thing)?
• Not wearing apparel from their company?
• Not mentioning their company if they give a talk? E.g. "Hello, my name is X and I work for company Y."

Any of these seem a little too much to put on the individual.

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/update-on-sponsorship-policy-discussion/42704/1

[paperdigits]

@piegamesde thank you. I'm sorry for redirecting/demanding work, that is never OK and I'll strive to be more mindful in the future.

However, I'd like to see 5.2 "Employees of rejected companies must be welcomed as community members in the community and at events, as long as they drop all public affiliation with their employer within these spaces." incorporated into this proposal in some way or another.

This seems like it'd be impossible to enforce. How do you tell an employee from a fan of the company? And at what point is it considered encroachment upon that person's interests?

[crertel]

I'm just somebody quietly using Nix over in the corner and reading some of the work on this problem, but I'm not comfortable with a de-facto power grab by the moderation team here for vetoing events and then enforcing opinions on whether or not somebody can freely associate themselves with their employers at an event.

This is extremely unusual from everything I've seen at industry conferences, and the one time I recall it happening (over in the Elixir community) we still make fun of everybody involved in that decision and the awkwardness it caused.

I'll also point out that, as written, it seems that any rejected sponsorship would mean you couldn't fly your flag--for example, if I had some small business whose proposed sponsorship is of too little value to handle and got rejected this proposed change would mean I couldn't wear my company T-shirt. That seems dumb.

---

Taking a step back though, I think y'all are overthinking this and making mountains out of molehills. It's hard enough to get events organized and harder still to get funding for those events. Maybe just focus on getting a few years of events with sponsorship up reliably, assume good faith from everyone involved, don't overindex on what might happen with a sponsor that some people might not approve of, and then check in to see if anything needs to be changed if you can do that without goofing up a few times in a row at scale.

I guarantee that the handwringing over theoretical bad PR or whatever is a distraction from the already hard work of setting up events and getting paid. Few people know about Nix at all outside of a few corners of tech, fewer people still are actually running it meaningfully in production, even fewer of those are doing so for companies.

The ugly but reassuring truth is "basically nobody cares in the world at large." Take refuge in that, go outside and touch grass, and maybe consider removing roadblocks to sponsorship and avoiding adding processes that allow subsets of the ecosystem not doing the work to make it harder for those actually doing it.

[RaitoBezarius]

I'm just somebody quietly using Nix over in the corner and reading some of the work on this problem, but I'm not comfortable with a de-facto power grab by the moderation team here for vetoing events and then enforcing opinions on whether or not somebody can freely associate themselves with their employers at an event. [...] The ugly but reassuring truth is "basically nobody cares in the world at large." Take refuge in that, go outside and touch grass, and maybe consider removing roadblocks to sponsorship and avoiding adding processes that allow subsets of the ecosystem not doing the work to make it harder for those actually doing it.

Thank you for your comment, it is just sad that your comment contains so much contempt and seems to discard the possibility that what happened had real effects on people over the past months. I recommend taking the time to review the existing discourse to avoid writing a message which seems to be ignoring all the feelings of people working on the matter and pouring their soul in this thing for the past days.

Furthermore, I recommend also to forego suggestions that it's sufficient "to touch the grass" for problems to disappear in a community.

[crertel]

Thank you for your comment, it is just sad that your comment contains so much contempt and seems to discard the possibility that what happened had real effects on people over the past months. I recommend taking the time to review the existing discourse to avoid writing a message which seems to be ignoring all the feelings of people working on the matter and pouring their soul in this thing for the past days. [...] Furthermore, I recommend also to forego suggestions that it's sufficient "to touch the grass" for problems to disappear in a community.

You have not commented on the substance of my post:

• This is a defacto power grab.
• Similarly draconian measures are uncommon and have not worked out in the past for other communities and events.
• Nix is quite small right now in the grand scheme of things so we have room to assume positive intent and focus on getting any event up.
• The existence of emotional investment and long discourse and debate should not be conflated with the concrete problems of getting any event going with any amount of sponsorship from any group.

[AshleyYakeley]

I'm not comfortable with a de-facto power grab by the moderation team here for vetoing events and then enforcing opinions on whether or not somebody can freely associate themselves with their employers at an event.

I agree with this. The rationale given is that the moderation team somehow represents the community? But I think it's the other way around: as an inevitable consequence of moderation, the moderation team has a great deal of influence over what counts as "the community" and who may participate in it. This isn't a bad thing, moderation needs to be done, but we should recognise that the community has no reciprocal influence over the moderation team, and so the latter should not be considered representative of the former.

[djacu]

This is extremely unusual from everything I've seen at industry conferences, and the one time I recall it happening (over in the Elixir community) we still make fun of everybody involved in that decision and the awkwardness it caused.

