There ends Wikipedia’s argument that it is independent and without bias or incentives.
Yeah, on one hand, they are so far only granting them API access, where the AI companies would have just scraped it ad-hoc.
On the other hand, they are now financially dependent on these mega corps.
It’s also a huge demotivator for people working on a collaborative project when large mega corporations are involved. Not quite 1:1 with fandom which operates for-profit but I can say almost any wiki on fandom is largely edited by anywhere from 0-4 people, and in many cases it is 0 because there is significantly less motivation to contribute to something that isn’t a wholly community project and is instead benefitting some corporation.
It will only get worse for wiki from this kind of starting point, more and more will slide in and if it does you’ll see a significant drop off in contributors.
I’d wager some contributors are already saying to themselves “I didn’t write hundreds of thousands of words for microsoft”.
Not quite 1:1 with fandom which operates for-profit but I can say almost any wiki on fandom is largely edited by anywhere from 0-4 people, and in many cases it is 0 because there is significantly less motivation to contribute to something that isn’t a wholly community project and is instead benefitting some corporation.
Oh yeah, I’m aware, I was involved with some Fandom wikis too in the past, even back when it was still called “Wikia”.
Unfortunately, for many communities there isn’t a good alternative, since not many people have the time and resources to host one themselves.
The thing is, if you solve that income-tech barrier then it will change the game. Most communities do want a wiki and will work on one in much higher numbers than fandom gets if they think it’s actually there to serve the community instead of make money for someone. So if that “I don’t know how” and “I don’t have money to host” and “it’s scawy!” barrier can be removed you’d see a lot more work go into basically all of them.
Maybe Ibis (federated wiki) actually does have a future!
Wow, that looks like it could also be a version of fandom.com that doesn’t suck.
Edit: just realized I skipped the comment from Awoo in the chain you’re replying to that mentions fandom.com, lol.
It’s probably fine.
It’s not like AI scrapers weren’t already scraping Wikipedia and using that data anyway, so Wikipedia might as well try to get paid for it.
“What are your thoughts on doing the same but in a pod-like thing that harvest your bio-electricity to keep data centers on?”
I expect to see a “ai summarize” button on Wikipedia in the coming months.
Wikipedia editors at least have rejected ai content afaik, so I think the actions of the foundation and the actions of the site will remain out of sync for the moment.
It is kind of amazing how little respect the foundation has for the works of their editors and how little they actually do for the site.
Yeah I wouldn’t be excited…
Quick-summarizing the article on Katyn as I pwn some tankies on Reddit without a single need for my own thought. Just lettin’ the LLM flow through me.
Probably better to make a deal and offer it via some dedicated service than webscraping but it’s funny that they’ve been running ads for ages talking dismissively about how AI “runs on Wikipedia” at the same time they made their first deal with Google :^|
Wikimedia named former U.S. Ambassador to Chile Bernadette Meehan its new chief executive, effective on January 20, Reuters first reported last month.
Lmao
Jimbo Wale$
