Hacker News Reader

So, is it illegitimate, then, to have a FOSS project with an intentional/cultivated "culture of abuse", where the idea is that nobody is kind to anybody and that's how everybody likes it?

The thing about this hypothetical is that it's wildly different from what Linus Torvald mailing list abuse winds up being. The approach is "we have a culture of anything goes, not because anyone likes abuse, but it's best way to get things done, and because most competent people do things that way, starting with Linus himself, who's clearly ultra-competent".

Essentially, while there might be some similarities to each, the abuse in open source culture is far more similar to cult abuse than BDSM activity. In cult-style abuse a person entering a project would confronted by escalating abuse and they either flee or accept and begin to normalize it whereas in BDSM, a person is given a description of what they're getting into from the start (though shading towards other "styles" is a hazard here too, of course).

Thing is there are a spectrum of approaches on the Internet but with the Internet no longer young, it's understandable people are concerned about where various styles end up. I have a friend who frequents 4Chan and says they get a lot out of the anything goes styles - "you just have to be willing to accept being called a fag and not taking any of it seriously". The problem is it's known that a substantiale-enough-to-dangerous group of people take the abuse as exactly that point and rather than seeing at restricting it naming calling in forums, live stream mass murder (say, they guy in Australia). This takes people a little aback.

BDSM irl has a lot of controls to prevent people from taking it "seriously", the point isn't to have some who want to torture in a no-limits fashion. If you could have some equivalent way to filter out people who can't separate play from reality, that would be great but I haven't the more abusive nooks of the Internet being very good at that.

Linux might be where many people's minds go by default in this comparison, but it's not the example I meant to conjure here.

I'm talking more about, for example, the FOSS communities around game-console homebrew tooling, where any given project is usually an outgrowth of some private Discord somewhere—some pre-existing group of friends—and so everyone on the project calls one-another names in the [public] GitHub issues; often actively trolls one-another with counter-productive PRs or issue-filings; sometimes creates new projects from scratch to replace/obsolete someone else's project just because A disagrees with B about some minor non-software-related thing, e.g. taste in music, and so "can't respect them or their software"; etc.

These people are clearly playing around as they make this software—and often producing the software rather sub-optimally in the process—but the goal of most of these project communities is clearly to have fun with the process of making software, not to end up with useful software.

And now, back to the question: are these projects, and their processes, to be considered societally acceptable? Or are they "canceled?" Would adding a Code-of-Conduct stating clearly what they're doing in the project even help? (IMHO it doesn't seem like it'd make much of a difference; it's not like these projects get outside contribution anyways.)

https://filepush.co/2Hvm/degradation-of-hobby.jpg

My guess is something like this: Once the project becomes big/useful enough and normal people are now using and relying on it, then they can start ousting the eccentrics/untolerables because at a large enough population, normal people just naturally outnumber them.

Disclaimer: Image is just an example to express an idea/concept. I do not endorse whatever is in the image and my post does not represent my employer, Unemployed Inc.

The thing with semi-abusive, semi-fun-loving group of friends and the perhaps multitude of kind of ambiguously abusive relationships one might notice small circles of people, is that these seldom scale to millions of people nor do they scale to long periods of time.

Any individual or group can do whatever they want in the OSS world, nobody is stopping anyone from being a jerk except those communities that voluntarily self-govern in such a fashion that excludes jerks. Where's the problem?

Mostly the https://en.wikipedia.org/wiki/Abilene_paradox.

I've seen a few cases now where a FOSS project community—one that is nominally open to new members, but is in practice a closed and static group of close-knit friends—starts off all perfectly happy with a sort of raunchy/abusive internal demeanor; but then someone comes to them with a suggestion of some Code of Conduct or other standard of decorum; and then everyone suddenly decides that it's a good idea, not because they personally want to change how they interact, or even to see others interact differently with them, but because they don't want to be the one caught arguing against the CoC if it turns out that everyone else wants it.

That's still a voluntary choice made by those individuals. If the group doesn't want to follow the suggestion of an outsider they don't have to do that.

Well, yes; my point was that there's certain societal expectations "out in the environment" that cause groups to do dumb things; and the way to fix the problem created by that is to change the societal expectations.

Society is very diverse, it's not that realistic to change society such that you stamp out the ideas you disagree with, it's much more feasible and effective to use the power you have within your own community to stand up to ideas you disagree with.

