"We banned nearly 37M accounts between January and March of this year"
If you've ever wondered what it's like to fight spam at scale, here's what Reddit deals with.
When I occasionally say that I've very strong concerns about the Fediverse dealing successfully with abuse at this scale, this is the sort of thing I'm talking about.
@dredmorbius Probably not all spam, of course. There will be other vaguely defined violations of terms service bannings. (Remember Reddit also took down chapotraphouse and associated subreddits.)
@emacsomancer This chart shows takedowns of one specific spammer, which is obvious when reading TFA.
@dredmorbius As long as instances stay small enough for moderators to keep on top, and instance admins are proactive about blocking other instances that permit abuse, I am optimistic about the Fediverse's future. We don't have to deal with it all at once by ourselves.
@dl That was Usenet's model.
@dredmorbius Well, that's why there should be many dedicated instances, just like there are many websites: Of course there are scammy ones, but the WWW seems to filter itself fairly well.
@janek "Should be" is an ought vs. is wishful thinking argument.
Technology scales by providing efficiencies. Larger instances and centralisation will occur.
That's amplified by human nature which can keep track of only a small number of options.
The Web was promoted as a decentralising technology. It's proved otherwise.
There is no empirical basis to support what you're arguing.
I'm going to posit that if you're looking for a humanity-scale network, you need to be able to handle at least 1010 individual nodes. (Probably a few multiples of this given multiple roles, associations, and collective organisations, whether commecial, social, or other orientation.)
And that present Fediverse / Mastodon architecture doesn't accomodate this well.
@dredmorbius On the other hand, my account has been banned for a couple of years now, and no appeal has been able to resurrect it. I have never posted anything in the least bit controversial, and all of my posts have been on either a) FLOSS or b) educational fora. I am a University instructor and a FLOSS developer, so...
@elb One view I've come to have of online identity is that flushing profiles every so often may actually be a relatively good thing.
Yes, it gives the bad actors a second shot, but it also gives the socially-oriented an opportunity against erronious policing and moderation.
On the internet, everyone knows you're a cat — and that's totally okay.