[bawolff]

The problems mentioned aren't even real problems:

- jimbo making CoI edits. He got caught, and edits were reviewed for appropriateness. System worked as it was supposed to

- college drop out on arbcom (for those not in the know, arbcom is kind of like an internal court to solve user disputes. They are not allowed to solve content disputes or say what an article should say, only user behaviour problems). How is that a problem? You don't need a degree to mediate user disputes.

(In fairness, the Gibraltar thing they mention was pretty bad)

Wikipedia certainly isn't perfect, but for some reason they chose some of the silliest controversies.

[chris_wot]

There are now hundreds, if not thousands, of decent editors who have stopped contributing due to toxic, non-productive people. This could be a way for them to continue contributing in a less toxic environment.

The problem with Wikipedia is not consumption of content, but the contribution of content.

[tux3]

Wikipedia has a similar problem to Stackoverflow, though nowhere near as bad, where the active community members really care about rules and have a whole established process and tooling for efficiently dealing with new contributions that don't necessarily meet that bar.

It all sounds utterly reasonable from the point of view of the community, who is most exposed to very low-quality content, spam, and vandalism. But newcomers mostly see a big bureaucratic machine rejecting their first attempt, per compliance with some long established policy whose full printed details could threaten a rainforest.

The problem is the rules are often (not always) there for a reason, and everyone involved has good intentions (assume good faith! you can generally assume good faith!). But it's definitely not always a pleasant experience for new users, and that's not an easy problem.

[Loughla]

The problem with this, with Wikipedia specifically is the good faith part.

It isn't adherence to strict rules that is the problem. It's the massively toxic little kingdoms that have become established among power users/moderators.

Like it or don't. Them is the facts.

[The_Colonel]

> It's the massively toxic little kingdoms that have become established among power users/moderators.

I wonder how a distributed architecture where each king operates its own kingdom is going to help with this problem.

(And yes, it is a problem, but also essentially a part of human nature. You can't fix it, only mitigate it)

[tux3]

On the other hand, it is very democratic. With many of the flaws of a democracy,

If you see a kingdom, in principle and in practice the emperor has no clothes. No single person has any special power or ownership over articles. If someone acts that way, you get to tell them what you think of it, if you like, and you'll very much be in the right. And if you are, others should agree and side with you.

There aren't moderators as such, people are supposed to talk to each other, directly. If you don't like how someone seems to act like they own the place, it won't change unless you tell em how you feel!

The thing is people who aren't afraid to give their opinions tend to talk over quieter people. I'm afraid I don't have a very good cure to offer, but please don't feel like anyone's too big of a power-user to have a little respectful, civil chat with. Please, template the regulars! They either respond politely, or they lash out and make an ass of themselves in public.

The problem is the rules apply equally to everyone, but the power users know exactly where the line is and how the bureaucracy works. "The law, in its majestic equality, allows administrators and newcomers alike to revert edits per policy, report 3RR violations, and open AN/I threads about people they don't like"

[chris_wot]

Except you get mobs forming on AN/I and they allow toxic users to fester. BrownHairedGirl was allowed to fester on the project for years. Innumerable productive members left or were banned as a direct result of her and her supporters.

[vkou]

That makes it no different from a run-of-the-mill legal system.

Which, for their flaws, do sometimes produce results.

[BobaFloutist]

It's the ancient mass-internet-moderation problem. I have yet to see a system that, even with the best of intentions, is able to do both of the following:

1. Be sufficiently hardened and responsive to mass bad-faith attacks, from trolling to toxicity to coups. 2. Be gentle, welcoming, and patient with newcomers, making it easy to join the community and learn the norms.

Most systems fall somewhere imperfect on the spectrum between the two, with rare exceptions going almost entirely to one extreme or the other.

[chuckadams]

I have corrected typos on Wikipedia and seen them reverted immediately without comment. Decided pretty quickly it wasn't worth the effort anymore.

[the_biot]

> There are now hundreds, if not thousands, of decent editors who have stopped contributing due to toxic, non-productive people.