@crertel I tried finding something related to this in the Elixir forms but was fruitless in my efforts. If this was documented somewhere, or if there is anything in the Elixir docs related to governance or policy, please share.

[crertel]

@crertel I tried finding something related to this in the Elixir forms but was fruitless in my efforts. If this was documented somewhere, or if there is anything in the Elixir docs related to governance or policy, please share.

It's oral history, and I'm not interested in airing out old drama with strangers--the main point being, it's weird to tell people at a conference they can't rep their employers if they want to.

If that's not good enough (and I wouldn't blame you), it should be sufficient to see other industry and event precedent: what other industry and community events use this proposed policy of controlling their attendees' visible affiliation?

[RaitoBezarius]

For information, I will take a proactive posture here to moderate the off-topic, outdated, abusive comments in this PR. We are focusing on getting a sponsorship policy off, concerns about whether a sponsorship policy makes sense can be done once, but once it is starting to run into circles, I will hide them.

Whether the moderation team or not is not representative of the community is another debate that has not its place in this PR, consider pushing proposals that encompass a condition that the selection committee is representative of the community in its decision-making, but let's not open all the debate at the same time, or we will not converge.

Dismissive comments are not tolerated anymore, and I will strongly moderate them, I include comments that disparage entities without proof, including:

• power grab fantasies
• reasoning about the perceived size of an entity to decide whether it makes sense to do proactive policymaking
• discussions that do not acknowledge the effective impact of the absence of sponsorship policy on the various operations of our community, e.g. nixpkgs

In addition, I will mention that what other communities does is, at some point, of limited usefulness to us. We have this community and not the community of other projects, it's up to the leadership of a project to answer with policies that make sense for their community, not to tell to community members that others do it like this, and therefore we must follow the "global ambient culture". If you disagree with this, please provide constructive elements to enable comparisons, otherwise, I will deem such comments offtopic too.

Again, I deeply apologize for the tone, but I will be rigorous to give everyone a chance to participate into this without being drained and dropping out of the discussion needlessly. This is not to say that those discussions cannot take place, but this might not be the right tool / right place to have such debates.

[AshleyYakeley]

Whether the moderation team or not is not representative of the community is another debate that has not its place in this PR

The problem is, nobody has given a good rationale for why the moderation team, of all people, should have a veto over sponsor selection. On the face of it, that sort of executive decision is the board's job, not the moderation team's job.

Other (bad) rationales for this:

• that the moderation team somehow has their finger on the pulse of the community in a way that the board does not -- if this even were true, the board is already capable of consulting the moderation team to help inform their decisions. Or indeed just reading the forums themselves.

• that the moderation team "bear most of the consequences if things go wrong, and thus have the highest stakes in the game" (from comment above) -- it's true that as moderators, they have somewhat more work if a lot of controversy is expressed on the various forums they moderate. But that's just part of the moderation job -- it's not really "high stakes" and it's not a good basis for making decisions.

  The people with the highest stakes are the board, since they're the ones ultimately held responsible for decisions.

[piegamesde]

they have somewhat more work if a lot of controversy is expressed on the various forums they moderate. But that's just part of the moderation job -- it's not really "high stakes" and it's not a good basis for making decisions

Let's be frank: If the moderation team were paid even something remotely reasonable for its job, the additional work of the fallout from the NixCon NA sponsorship would have been more expensive than the sponsorship money we got from Anduril.

I really really wish this was only "somewhat more work". But it's not. I wish the moderation team was in a better position, but in the last year the new recruitments have been barely sufficient to keep up with the burnout rate.

[crertel]

Dismissive comments are not tolerated anymore, and I will strongly moderate them, I include comments that disparage entities without proof, including: power grab fantasies reasoning about the perceived size of an entity to decide whether it makes sense to do proactive policymaking discussions that do not acknowledge the effective impact of the absence of sponsorship policy on the various operations of our community, e.g. nixpkgs

This is dismissing core, important issues.

• Re: power grab. The proposed changes would give the moderation team (currently, to my knowledge, the most vocal group) veto power. This power did not previously exist, and to my knowledge there is not a reciprocal veto power on moderation team activities. Even if you do not wish to classify this as a power grab, the core question remains: should the moderation team have this veto power?

• Re: scale. Proactive policymaking has a cost--as you and others keep pointing out, we apparently care about the cost to the moderation team, so apparently cost is a factor. Smaller orgs have a smaller cost budget than larger orgs. Similarly, smaller orgs have a smaller blast radius when things do go wrong, as can be seen in the difference of badness between a PR gaffe by myself, by my small employer, or by some giant company like Amazon. If we want to talk about how damaging bad PR is--e.g., talk about the blast radius--it makes sense to take into account how large the org is. It's a valid area of discussion to check cost vs benefits.

