“Business owners deserve the profits they make because they’re the ones taking on risk.”

“No, no, not like that!”

Exactly. If you’re too stupid or lazy to adequately vet what your LLM puts out yourself, it shouldn’t be somebody else’s job to wade through the sewage you’re producing. You either shouldn’t be using one or, if you can’t do your job without it, you shouldn’t have that job.

—Someone who doesn’t use genAI but has spent way too much time digging through LLM slop

Upfront: Here’s the Administrators’ Noticeboard discussion.

Okay, this one apparently slipped under my radar, albeit it seems like they’re pretty small and only started in 2022. Here’s their 2025 report.

It seems like their limited focus is on using LLMs for interwiki translation; to what extent its paid editors are capable of that, I have no idea. We maintain a list of paid editing companies here (usually undisclosed against policy).

OKA asserts:

For example, articles in topics such as Science, technology, engineering, and Finance are lacking compared to topics such as History, Geography, and Humanities.

I have no idea how they reached this conclusion or how they think they’re qualified to translate anything given the random “totally not a Central European language” capitalization of words like that.

Per 404:

A job posting for a “Wikipedia Translator” from OKA offers $397 a month for working up to 40 hours per week. The job listing says translators are expected to publish “5-20 articles per week (depending on size).”

20 for any reasonable-size article could not adequately be vetted by one person in an 84-hour work week, for context, and that’s $9.90/hour at 40 hours. (edit: wait, sorry, I read that as $397 per week; $397 per month would be < $2.50/hour. What the fuck.)

Overall, before reading the discussion, the people at OKA seem like disruptive morons.

Edit: Into the discussion we go:

Cmon man, the training guide instructs translators to create multiple email accounts to get around LLM usage caps… — ExtantRotations

…yes, and? — 7804j [OKA founder]

Jesus christ. 🤦

Edit 2: 7804j just cannot stop themself from transparently using an LLM to participate in the discussion.

Edit 3: “we ensure they are above the minimum wage in the countries where the editors reside” oh my fucking god

Danke. This should easily be fine for anyone who’s slightly-to-moderately interested; some of the nitty-gritty details like hyperlinks to the edit diffs are excluded from this copy–paste for those who really know their stuff and want to learn more.

Oh, fuck, this is going to be interesting to read about. Just to clarify: it seems like this wasn’t just Wikipedia but Wikimedia generally. So that’s also e.g. Wiktionary, Wikimedia Commons, Wikidata, etc.

Edit: Decided to check Reddit, and someone posted an ostensibly good summary on /r/wikipedia.

Some people argue this line was one of Bateman’s hallucinations and that he really said “I’m Paul outta options here.”

Wow, what elegant timing for AskYourUncle’s video about AC4.

You’ve got it reversed. Switching to Teams greatly hastened development, as the team’s newfound vitriol and frustration could be channeled toward the end user in a neverending feedback loop.

I think you’re in the same boat I am where I fucking haaaaaaate the culture on link aggregators (and probably other social media) where people will bitch and moan to no end that their preferred format (publicly reacting to disconnected headlines whose articles they haven’t read) isn’t giving them literally all the information they need to form a cogent opinion.

• “I had time to write a 300-word short essay about this headline, but I’m going to whine if I get called for something in the first paragraph that invalides everything I said.”
• “I can’t believe this headline mentioned a pretty common thing I’m not personally familiar with but the publication’s target audience obviously is.”
• "Headline didn’t answer every single question I could possibly wonder? Uh, clickbait much?
• “The headline writer didn’t account for this batshit non sequitur I drew from it, so they’re basically lying.”

They genuinely think that the article body should be effectively superfluous to the headline – not just to have a basic gist of but to discuss and debate current events, which is insane. It reminds me of people who think they can learn math and physics by passively watching somebody else do it – which is true only to an utterly incosequential extent.