While this is likely true (though not quantifiable), I would add there is also a pattern I've seen play out where somebody gets really into Wikipedia, starts stablishing turf on some subject, goes a little power mad, then gets knocked back by rules designed to keep such people out of power. Said editor then leaves Wikipedia in a huff, and a prolific contributor is lost -- but it's broadly a good thing.

[j_maffe]

Care to share details/sources for this claim? I haven't experienced any of this toxicity while editing there myself.

[matt_heimer]

I think AARoads might be a good example. https://slate.com/technology/2023/12/wikipedia-road-highway-...

To me the main argument for decentralized wikis is when a community wants to capture deep subject focused knowledge. It's why I created https://wiki.osdev.org. Wikipedia is great but at it's core it will always be an Encyclopedia with a large breadth of knowledge. Specialization in a field is best handled elsewhere with links out from Wikipedia if possible. Especially when there will be original content.

Hyperlinking is all the decentralization you need in some cases. Activitypub for wikis is of interest for people contributing to a collection of specialized knowledge. No one is going to subscribe to all the specialized wikis of the world except maybe search/aggregation systems.

[bawolff]

Ultimately though, i still think the existing system works well here.

Wikipedia wants to be a specific thing with a specific scope. Where that line should be is debatable, but it will eventually be drawn somewhere. No matter where you draw it, someone will be on the other side of it.

However, mediawiki is open source. The licenses of content are cc-by-sa (similar to gpl). You can start your own wiki. Similarly licensed content can be moved back and forth if rules change.

Perhaps this is just manual "federation" in a sense, but the ecosystem supports it. I persinally believe this is one of the reasons why mediawiki as an open source project should be a core part of Wikimedia's mission.

[account42]

Wikipedia's notability requirements in particular are quite arbitrary. It doesn't make sens for an online encyclopedia that isn't limited by physical space restrictions to not document everything. If you must, layer curation of what is notable on top of that.

[tux3]

The main concern Wikipedia has with notability is that if you don't have reputable secondary source writing about the topic, then you won't have references for any claims in the article. People just starting adding things that they know, or that "everyone knows".

When people disagree about content, Wikipedia always falls back to just reflecting what reliable sources says. And if they disagree, people can always collaborate to try to give due weight to both viewpoints.

But if there are no reputable sources that have already written about the subject, articles risk becoming someone's personal blog about their favorite topic. The notability criteria isn't a bar about what's important enough or 'deserves' a Wikipedia page, it's a super practical matter about verifiability, content disputes, and generally just being on solid ground if any claim is challenged.

[lupusreal]

But "has reputable secondary sources" isn't even close to Wikipedia's real 'notability' determination.

[The_Colonel]

There are many practical reasons to limit the scope:

* keeping at least minimum quality standards

* avoiding abuse by storing your files disguised as some "knowledge". At some scale the organization is not able to check even for a trivial instance of abuse.

* abuse legal problems by publishing illegal content (again, nobody will be able to do even basic checks if there's no limit)

* if anything goes, you get limited by physical needs like storage

> If you must, layer curation of what is notable on top of that.

Wikipedia is the curation. Store the non-curated data elsewhere.

[DonsDiscountGas]

It could be, in the same way that drinking a cup of tea every morning could make me make rich somehow. Why do you expect a federated system to be better in this respect?

[bawolff]

Wikipedia has certainly not solved the problem of getting huge groups of people to get along (especially when the people aren't getting paid and are instead presumably motivated by ideology). However i don't think anyone else has either.

[arp242]

A decentralized wiki with little to no oversight sounds like a significantly more toxic environment to me. All those assholes and bad faith editors will be on your new decentralized network too.

[DonsDiscountGas]

> How is that a problem? You don't need a degree to mediate user disputes

You probably shouldn't be lying about having one though

[BeetleB]

> jimbo making CoI edits. He got caught, and edits were reviewed for appropriateness. System worked as it was supposed to

Here's a better example:

https://en.wikipedia.org/wiki/Talk%3AAlan_Dershowitz%2FArchi...

Jimbo essentially moved the whole history of an article elsewhere, and rewrote it. So if you clicked the History on the article, you wouldn't see the prior edits.

[arcfour]

> jimbo making CoI edits. He got caught

I mean he made the edits under his own name, not sure there's much to "catch" there. This made it sound like he was editing via a sockpuppet.