• Re: prior work by other communities. If I recall reading correctly, a lot of this discussion and its context uses other communities, notably the ASF, as a touchstone. In addition, if nobody else has a policy about not wearing silly hats, maybe that's a valid data point that a policy about not wearing silly hats is maybe not that important. Dismissing prior art here seems rather arbitrary.

• Re: discussions that do not acknowledge the effective impact of the absence of sponsorship policy on the various operations of our community, e.g. nixpkgs. Nixpkgs seems to be doing just fine. Can you enumerate the exact concerns, bullet-point style, of how lacking this policy hurts nixpkgs? I suspect the answer is (given other comments here) "it makes moderation harder". Okay, how important is that? Is the comfort of moderators more important than all other concerns (because again, that's what the proposed language implements with the veto setup)? Maybe it is, maybe it isn't, but that's a topical discussion worth having.

[RaitoBezarius]

Dismissive comments are not tolerated anymore, and I will strongly moderate them, I include comments that disparage entities without proof, including: power grab fantasies reasoning about the perceived size of an entity to decide whether it makes sense to do proactive policymaking discussions that do not acknowledge the effective impact of the absence of sponsorship policy on the various operations of our community, e.g. nixpkgs

This is dismissing core, important issues.

Thank you for improving your prose on the matters you care about.

• Re: power grab. The proposed changes would give the moderation team (currently, to my knowledge, the most vocal group) veto power. This power did not previously exist, and to my knowledge there is not a reciprocal veto power on moderation team activities. Even if you do not wish to classify this as a power grab, the core question remains: should the moderation team have this veto power?

I am a project team lead of NixCon, the answer is yes. If folks want to organize conferences and find out that the moderation team having this veto power is excessive, we can revisit. This decision does not enshrine the moderation team as some sort of unchangeable god.

I hope this is clear enough and we don't enter into a debate on that, I am not interested into that.

• Re: scale. Proactive policymaking has a cost--as you and others keep pointing out, we apparently care about the cost to the moderation team, so apparently cost is a factor. Smaller orgs have a smaller cost budget than larger orgs. Similarly, smaller orgs have a smaller blast radius when things do go wrong, as can be seen in the difference between a PR gaffe between myself, my employer, and one of our vendors (say, AWS). If we want to talk about how damaging bad PR is--e.g., talk about the blast radius--it makes sense to take into account how large the org is. It's a valid area of discussion to check cost vs benefits.

Yes, I am a project team lead of NixCon and a board observer of the NixOS Foundation and a quite active community member.

I can tell you that the cost of not doing that is devastating for most of my operations that were ongoing before the event: we lost weeks of work of many individuals whose time is extremely expensive and valuable.

You can measure it quite easily by reviewing all the reactions over the situation over social media, Discourse, GitHub, Matrix, Discord and any platform. If you dismiss this and say 'this is not a big deal', I'm afraid we cannot come to a mutual understanding of the concept of cost and I think there's no further conversation to have over this matter.

• Re: prior work by other communities. If I recall reading correctly, a lot of this discussion and its context uses other communities, notably the ASF, as a touchstone. In addition, if nobody else has a policy about not wearing silly hats, maybe that's a valid data point that a policy about not wearing silly hats is maybe not that important. Dismissing prior art here seems rather arbitrary.

Yes, those policies were used for starters, and they created this situation, so what you are saying is you want the community to accept the policies of other communities and this is dismissive of our community.

I don't think we are in the business to force communities to agree to our views.

• Re: discussions that do not acknowledge the effective impact of the absence of sponsorship policy on the various operations of our community, e.g. nixpkgs. Nixpkgs seems to be doing just fine. Can you enumerate the exact concerns, bullet-point style, of how lacking this policy hurts nixpkgs? I suspect the answer is (given other comments here) "it makes moderation harder". Okay, how important is that? Is the comfort of moderators more important than all other concerns (because again, that's what the proposed language implements with the veto setup)? Maybe it is, maybe it isn't, but that's a topical discussion worth having.

As you don't seem to be a contributor, I suppose you are not aware of what's going on, I will put it shortly like this: long term trusted community members who are critical to our operations are quitting the project over this situation and the lack of swift decisionmaking from our side. You cannot know that, you are not participating into this part of the community, but from my PoV, this part of the community is critical and justify why we are going about this like this.

If everyone who is not engaging into Nixpkgs operations on a daily basis will come and ask to rehash everything that's going on, that's not tractable, I will ask that folks who don't know, don't interact that much, to step aside and leave the space for people who are actively working directly on the project. Otherwise, you are wasting time of others, as time is of essence here.

---

So far, I feel like you are just not really aware of what's going in this community, this is normal for a newcomer, but I wouldn't expect a newcomer to come and try to give us lessons about how we do such a thing, I do not recommend doing this or engaging into this.

[crertel]

the lack of swift decisionmaking from our side.

