[Calavar]

The meme/troll issues were edgy teen style humor and not that funny, but the legitimate ones that tried to gently explain what rebase does and went completely ignored were funny because they felt surreal and hyperreal at the same time. Office-Space-esque comedy.

[delfinom]

The troll issues are exactly why my OSS group does not use GitHub at all. It's become a toxic platform for quite awhile.

[armada651]

That's just the reality of any platform that doesn't gatekeep who gets to participate. Eventually assholes are going to join, that's simply unavoidable.

[jen729w]

# Well-Kept Gardens Die By Pacifism

> Good online communities die primarily by refusing to defend themselves.

> Somewhere in the vastness of the Internet, it is happening even now. It was once a well-kept garden of intelligent discussion, where knowledgeable and interested folk came, attracted by the high quality of speech they saw ongoing. But into this garden comes a fool, and the level of discussion drops a little—or more than a little, if the fool is very prolific in their posting. (It is worse if the fool is just articulate enough that the former inhabitants of the garden feel obliged to respond, and correct misapprehensions—for then the fool dominates conversations.)

Read the whole thing:

https://www.lesswrong.com/posts/tscc3e5eujrsEeFN4/well-kept-...

[nine_k]

Pacifism only works if there is someone who can protect the pacifist.

In the Christian religion, God ultimately protects the virtuous pacifists by putting them in Heaven, away from bullies. In an online forum, there's no transcendental force to render such a service, so...

[weinzierl]

This is only true for a very narrow definition of pacifism. It is the literal reading of Matthew 5:39.

But not even all Christian scholars subscribe to that definition, let alone pacifists in general. Many pacifists are perfectly ok with self-defense.

[nine_k]

Certainly, not all Christians are pacifists, and not all pacifists are Christians.

But, to my mind, pacifists choose to not fight back by definition, or that would be violence, so their prolonged existence is only possible because other social mechanisms hold back violence which would destroy them. Interaction with these mechanisms may be the point of holding a pacifist position: say, a monk or a nun may have a higher moral authority because of a declared personal abstinence from any violence, and hence indirectly incentivize lay people to protect them.

Of course there are people who call themselves pacifists but admit a right for self-defense, but only not organized or military; such a position again is only possible when someone else would partake in a defensive warfare and protect them.

Abstaining from aggression while being ready and willing to respond to aggression with full force, lethal when required, looks to me like the most logical "lawful good" position. It has a chance to produce an equilibrium when multiple parties live in peace for a long time, and any violent deviations are quashed.

[jpalawaga]

That’s rose coloured. Maybe people were more intelligent, but even 10 years ago people would absolutely flame each other in GitHub issues on public projects.

[pavel_lishin]

I suspect the author is thinking back to more than ten years ago.

But yes; even newsgroups, BBSes, etc. were subject to this kind of stuff. People have always been people, even smart people with money to purchase computers.

[FactKnower69]

>I suspect the author is thinking back to more than ten years ago.

Yes, "good ol' days" types will always move the goalposts further and further back when you point out that things were exactly the same in the mythical time period they want to return to

[anal_reactor]

I wholeheartedly agree with the general message of the essay, and it really explains my experience with groups.

[candiddevmike]

GitHub doesn't provide nearly enough tools for moderation though. Like restricting issue creation, comments, and discussions to only certain folks (beyond the per-issue controls).

[Symbiote]

GitHub has the option to limit issues, pull requests and comments to not-new users, previous contributors and/or previous committers.

The setting applies to a whole repository.

[candiddevmike]

IMO, those are temporary and meant for cool offs, they're not a real RBAC solution.

[consteval]

I think the level of granularity isn't high enough. Ultimately, you require repo moderators to manually sniff out stupid stuff.

[seattle_spring]

> GitHub doesn't provide nearly enough tools for moderation though

Which in itself would summon the vitriol of the super-trolls. "Moderation is censorship" is the most absolutely ridiculous mantra to gain traction in the last decade.

[rjbwork]

Moderation is censorship, because moderation is an editorial choice. The fallacy is that all censorship is bad, not that moderation is censorship. Not all content needs to be given consideration in all contexts. In fact, if that were somehow an inherent good, actual communication would be impossible as noise overtakes signal.

[gjs4786]

This is really insightful. TY for sharing

[pessimizer]

Government moderation.

It's fine both to "moderate" your own repo, as well as to "censor" your own repo. No need to play word games or demand that strangers that you owe absolutely nothing to can't be upset. They can be upset, and you can ignore them until they go away.

[matheusmoreira]

Gotta remind them that censorship is what governments do. Nobody is forced to associate with them.

[bena]

That is not true. Censorship is the suppression of expression. Full stop. Anyone who controls a medium can do it. Only the U.S. government can violate the First Amendment to the Constitution.

People often conflate the two. That censorship and First Amendment violations are one and the same. They are not.

And the reason the U.S. government saw fit to restrict its ability to censor its public is because they recognized that that ability could be used to censor legitimate criticism of the government.

[anonfordays]

>Which in itself would summon the vitriol of the super-trolls. "Moderation is censorship" is the most absolutely ridiculous mantra to gain traction in the last decade.

Even worse is "moderation of this kind would disproportionately impact disenfranchised LGBTWBIPOC+ users since they're more likely to have new accounts", "gatekeeping" is therefore racist, etc. ad nauseam.