I mean, the whole point of the Abilene paradox is that it's something that happens when everyone's level of regret they'd have from doing the thing doesn't exceed their level of regret from social approbation if it turns out everyone else did want to do the thing and they had to be the one to argue against it. So in the end everyone loses a little utility—rather than everyone expecting to lose a lot of utility (in the form of personal social capital), and then actually losing nothing-at-all when it turns out nobody's on the other side of the argument.

If you want to stand up for a lack of decorum in your community, there's nothing stopping you; but the behavioral economics just don't work out in favor of anyone following after your example, and do in fact work out in favor of the rest of the group tearing down the presumed-social-more-violator to protect themselves from being associated with you. (Again, even in cases where everyone turns out to secretly agree with you!)

People will have opinions about things, it's not feasible to eliminate the possibility of exposure to ideas you disagree with. If the Abilene paradox is so crippling that it precludes the possibility of expressing your own opinions within your own community then what hope is there to "change societal expectations"?

Divine grace.

Groups that don't have a code of conduct (which in my experience increases rather than reduces the amount of jerkdom) get pressurised to add one. Conferences that don't have one attract twitter harassment campaigns. It's not really a free choice.

It is absolutely a free choice, social pressure isn't the law.

"The law" is ultimately just a particular form of social pressure, there's nothing magic about it. If someone is forced into bankruptcy because a twitter campaign got them fired, that's practically the same consequence as being forced into bankruptcy because a court imposed a large fine.

> "The law" is ultimately just a particular form of social pressure, there's nothing magic about it.

This is wrong. The "magic" (i.e. the thing that distinguishes it from social pressure) is threat of violence or imprisonment from the state if you don't comply with the law.

> If someone is forced into bankruptcy because a twitter campaign got them fired, that's practically the same consequence

No it isn't. You are employed at-will by your employer, they are free to fire you if they decide they want to do that, but they don't have to. If the business makes the decision to fire you based on public pressure that's their right unless you're making an argument for stricter labor laws (I suspect you aren't).

> This is wrong. The "magic" (i.e. the thing that distinguishes it from social pressure) is threat of violence or imprisonment from the state if you don't comply with the law.

Again, though, any violence is ultimately going to be carried out by humans for social reasons. While these online outrage mobs are, as yet, nowhere near as powerful as governments, they're already making credible threats of violence, so that's not the clear bright line you seem to think it is.

> No it isn't. You are employed at-will by your employer, they are free to fire you if they decide they want to do that, but they don't have to. If the business makes the decision to fire you based on public pressure that's their right unless you're making an argument for stricter labor laws (I suspect you aren't).

I am, among other things; at-will employment has always seemed unreasonable to me, as proven by the very existence of protected classes. It's popular to say "asshole isn't a protected class", but as far as I can see the whole rationale for protected classes applies just as strongly to those who have said legal but unpopular things.

> any violence is ultimately going to be carried out by humans for social reason

Every interaction with another human falls under the label of "social", the word ceases to mean anything when used to describe everything, abusing that rhetorical ambiguity to equate tweets on the internet with civil order enforced by armed agents of the state is absurd.

> While these online outrage mobs are, as yet, nowhere near as powerful as governments

Correct, not even close, and never will be in any conceivable future. The idea that a twitter mob could somehow ever be comparable in power to a government demonstrates a serious disconnection from reality and the power that governments exercise over their people all around the world and all through the annals of history.

> I am, among other things; at-will employment has always seemed unreasonable to me,

So what are you proposing?

> It's popular to say "asshole isn't a protected class", but as far as I can see the whole rationale for protected classes applies just as strongly to those who have said legal but unpopular things.

An asshole is not a "class" of people, it's a behavioral disposition. Nobody needs to be an asshole. The social advantages of being an asshole come with the consequences of people not wanting to interact with you.

> Correct, not even close, and never will be in any conceivable future. The idea that a twitter mob could somehow ever be comparable in power to a government demonstrates a serious disconnection from reality and the power that governments exercise over their people all around the world and all through the annals of history.