If the concern is swift decisionmaking, the earlier suggested language of this PR--having the Board make a vote--seems to be the faster thing (consensus happens faster with n=1). If the problem is the Board not doing its job quickly enough, I'm unsure that this solves for that.

dismissive of our community.

If you limit community to the people you engage with on social media, then yes I am not a member of the community. If you include people who have been running Nix quietly in production for users for years, and who was first introduced to Nix (the language) around 2011, then yes I'm a member of the community. Consider me part of the "dark matter" of Nix developers--and if that isn't good enough to be considered part of the community, that's unfortunate but doesn't impact my servers.

If folks want to organize conferences and find out that the moderation team having this veto power is excessive, we can revisit.

Seems simpler to go with the faster approach originally in the PR, and add the moderation team veto in if it turns out to be a problem.

long term trusted community members who are critical to our operations are quitting the project over this situation and

It's unclear if you mean "our" (the moderation team) or "our" (the folks involved in making sure I can still pull from the repo and so forth in a more general sense). I understand your alarm and concern for your team, if that's been a problem.

[RaitoBezarius]

the lack of swift decisionmaking from our side.

If the concern is swift decisionmaking, the earlier suggested language of this PR--having the Board make a vote--seems to be the faster thing (consensus happens faster with n=1). If the problem is the Board not doing its job quickly enough, I'm unsure that this solves for that.

Right, unfortunately, we can only do what we can with what we have.

dismissive of our community.

If you limit community to the people you engage with on social media, then yes I am not a member of the community. If you include people who have been running Nix quietly in production for users for years, and who was first introduced to Nix (the language) around 2011, then yes I'm a member of the community. Consider me part of the "dark matter" of Nix developers--and if that isn't good enough to be considered part of the community, that's unfortunate but doesn't impact my servers.

Yeah, so, I wouldn't say that I limit community to that. I am very happy you are running NixOS in production, but, unfortunately, in the project, I have to work with the people who are affected and care about things. If this doesn't impact your servers or yourself, I would like to ask you to let people who are affected to express themselves or to be representative of such folks.

If folks want to organize conferences and find out that the moderation team having this veto power is excessive, we can revisit.

Seems simpler to go with the faster approach originally in the PR, and add the moderation team veto in if it turns out to be a problem.

So following this reasoning, we should include the moderation team veto now because this policy is not really far from the prior informal in existent policy and proved to be a problem. Happy we came finally to an agreement.

long term trusted community members who are critical to our operations are quitting the project over this situation and

It's unclear if you mean "our" (the moderation team) or "our" (the folks involved in making sure I can still pull from the repo and so forth in a more general sense). I understand your alarm and concern for your team, if that's been a problem.

I am not part of the moderation team, "our" refers to operations of the whole NixOS project, that includes the build infrastructure team for example who lost two members over this situation. But that's not limited to it, there's plenty of people who are maintaining stuff who are throwing the towel. We can pull repo today, whether we will be able to continue to maintain the complicated nixpkgs machinery in six months is questioned here.

[numinit]

I can tell you that the cost of not doing that is devastating for most of my operations that were ongoing before the event: we lost weeks of work of many individuals whose time is extremely expensive and valuable.

You can measure it quite easily by reviewing all the reactions over the situation over social media, Discourse, GitHub, Matrix, Discord and any platform. If you dismiss this and say 'this is not a big deal', I'm afraid we cannot come to a mutual understanding of the concept of cost and I think there's no further conversation to have over this matter..

Conflating social media reaction with the moderation team's desired representation of the community is a mistake. It's important to be clear: while community representation is important, setting a precedent for those who yell the loudest on social media getting their way can easily be abused by starting an internet mob in order to gain more representation through social pressure. People know this tactic works. And, as we have seen, this results in a lot of yelling over each other and division, and is hard to moderate.

Yes, those policies were used for starters, and they created this situation, so what you are saying is you want the community to accept the policies of other communities and this is dismissive of our community.

I think we need to be clear about what created the situation: no policy to start for NixCon EU, and a policy that many thought inadequately addressed the problem thereafter for NixCon NA. Now, it's clear we need some sort of policy, and everyone who feels slighted now wants it to be their policy.

---

As a frequent contributor to nixpkgs, I have not really loved how these discussions have gone for quite some time, and dislike what seems like the community yelling over each other, dueling via public open letters, squabbling about each others' perceived differences in morality, and the entire thing generally turning certain sections of the Discourse and Github into a corrosive mess. Everyone's trying to grab the steering wheel for a problem that will never be adjudicated with satisfaction to everyone, and I much prefer contributing to nixpkgs over expending more words on this stuff...

