I always wonder what goes through people's heads when they write replies like this.
Like, do you not understand what "microblogging" is? Do you not understand how your content management tools work if you don't want to see things that I write?
You aren't even challenging me factually, you're just saying you don't care. Well, 1) I don't care about your opinion 2) I _do_ know how to block
Mate, what exactly are you hoping to accomplish here other than getting blocked and possibly reported?
https://media.hachyderm.io/media_attachments/files/111/183/511/498/675/932/original/54b388c00a65125f.jpeg
@134318c2 I really don't like the framing of "replacement" or treating stats as somehow "easier" at all.
Like I've seen people who have masters degrees in other disciplines trip over themselves with even basic probability and stats.
I'd love to see greater literacy here (I could see an entire class based around Utts's "What Educated Citizens Should Know" paper https://ics.uci.edu/~jutts/AmerStat2003.pdf )… but it isn't a capstone, a replacement for calc, etc… and as you point out we don't have the buildup.
Failure modes of #blocklists and fediblock
When a blocklist that didn't list its perspective on the issue upfront gets involved in an intracommunity dispute
Especially but not exclusively when people outside of a community get involved in intracommunity disputes
There's a fine line to walk on these issues generally—a lot of fine lines, actually, and a lot of nuance that I'm not going to get into right now—but the caution is that blocklists are _very easy_ to weaponize here
#fediblockmeta
1/
Between #IFTAS and #FSEP and #Fediseer let's also look at another threat model that I think people don't fully appreciate with #blocklists
How much do you trust the blocklist source—not its upstreams, but the actual place you get it from—to do what they are telling you it does?
How much do you trust the maintainer to not perform a MitM attack?
How much do you trust others who have access?
If a MitM attack _were_ performed, how would you know about it? How would you catch it? How quickly?
1/
Getting what I want for this data model to fit inside of #ActivityPub the way that it has been adapted to #mastodon and other #fediverse solutions is going to be a trick.
There's the transport model, but there's also just… how do you represent this.
We need finer grained controls than what are provided today by #mastodon for #blocklists
For instance, what if the tooling allowed individuals or instances to say:
* "Only allow direct messages from this instance from people who are followed or where there is explicit consent to receive a direct message"
* "Automatically hide all content (under something like what's called a CW today) from this server unless there's explicit permission to show it from the user"
* "Block replies to this post"
I can agree with the problem but still think that the solution causes unintended consequences.
I can agree with the solution but think that the implementation causes unintended consequences.
I can agree with the implementation, but think the context around how it is being used or proposed to be used does not address the problem (or causes unintended consequences).
It's hard to have these conversations when a bunch of racism and transmisogyny are being thrown around, however.
So stop that.
Part of the problem with #blocking in the fediverse, unique here as opposed to other social networks, is that there is a significant amount of "bycatch" in a lot of our instance blocking
If you block an instance you are blocking every member of that instance.
Maybe that's what you and your users want! That's what you are going for!
But it also means that to some extent _most_ servers are going to have a group of people who are essentially false positives: they aren't going to harass anyone
1/
Oddbean new post about login | logout
Notes by Hrefna (DHC) | export