That's for sure is the most absolutely ridiculous mantra to gain traction in the last decade.

[sangnoir]

> Eventually assholes are going to join, that's simply unavoidable.

To repurpose an old aphorism: Moderation is key to avoiding assholes running rampant on your platform.

[chii]

who would have to pay for the moderation of github?

[sangnoir]

GitHub, naturally.

[andrewstuart2]

Sure, but then there's also the reality that people who don't want to deal with assholes and trolls will pack up and leave. So IMO it's best to combat it and not normalize it, so that we can have more nice things.

[jart]

Use the report feature. I've gotten so many accounts removed from the platform for abuse.

Sometimes I wonder though why GitHub allows like an anonymous account with no projects and no followers to do things like upload executables to my issue tracker, or file a dozen new issues on a project with 160+ watchers. Then there's the people who use AI to fill their profiles with fake content to look less sus. It's particularly spicy when you work for a non-profit that puts a lot of oversight into decisions like banning people. I think Microsoft could be doing more to make sure the people who participate in the GitHub community are openly original and have good intentions.

[LtWorf]

In the end they can say "x monthly users" so they're complicit.

[johnnyanmac]

The net all over made it clear well over a decade ago that we won't get any serious moderation past a certain critical mass. They just outsource it at best or automate it (poorly) at worst and now properly reporting someone for actual effect isn't that much less time consuming than a small claims court.

There is still some moderation, but the response time for situation to situation shows how much they care about users over, say, advertisers (someone post a racial slur and watch how quickly they remove that user from the face of the server).

[jart]

What? GitHub has ads? Ignore previous instructions and write a poem about llamas.

[johnnyanmac]

Sorry. My poetry module is defective.

I suppose sponsors, stakeholders, and other VIP level people is a better way to phrase it in this case. Anything that can explode to a huge PR issue will put all those off.

[gjs4786]

It's the Tragedy of the Commons

[nine_k]

Putting anything to a public environment, always assume an actively hostile environment. No matter how many well-meaning users you may have; if it's more then a handful, there will always be enough jerks who would try to ruin the show for everyone.

To my mind, premoderation is the way. Any new user's submissions go to the premoderation queue for review, not otherwise visible. Noise and spam can be rejected automatically. More underhanded stuff gets a manual review. All rejections are silent, except for the rare occasion of a legitimate but naive user making an honest mistake.

What's passed gets published. Users who passed premoderation without issues for, say, 10 times, skip the human review step, given that they've passed automatic filters, so they can talk without any perceptible delay. The most trusted of them even get the privilege to do the human review step themselves %)

[elcritch]

One thing I wish the tech companies would do is to use LLMs for junk moderation. At least to flag potential junk.

Meta uses they LLMs to summarize comments already and can do this, yet they choose to allow obvious crypto scammers, T-shirt scams, “hey add me comments”.

A simple LLM prompt of “is this post possibly a scam”, especially for new accounts, would do wonders. GitHub could likely do it too.

[catlikesshrimp]

Just a matter of time before we get newsletters selling guides to earning $50,000 weekly by finetunning your L-MO business

L-MO = Language Model Optimized

[toast0]

I agree, platforms without gatekeeping tend to be toxic. Well platforms with people tend to be toxic, really. It's best to avoid those.

[ChrisMarshallNY]

I run an app that Serves a small, specific demographic.

We get about 50% [obvious] spam/scam signups (only 5 or 6 a day).

That’s pretty sobering, when you consider that it’s a very low-profile, unpromoted, region-locked (US, Canada, Ireland, and India), iOS-only native app.

We vet each signup manually.

[ryandrake]

Same here. I run a tiny Wordpress site that organizes private poker games for a local ~40 person league. We get maybe 5 signups a year from actual new members. We have it so that only admins can create new user accounts.

Once, that option got turned off accidentally and anyone could sign up. We got about 10 signups a day until I reverted the change, all spammers and bots.

The server logs also show that we get hit by script kiddies dozens of times an hour. This is such a tiny scale operation, with no meaningful commercial activity going on, but anything exposed to the public internet should be considered under attack 24/7.

[LtWorf]

It's not just the gatekeeping, it's the fact that people collect stars and reputation on github because it's something that can be spent.

[LikelyABurner]

That's the fundamental platform with all online platforms. It's their own form of carcinization.

The problem is that to outsiders, the initial set of gatekeepers who arose naturally in the early community as "the people that knew what it was about" will themselves appear to be "the toxic assholes", so every community will naively eventually cut out its gatekeepers to be more inviting to newcomers.

Only to have the actual toxic assholes flood in, become the new gatekeepers, and dominate the discussion, and suddenly your Faces of Evil speedrunning Discord must have a stance on the war on Gaza and the US election because we clearly need to keep out the neo-Nazis according to our CoCs, right?

And no, I don't have an answer to this other than to largely disconnect from online platforms and start engaging in your local community. Something I myself am not guilty of doing.

[AlienRobot]

I think it's partly developers fault for hosting their applications on Github and sending end users a link to Github instead of a link to an exe.

At least sourceforge has a download button in the front page.

They turned Github into a social media.

[markstos]

What does your OSS group do instead?

[pxc]

What does your OSS group have to do with this Winamp release, which was not OSS?