So, to build some consensus, my only recommendation is for the project to pick its battles:

• If the policy is too complex by including a bunch of hidden vetoes, rules that come down to differences in opinion (mostly, the ones that have been squabbled over), and opportunities for backhandedness, we'll lose more trust and drag this out for longer. If something is empirically being squabbled over excessively, maybe it's best just left out.
• If any team needs representation, that's fine if and only if how it's done removes all scope for manipulation. Transparent decision-making processes and a policy that establishes clear process will help. People just yelling about the issue or shutting each other down because of differences in opinion should not hinder decision-makers being neutral.
• We should set expectations that not everyone will get their way. Compromise doesn't have to come at the expense of trust. See the two previous points.
• And... just be kind to each other, this whole thing is a pain and doesn't need to be nearly this contentious.

I'm grabbing a beer, logging off for tonight, and installing NixOS on a fanless i7 SBC router I'm setting up... :smiley:

[RaitoBezarius]

I can tell you that the cost of not doing that is devastating for most of my operations that were ongoing before the event: we lost weeks of work of many individuals whose time is extremely expensive and valuable.

You can measure it quite easily by reviewing all the reactions over the situation over social media, Discourse, GitHub, Matrix, Discord and any platform. If you dismiss this and say 'this is not a big deal', I'm afraid we cannot come to a mutual understanding of the concept of cost and I think there's no further conversation to have over this matter..

Conflating social media reaction with the moderation team's desired representation of the community is a mistake. It's important to be clear: while community representation is important, setting a precedent for those who yell the loudest on social media getting their way can easily be abused by starting an internet mob in order to gain more representation through social pressure. People know this tactic works. And, as we have seen, this results in a lot of yelling over each other and division, and is hard to moderate.

Correct, though, this is unrelated and missing the point.

Yes, those policies were used for starters, and they created this situation, so what you are saying is you want the community to accept the policies of other communities and this is dismissive of our community.

I think we need to be clear about what created the situation: no policy to start for NixCon EU, and a policy that many thought inadequately addressed the problem thereafter for NixCon NA. Now, it's clear we need some sort of policy, and everyone who feels slighted now wants it to be their policy.

Correct, you are just rehashing what everyone has been saying in the past. The important criterion is whether a policy would have fixed the situation for NixCon EU & NixCon NA. Whether folks want their policy or not is of little interest to me. What I care about is whether this is an effective policy to address community concerns and get everyone back to productive work.

As a frequent contributor to nixpkgs, I have not really loved how these discussions have gone for quite some time, and dislike what seems like the community yelling over each other, dueling via public open letters, squabbling about each others' perceived differences in morality, and the entire thing generally turning certain sections of the Discourse and Github into a corrosive mess. Everyone's trying to grab the steering wheel for a problem that will never be adjudicated with satisfaction to everyone, and I much prefer contributing to nixpkgs over expending more words on this stuff...

Sure, we all are in the same boat, but those discussions are of political nature and are important. Somefew have to take care of them.

So, to build some consensus, my only recommendation is for the project to pick its battles:

• If the policy is too complex by including a bunch of hidden vetoes, rules that come down to differences in opinion (mostly, the ones that have been squabbled over), and opportunities for backhandedness, we'll lose more trust and drag this out for longer. If something is empirically being squabbled over excessively, maybe it's best just left out.
• If any team needs representation, that's fine if and only if how it's done removes all scope for manipulation. Transparent decision-making processes and a policy that establishes clear process will help. People just yelling about the issue or shutting each other down because of differences in opinion should not hinder decision-makers being neutral.
• We should set expectations that not everyone will get their way. Compromise doesn't have to come at the expense of trust. See the two previous points.
• And... just be kind to each other, this whole thing is a pain and doesn't need to be nearly this contentious.

I am sorry, I fail to understand your recommendations, you are hinting at a mix of things and I cannot see the relation between your words and the current state of things. From my PoV, this is not actionable, but thank you for your input.

[numinit]

Let me be very clear, then, and explicitly spell it out: You are likely losing contributors because a lot of the discussion is toxic and people are failing to compromise. This is being egged on by the people who are supposed to be in charge, yourself included, who are not actually discussing things in a way that facilitates compromise. I'm providing actionable ways for people to come to compromise, including recommending setting expectations that not everyone will get their way. So, responding to that with:

Sure, we all are in the same boat, but those discussions are of political nature and are important. Somefew have to take care of them.

Completely disregards that a discussion can be of "political nature" and not just people yelling past each other.

And:

I am sorry, I fail to understand your recommendations, you are hinting at a mix of things and I cannot see the relation between your words and the current state of things.

This is a major reason the discussion has turned toxic. Here's a concrete example:

Responding to newcomers with shutting entire responses down:

I hope this is clear enough and we don't enter into a debate on that, I am not interested into that.

Otherwise, you are wasting time of others, as time is of essence here.

I wouldn't expect a newcomer to come and try to give us lessons about how we do such a thing, I do not recommend doing this or engaging into this.

And, then twisting words:

Happy we came finally to an agreement.

is a concrete reason that this discussion is toxic.