That's a trick of definitions: when an organisation acquires particular kinds of power we call it a "government", regardless of how it's structured or operates. Historically it's completely normal for states to be taken over by religious groups, by social movements, by corporations, by criminal gangs. And in most of history there has been very little respect for freedom of speech, for the right to say unpopular things and not suffer devastating personal consequences, for the idea that there's such a thing as a loyal opposition, or an honest mistake. If we believe that's important - and I do - then it's not enough to just protect speech from governments. Maybe 200 years ago only governments and major religions wielded enough power to ruin lives, and so constitutional protections and strict laws on freedom of religion were enough. But in these days of megacorps and large social movements that's no longer true.

> So what are you proposing?

That firings should be required to be for cause, and there should be reasonable protections even for fired workers in terms of notice periods, payments and the like - similar to the way it works in France (or most developed countries really). That really vital things (in particular, healthcare) should be done by a provider that's obliged to serve everyone rather than a private company that's permitted to arbitrarily withdraw coverage.

> An asshole is not a "class" of people, it's a behavioral disposition.

Or it's anyone you don't like. In any case, the rationale for protected classes isn't about the fact that they're classes, it's about the risk of suffering coordinated action from others - which nowadays applies to people who've said unpopular things too.

> Nobody needs to be an asshole.

Nobody needs to be a particular religion, but that's still a protected class.

It's absolutely legitimate to have a F/OSS-licensed project with such a culture. But it shouldn't count as part of the F/OSS movement(s), any more than an MIT-licensed website about how you need to pay SCO for Linux licenses would be part of the movement.

> But it shouldn't count as part of the F/OSS movement

Why not? If the end result is a high quality piece of open source software, why do I need to care about the politics or behavior of the developers?

The only reason I can think of would be from a risk-assessment standpoint, i.e. is another censorious asshole going to waltz on in and damage the team that produces this product so that it dies and becomes orphaned.

Why anyone would want to contribute to a project they don't like the developers of is beyond me; either fork it or go do something else. It's not a requirement. This is not your livelihood, for the most part. Just walk away.

Maybe if censorious assholes could just step the fuck off, and let people be, this would be less of a problem. As far as I can see, they've caused far more problems and far more discontent than anyone posting aggressively has ever done.

You don't have to care, and I am not saying you should.

I'm saying it doesn't count as part of the F/OSS movement. Whether you want to contribute, not contribute, etc. is your own choice and doesn't need to be correlated in any way.

How do you define the F/OSS movement?

> But it shouldn't count as part of the F/OSS movement(s)

Why not? Who gets to determine what counts as being part of the movement? I thought that, broadly speaking, the movement was about the method of producing and licensing software. If a certain group produces F/OSS software, but is internally abhorrent, they're still part of the F/OSS movement.

The comparison to BDSM is incorrect. BDSM is about "abuse" that one enjoys due to established parameters. I don't think anyone at the receiving end of abuse within F/OSS enjoys the abuse and seeks it out. That's why the codes of conduct exist, to... bring F/OSS in line with the expected ethical standards of scenes such as BDSM, so to speak.

Certainly. I wouldn't expect a big, open-membership FOSS project to "get away with" being abusive in a non-consensual way. That isn't what BDSM is—which I had hoped was carrying some of the weight in explaining what exactly I meant here, by making you picture what an actually-analogous-to-BSDM software community would be like.

But my question still stands. Let me be more precise with it. Is Code-of-Conduct-space actually wide in practice, or is it narrow? Is there really only one societally-acceptable way to run an open-membership out-in-public software project, with the differences coming down only to the fine details; or is it actually possible to do something very different, e.g. having a CoC containing "We like to call each-other names here. Only join if you're into that."?

Would a project that claims different limits than the ones of wider society, be tenable? Or would it, and every other project attempting such a thing, be boycotted, simply for its members being weird and liking things people don't normally like (and therefore making it squicky to contribute to the software if you don't like what they're doing, which must therefore be a moral failing on their part)?

It think it would be difficult if it isn't feminist, in particular, considering prominent biases in the software development community.

What does "feminist" mean in this context? Many of the people who claim to be radically feminist on places like Twitter, etc. are in practice quite hostile to the bulk of female engineers and developers, especially to those who aren't very vocally aligned with their political stance. It's not real feminism, it's just extreme hypocrisy and pushing fringe politics.

I think it’s pretty clear that OSI does not want to be a “BDSM but we build something along the way” group.