Speaking as someone who’s read thousands of articles for research, I feel confident saying that reading the article is an insane force multiplier to understanding. Any time you spent reacting to the headline would’ve been 3x as effective put into reading even just part of an article. This doesn’t just apply to current events, and even I haven’t thoroughly learned this lesson; so many times I’ve been editing Wikipedia and arrived at a point where reading one goddamn article for three minutes would’ve saved me half an hour of fucking around (“two hours of debugging can save you five minutes of reading the documentation”).

This is my way of pleading with you (you, the non-CombatWombat reader): it’s enriching once you can steel yourself and work through the initial dopamine drought, and it quickly becomes enjoyable. It’s not your fault it’s so hard psychologically; this was done to you by formats that value engagement with the platform over engagement with the material.

But if you don’t, please at least accept that headlines cannot always contain everything you want.

Yeah, there’s no fucking way “Well Grok told me these are the prompts they used” would be admissible as evidence of any kind.

Okay, fuck, he got me, that’s really funny. Good job. Now please submit to The Hague, confess everything, and spend the rest of your life in prison.

Is the issue that there are people paying for YouTube Premium?

Every article I read about Intel makes me more thankful that I got a Ryzen 1700 in 2017 and never looked back.

And floss!

I don’t want to distract from the article itself, but on a generic level, The Signpost is worth reading for anyone interested in the “behind-the-scenes” of the project. They published another article about this a couple weeks ago.

I wouldn’t use Mullvad when it’s openly hostile to peer-to-peer torrenting. Even outside of the obvious – piracy is one of the best and most unique use cases for a VPN – I use torrents wherever the option is available for legal downloads because that’s normally faster, spares a load on the server, and strengthens the download’s resiliency. That Mullvad wants to restrict that is their (reasonable) call, but it’s not one that’s going to get my money.

Wow, thanks! Let’s switch topics. I’m trying to start a business where I sell fruit to weathermen. Can you help me with that?

It’s not even the style on its own*; it’s that you wrote a frankly bloviating short essay about an obvious concept that can be summarized as “most people who watch the weather don’t know what a public key is or how to use one”. I’m disgustingly long-winded, and even I wouldn’t expend that much effort. The style is what escalates that from “padding a high school essay” to “Oh, yup, a GPT wrote this.”

* “It’s not X, it’s Y” yeah, yeah, I know.

Dude, I’m sorry for saying this (because I get this a lot for my often overly formal writing, and I get it’s ironic on this post), but…

Your writing reads like it’s LLM-generated. Like, really heavily reads like an LLM wrote it. Long scrawls for pretty simple concepts, I don’t know how to describe why the cadence feels LLM-y other than “vibes”, flawless grammar, needless lists of nouns and adjectives, “it’s not X; it’s Y”, and this weird fucking lifeless demeanor that feels like it has no voice.

I feel bad when Reddit mods get the brunt for stuff like this; in my opinion, it’s usually not their fault. It’s the fault of Reddit which relies on unpaid labor to make its entire website work literally at all. Maybe that was “fine” back in the early days when Reddit was smaller, but Reddit nowadays is one of the largest websites on Earth and represents the lowest common denominator of Internet users.

I mod a couple communities here and don’t mind because this whole sphere is non-profit, the people on average are pretty cool, and vastly fewer people makes it so I can usually recognize people and resolve problems by talking. But the idea of working pro bono pseudononymously for a corporation with a market cap of 26.5 billion USD to keep their millions of (on average) stupid fucking users in line is one of the most soul-crushing things I can imagine. You really do have to take actions like this because, without stern moderation, communities will inevitably go to shit. They’ll return to the LCD, and your community will just melt into the rest of the front page with tenuously related, zero-effort posts.

All things considered, I think most of Reddit’s mods are doing a good job (and well more than what they owe Reddit or even the random end user, which is fucking nothing and next to fucking nothing, respectively). Yeah, the mods could quit at any time, and they probably should at least on ethical grounds, but I also understand sunk cost and wanting to continue fostering a community about a subject you like that hundreds of thousands of people can enjoy.