[RaitoBezarius]

Let me be very clear, then, and explicitly spell it out: You are likely losing contributors because a lot of the discussion is toxic and people are failing to compromise. This is being egged on by the people who are supposed to be in charge, yourself included, who are not actually discussing things in a way that facilitates compromise. I'm providing actionable ways for people to come to compromise, including recommending setting expectations that not everyone will get their way. So, responding to that with:

Sure, we all are in the same boat, but those discussions are of political nature and are important. Somefew have to take care of them.

Completely disregards that a discussion can be of "political nature" and not just people yelling past each other.

I fail to understand what constitute 'people yelling past each other' vs. 'a discussion of political nature'.

And:

I am sorry, I fail to understand your recommendations, you are hinting at a mix of things and I cannot see the relation between your words and the current state of things.

This is a major reason the discussion has turned toxic. Here's a concrete example:

Responding to newcomers with shutting entire responses down:

I hope this is clear enough and we don't enter into a debate on that, I am not interested into that.

Otherwise, you are wasting time of others, as time is of essence here.

I wouldn't expect a newcomer to come and try to give us lessons about how we do such a thing, I do not recommend doing this or engaging into this.

And, then twisting words:

Happy we came finally to an agreement.

is a concrete reason that this discussion is toxic.

OK, I am sorry you think the discussion is toxic.

Unfortunately, I will make it clear again, newcomers doesn't have a lot of useful inputs to add here except if they're experts on the subject (sponsorship policies, open source governance models, etc.) because they don't have the full context to comment on the topic without forcing everyone else to rehash the context for them. This makes the discussion extremely draining for people with the context who want to move forward on the points we need to compromise on.

Does this create the risk of losing contributors over this discussion? Yes. Do I want that? No. Am I forced to answer those messages? Yes.

Nevertheless, everyone is free to observe this discussion and decide what they want to do with their free time. If you are not affected, if you are not a stakeholder of that discussion, please consider NOT chiming into that discussion to let folks who are affected, who are stakeholder of that discussion drive this discussion.

The contributors we are losing right now are contributors who are explicitly (:= this is what they told me) tired of your answers, the answers of newcomers who have no idea of what's going on in this community.

I don't know who are 'the contributors' you are mentioning in your header, I don't know if they are maintainer of the critical packages of nixpkgs, I don't know if they are newcomers who would become contributors. But again, as I said, this is not actionable to me. Feel free to contact me in private if you want to add more to this that you do not want to share publicly, my DMs/emails and anything and everything are always open.

I cannot make it simpler than that, if this is not possible to understand, I will moderate systematically until we get to this point. This is my last warning on the offtopic discourse.

---

Either:

• react to actionable items to make the sponsorship policy move forward
• start a new topic about current moderation team position in the community
• start a new topic about what should be the moderation team in the future in the community

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/community-calendar/18589/130

[AshleyYakeley]

I would like to point out that the current PR is a huge improvement over having no policy. It clearly empowers both event organisers to decline sponsorships that they believe are locally unpopular, and the board to decline sponsorships that they believe would be detrimental to the larger community, where this wasn't clear before.

The important criterion is whether a policy would have fixed the situation for NixCon EU & NixCon NA.

Trying to guarantee a single particular outcome, especially for a past event, is a bad approach to making policy. Ask yourself, if the moderation team happened to be very pro-Anduril, as is perfectly possible in the future, would you still want to make this change?

[RaitoBezarius]

The important criterion is whether a policy would have fixed the situation for NixCon EU & NixCon NA.

Trying to guarantee a single particular outcome, especially for a past event, is a bad approach to making policy. Ask yourself, if the moderation team happened to be very pro-Anduril, as is perfectly possible in the future, would you still want to make this change?

If the moderation team was pro-Anduril in its actions, nixpkgs would be defunct.

I wonder why is this so hard to understand given all those reactions from contributors of large ecosystem in nixpkgs who form a network of even larger contributors of signifcant subsystems of nixpkgs.

I wonder also if the motivations of certain individuals here is to force a fork of nixpkgs to the detriment of the community or if this is just a benign misunderstanding of not being someone who contribute and interact on a daily basis with nixpkgs developers.

[paperdigits]

Nobody is going to fork nixpkgs.

Please, let's try not to be so flippant. We're in this situation because what was in/not in place previously didn't work. Now we're trying to find a way forward.

If the moderation team was pro-Anduril in its actions, nixpkgs would be defunct.

This feels a little bit too dramatic of a statement. Perhaps the team would sustain a substantial workload hit, but nixpkgs itself, the code base, would still be around.

While Andruil is the past, its quite easy to imagine other future scenarios where people's current feelings might be flipped. If there is a NixCon in Iran or Russia or another non-Western country where a government contractor or other business is using nix but their output is objectionable.