Did I say they did? The GP poster just made a unilateral assertion, and I was trying to point out a case where the assertion is false. That case doesn't apply here; but nor should you implicitly accept the GP's conclusion (that ESR should be kicked out of this community), since one of the premises used in their argument (that all FOSS "must be kind") isn't true-as-stated.

When you want to do something as serious as exiling someone from cyberspace, you need a high standard of rigor in your arguments; and you need to be wary of thinking you can "read through" rhetoric to "what they clearly meant", without having false claims made for rhetorical purpose still affecting your judgement.

I invite the GP to restate their argument with more precision, and see if it still sounds compelling.

> So, is it illegitimate, then, to have a FOSS project with an intentional/cultivated "culture of abuse", where the idea is that nobody is kind to anybody and that's how everybody likes it? Sort of a... "BSDM but we build something along the way" thing?

It's nonviable for anything beyond a small hobby project, for the simple reason that such a culture would make it essentially impossible for anyone to participate in the project as a representative of their employer.

That assumes a monolithic design style, I think.

There are a lot of projects that are factored into small pieces with separate maintainership, where the project as a whole—or its "core"—might be corporate-sponsored; yet individual parts/plugins/connectors—including fairly-critical ones!—might turn out to be one person's hobby-project.

How many maintainers do you think the average Kafka connector has? The average Postgres extension? The average Redis or Nginx module? Etc.

These ecosystem components move the project-as-a-whole forward, and yet they don't necessarily need to be as "corporate" as the project-as-a-whole does to accomplish that. Even though, in the end, a corporation might use one of them, it won't really matter as long as nobody sees them doing so. (If they ever need to patch the un-corporate extension, though, they'll probably fork the thing and flense out all references to the original. I've seen that happen a lot, and done it a few times myself.)

> would make it essentially impossible for anyone to participate in the project as a representative of their employer.

Freedom to fork is a thing. Start a CoC-friendly fork of the project, for the benefit of those who require this as a condition of participation; but stipulate that both forks will freely share code between one another. (Meaning, no PNG-like clause whatsoever. This ensures that the choice between the "nice, CoC" and "sticks and stones" project leadership styles is kept free and open, with nobody being forced into one choice or the other. This of course is a compelling reason to actively reinforce the status quo of Persona-Non-Grata clauses in a copyright license being regarded as incompatible with the OSD or any other definition of FLOSS.)

In some cases that would be a feature.

> where the idea is that nobody is kind to anybody and that's how everybody likes it?

Crocker's Rules[0] as CODE_OF_CONDUCT? That's an interesting idea indeed. Especially if you want to actively preempt entryism by the usual faction of CoC-pushers.

[0] https://web.archive.org/web/20021202223742/http://sl4.org/cr...

> Note that Crocker's Rules does not mean you can insult people; it means that other people don't have to worry about whether they are insulting you.

People deciding to subject themselves to this does not allow them to be abusive towards others, nor would it be stretch to argue that for a project to be "open", there ought to not be a culture of actively being inconsiderate.

Whether FOSS implies open membership is a matter of semantics. FOSS can refer to the development model (bazaar vs cathedral), the community, attitudes ("patches welcome" as a way of saying "well do it yourself then", meant sarcastically as "patches welcome" is repeating the implicit), or ideological stance (blah blah freedom).

So I don't think "does FOSS imply open membership" even has an answer.

Maaybe, but you'd have to follow the "safe, sane and consensual" rules. Explain in advance that that's what you're doing. Ensure that everyone is on board. And not do it in public, because the public aren't consenting.

And if it rose to public prominence while remaining a men only club, that would probably cease to be OK.

(It remains unclear to me what the long term psychological effects of participating in a community where "nobody is kind to anybody" are, and what kind of trauma they are replicating)

> Explain in advance that that's what you're doing. Ensure that everyone is on board.

This assumes that people choose their way into this state. What if "software as performative BSDM" is the state of nature, that people are in unless they decide otherwise? (A bold claim, I know, but I can think of some good examples, and I bet you can too.)

> And if it rose to public prominence while remaining a men only club, that would probably cease to be OK.

What defines prominence? There are a lot of FOSS projects that are critically relied-upon by thousands of downstream projects, which nevertheless only have one contributor—and not even because that contributor doesn't accept PRs, but rather just because nobody else has ever thought to help them.

