Olives @olives@qoto.org

Let's try ANN for a moment.

First, there's a random captcha to even access the site. So, let's engage with that for a bit... Then, there's a video hovering in the corner, but to close that, you first have to scroll through the cookie consent prompt, otherwise it will attempt to set hundreds of cookies.

The video appears to be about online trolls harassing someone. So, Internet drama. Closed that.

Then, I find that there is no article that I'm that interested in. Scroll all the way down. Nope. Digging into the archives. Okay, one article potentially relevant.

Coverage of financial censorship seems poor, including the meeting with the regulator to express concerns about that.

I don't have much of an opinion on Niche Gamer (the site appears to have improved lately). They write stories I find relevant and I cover them here.

If you have other potential sites I could check, I don't mind checking them though.

For a company which collects so much data, Facebook is not terribly transparent.

The bigger issue is that Facebook runs obscure "data sharing" programs and no one really understands what they're doing. Then, they come out with language that is about as clear as mud.

So, when I see Facebook using weird suspicious language, I tend to get suspicious of that.

There was a lengthy tortured term which made me just think "wtf".

"realism" (quotes) came to mind because Facebook used the term in a suspicious manner a month ago, although it was just an example off the top of my head. I've said it before though, "realism" is a distraction which leads to musing over arbitrary distinctions (degrees? elements?) of "realism", rather than whether it is actually consensual / a fictional character.

In the case of NCII, I noted earlier this year that tech firms seemed to want to pretend that any false positive is "not actually a false positive", because it "looked close enough" by changing the definition of NCII to include the false positives. It's an absurd attempt to evade scrutiny of their moderative practices.

It looks like Bluesky hired a bunch of contractors lately, so if the moderation shifted for the worse (more censorious), that might explain it.

Instance is bugging out again.

I see a policy guy moved from the U.K. to Brazil.

Artificial "Intelligence".

Be careful about assuming why Bluesky is censoring any particular thing, particularly when this assumption shifts responsibility away from the company.

A lot of the time, it really is a bad managerial decision, also, assumptions can serve as a distraction from combating censorship more effectively.

Let's suppose a company quietly decided they didn't like AI, because they reckon a company didn't get permission from the artists whose works the model was trained on.

So, inexplicably accounts which post such content inexplicably start getting permanently banned but no one really knows why. That is what Bluesky's moderation is like.

One of the issues with Bluesky not actually being decentralized is that they can randomly start engaging in rather arbitrary forms of censorship.

So far, the company doesn't seem to operate any more clearly than any other Silicon Valley firm.

Once again, if you are complaining about censorship in some obscure place on page ten, chances are we are not going to see it.

https://qoto.org/@olives/113361636922921463
https://qoto.org/@olives/113430265249215742
I should get back to working on this master post (which includes "porn is spooky" scientific debunkers) sometime. I haven't posted it in a while so I'll do so again.

If Bluesky is engaging in puritanical censorship, is so-called "realism" an apparent factor in the censorship? Or is it more general than that?

I'm curious if there are any particular patterns in the puritanical Bluesky censorship. Not to excuse it, of course, but to figure out how to combat it more effectively.