For me personally, this boils down to the problem of contributing to the commons, which is problematic, but still the best idea we have. The more contribution to the commons there is, the better off we are. Even where we disagree with some entity's output, understanding their use of the commons enriches the commons itself.

My own fear, and the reason why I've taken the time to try and contribute to this conversation, is because I would not want to find the Con I attend to be canceled or not be able to afford a venue because a sponsor has been rejected. I appreciate and am envious that our European counterparts have the possibility of finding a large scale venue that'll host a con for little or no cost, but that isn't a reality here in America and probably other places around the world.

I wonder also if the motivations of certain individuals here is to force a fork of nixpkgs to the detriment of the community or if this is just a benign misunderstanding of not being someone who contribute and interact on a daily basis with nixpkgs developers.

I don't think we should take the conversation down this path. I understand your frustrations @RaitoBezarius

I think that in the interest of compromise and moving forward, that in Sunday's meeting we should work to agree and codify that the process (1) needs the formation of a selection committee and (2) the selection committee should include a community representation component. Once we agree on this, then we can take on defining the composition of the selection committee.

[RaitoBezarius]

If the moderation team was pro-Anduril in its actions, nixpkgs would be defunct.

This feels a little bit too dramatic of a statement. Perhaps the team would sustain a substantial workload hit, but nixpkgs itself, the code base, would still be around.

What value is there for a nixpkgs that does not receive the sheer amount of free work from the people who have been clearly against the sponsorship? Yes, sure, the repo would still exist. Then, you can count how many days / weeks you are left before you cannot make it work anymore, be it for infrastructure reasons or maintenance reasons.

Yes, we can have a nixpkgs without those people, but I should remind everyone to try to understand what makes nixpkgs magic and what constitute its value.

While Andruil is the past, its quite easy to imagine other future scenarios where people's current feelings might be flipped. If there is a NixCon in Iran or Russia or another non-Western country where a government contractor or other business is using nix but their output is objectionable.

For me personally, this boils down to the problem of contributing to the commons, which is problematic, but still the best idea we have. The more contribution to the commons there is, the better off we are. Even where we disagree with some entity's output, understanding their use of the commons enriches the commons itself.

My own fear, and the reason why I've taken the time to try and contribute to this conversation, is because I would not want to find the Con I attend to be canceled or not be able to afford a venue because a sponsor has been rejected. I appreciate and am envious that our European counterparts have the possibility of finding a large scale venue that'll host a con for little or no cost, but that isn't a reality here in America and probably other places around the world.

I think you are conflating two things.

(a) Letting everyone contributing to the commons: this is not what we were discussing here. (b) Making conference happen and possible to attend in expensive places of the world, respectfully of the community: This is the matter.

In (b), you are conflating two things again.

(a) Making a conference possible requires a target budget A (b) Accepting as many sponsorships as possible

I will remind that I was involved with the past NixCon and I came to NixCon NA. Given our current state of popularity, I do not believe there's any trouble to have (a) even if we don't do (b). (I am even involved into looking into NixCon Asia, so...)

Thus, I think your fear of not having any conference at all is only but a fear.

Assuming that we would have a problem, conference in place A or B requires more money than usual, this is a board matter to see if we can use the money of successful NixCon to help other NixCon to flourish, for example.

Of course, sometimes thing does not pan out as desired / expected, but this is hardly related to whether we have to offend the whole community by accepting a sponsor for a meager small K of dollars.

I wonder also if the motivations of certain individuals here is to force a fork of nixpkgs to the detriment of the community or if this is just a benign misunderstanding of not being someone who contribute and interact on a daily basis with nixpkgs developers.

I don't think we should take the conversation down this path. I understand your frustrations @RaitoBezarius

I don't think too, but I think we are having too many interactions from people who are not grasping the situation we are in. Nonetheless, people who are watching this issue and us and who are grasping the situation are unamused and are losing hope or being significantly affected by seeing folks spreading out dismissive comments while they are in NixOS/*, nix-community/*, working and trusting us to do the right thing and represent their opinion adequately.

So when I see so much out-of-touch participation, I can only wonder if I am being subjected to interactions of the same nature as https://lwn.net/ml/debian-vote/CA+f80t5gKqVHwDPLw9wHSUdWNkysN8mdoH+f-e34LSYTtk6Okw@mail.gmail.com/ here.

I think that in the interest of compromise and moving forward, that in Sunday's meeting we should work to agree and codify that the process (1) needs the formation of a selection committee and (2) the selection committee should include a community representation component. Once we agree on this, then we can take on defining the composition of the selection committee.

I think we already agreed on (1), (2), I believe we need to discuss the composition of the selection committee on Sunday's meeting as per the PR thread that sparked the meeting. If we need to roll back (1) & (2), I feel like we are taking a ton of step backward every time we are trying to move forward, and this is not a good sign.