And then there are the tiny projects everyone ends up using because they're transitive dependencies of one thing or another. This comes up a lot in the Node.js ecosystem.

And then there are literal art-projects that people end up relying upon for some strange reason. _why's libraries were this kind of thing: open-source in a sense, but not for the sake of improving them qua software, but rather instead to improve them qua the original artistic vision, if you understood and shared it; and otherwise only open to view and learn from.

> What if "software as performative BSDM" is the state of nature, that people are in unless they decide otherwise?

This is a stupid claim and you know it.

> "This is a stupid claim and you know it."

-BZZT-, CoC violation detected. ;)

> but you'd have to follow the "safe, sane and consensual" rules.

Those rules are entirely arbitrary and were made up as part of this new Code of Conduct bullshit. They didn't used to exist and plenty of productive developers contributed plenty of useful code. You're arguing the rules matter from within the context of the rules; that's begging the question.

> long term psychological effects of participating in a community where "nobody is kind to anybody" are

99% of people toughen up and move on with their lives and learn to not be so serious.

1% of people don't manage to do this. Some subset of that 1% get mad about it, and now they're using this as a hammer against massive numbers of normal, volunteer developers who just wanted to happily code without having to acquiesce to a particular brand of American progressivist moral licensing.

99% of people don't participate in open source, and much of the behavior we're talking about wouldn't be tolerated at any Western employer (except among the most senior untouchables).

I don't think that bolsters your argument. Having been around at its beginnings, FOSS itself wouldn't have been tolerated at any Western employer at that time. Consider that it was the rebels who didn't care about conforming or what was good for business/PR that made FOSS what it is today.

In the beginning there was no "FOSS" only Free Software. Only after the invention of Open Source as a more business compatible and less radical version FOSS came to life.

https://www.gnu.org/philosophy/floss-and-foss.html

I doubt OSI has stated a "culture of abuse" as their goal, their CoC even indicates the opposite.

If someone and his three friends want to be politically incorrect, great, let them.

If they open up their little club to the world at large, without any indication that their club is "special", then I guess the world is right to assume some standards of behaviour.

Why does opening up your rude little club imply anything of the sort? Why would you join a club that you don't enjoy the discourse within? I don't understand how people can't see that participation in all of this is 100% voluntary. This isn't like inclusion and diversity efforts in corporate enterprise, this is a different beast, and over-corporatizing it defeats the reasons why many people moved to working in / on open source stuff to start with.

> Why would you join a club that you don't enjoy the discourse within? I don't understand how people can't see that participation in all of this is 100% voluntary.

The topic under discussion is the OSI's policies and the application thereof. It seems pretty evident to me that the OSI sees that first part (people not enjoying the way they're spoken to or about) as the potential problem: they want people to voluntarily participate in a group, and feel like others in that group can be expected to treat them with dignity and respect even if they disagree.

Furthermore, to your "100% voluntary" part: OSI just revoked their voluntary association with ESR. It cuts both ways.

I don't think it's unreasonable for a large group that is open to the public to publish some pretty standard "don't be a rude and inconsiderate jerk" rules, and to enforce them. As you've pointed out, they can always go fork if they don't like it. Maybe ESR can go start the 'R(ude)OSI' if he thinks that's a productive use of his time, but I don't think most people in general want to have to deal with having textual abuse hurled at them just to participate.

Who knows, maybe I'm wrong and such an effort would be successful! I encourage someone who believes that to try; I'm not against the words, I'm against being rude to people who wish to come and contribute who are shunned or discouraged or driven away as a result of it. Many hands make light work. I'd like to see free software take over the world, and I truly believe that it won't be able to do that if many/most groups creating and promoting it give first-timers a terrible negative experience, which has been the status quo since I was a child.

I tend to favor the rude crowd on principle, but I agree with the logic supporting OSI. More than just "a large group", they're a corporate brand (and one of the most important ones in the open-source world) and as a brand they have to think about their image. ESR doesn't fit their image. OSI wants to present a softer image. That's their prerogative (and I agree, probably important for the open source community to grow with new people who mostly start young and stupid - that's a good thing).

That said, there is probably space for a ROSI: a group with similar benefits to OSI but which prioritizes software over feelings. Being tactful, even for those who can and want to, costs time and energy. Dispensing with that saves time.