[lf-]

This isn't going to happen. For example, nobody threatened to stop their free work in the open letter, because it would be obviously non-credible and come across as petulant. I don't know why people think I'm being flippant in making this point.

Two major members of NixOS infra team are currently on strike until this is resolved in a satisfactory way. There is a strong likelihood of there not being any 24.05 release because there's no committers who aren't on strike who would be release managers. Some committers who are on the moderation team can't help with releasing the software because they're too busy wasting their time dealing with bad-faith trolls on threads related to this issue and overall increase in background trolling as of late.

Currently the project is at its knees because of the foundation not resolving the issue of advertising for companies actively involved in imperialism and, some, in the genocide in Gaza. I'm not interested in discussing whether any particular companies are in either of those positions. The point here is that the foundation isn't listening to the most active contributors and they're suffering the consequences thereof (several tens of thousands of dollars of very much non-fungible labour withheld, significantly more than the sponsorships in question).

[paperdigits]

I think you are conflating two things.

(a) Letting everyone contributing to the commons: this is not what we were discussing here. (b) Making conference happen and possible to attend in expensive places of the world, respectfully of the community: This is the matter.

In my view, bringing people together at events is enriching the commons, which is why I brought it up.

In (b), you are conflating two things again.

(a) Making a conference possible requires a target budget A (b) Accepting as many sponsorships as possible

Please note that I did not say anything about accepting as many sponsors as possible. Rather my concern was specifically the opposite: if we have so few sponsors that rejecting one makes the conference not financially possible.

It seems we've only wanted to discuss things that have already happened & how not to find ourselves in that situation again, but I feel strongly that policy needs to be forward looking as well.

[mweinelt]

my concern was specifically the opposite: if we have so few sponsors that rejecting one makes the conference not financially possible.

Thanks for worrying about that, but I can tell you as a NixCon 2023 organizer, that this fear is unfounded.

We had 15 sponsors and excluding the one sponsor did not make a dent in the budget. The NixCon 2022 sponsor section also doesn't look like it lacked sponsors.

I recognize that NixCon NA had fewer sponsors, but that is also not a sign that the event would have collapsed without the sponsorship in question.

It seems we've only wanted to discuss things that have already happened & how not to find ourselves in that situation again, but I feel strongly that policy needs to be forward looking as well.

Weigh the cost of offending contributors versus the small monetary gain.

[paperdigits]

Thanks for worrying about that, but I can tell you as a NixCon 2023 organizer, that this fear is unfounded.

Again, I'd ask that the comments not be so flippant. Sponsors of the past are not necessarily sponsors of the future. Are we so stuck looking backward right now that what's ahead isn't even in sight?

Weigh the cost of offending contributors versus the small monetary gain.

It's not just the "small monetary gain," its about having a policy that is acceptable for the next thousand contributors, and the thousand after that; that is what I mean by "forward looking." Unless we're planning to revisit this often, we should try and make policy that works for a longer period of time.

[zimbatm]

Ok, that's enough.

If your comment isn't directly helping move the policy out of the door, it will be marked. And if you aren't a contributor to this project, be extra careful, I will ask the moderation team to ban you instead. We can't afford to have the conversation be detailed.

We need a simple and practical policy we can use at NixCon, that samples a bit more of the community for opinion. This is the first version and it can be improved based on feedback.

Let's aim at next Tuesday to get this over with and reviewed by the board.

[edolstra]

The sponsor rules as currently proposed by this PR seem reasonable to me:

1. Sponsors must be individuals or organizations that are legally able to operate in compliance with the laws and policies of their own origin, that of the NixOS Foundation, and of relevant third-party agreements.

since this is a non-discriminatory policy that doesn't exclude entire industries.

However, it's important to acknowledge that this policy does permit sponsorship by defense contractors, which is what triggered the current controversy. So the big question is whether committee members can reject sponsors on grounds not listed in the policy. (The proposal does mention a requirement to state "reasons why the sponsorship would impede the NixOS mission", but that's pretty vague, lacking a stated "NixOS mission".) If in practice the policy is "the ethics/politics of the committee members", then that's not a very transparent policy. Ideally it should be clear in advance to potential sponsors whether they're welcome.

[JulienMalka]

The sponsor rules as currently proposed by this PR seem reasonable to me

Just want to clarify/reiterate that the current state of this PR does not reflect the agreement that was found yesterday after countless hours of negotiation.

So the big question is whether committee members can reject sponsors on grounds not listed in the policy.

The committee members can reject sponsors on grounds not explicitly listed, yes.

[JulienMalka]

Please refrain from changing all the wording of the policy while we are discussing it. Let's try to focus on the open discussion points.

[zimbatm]

We have now integrated the remaining changes in #136, which I believe is very close to the consensus we reached on Sunday. For more context, please also read the discourse post. Thanks again for all your feedback, effort and patience.