Taylor LorenzCan anyone who understands how this platform works better than me explain why we can’t have crucial harassment tools like restricting replies to mutuals only, or mass blocking? Also any reason we can’t get link previews on here? Just curious if these are things that can ever be implemented 🙏🏻
TJ@taylorlorenz defer to others on the link previews (which should work on some interfaces) but if you're getting abuse from a certain instance, your instance admins should be able to ban that instance entirely.
Marko Vodanović@TJ @taylorlorenz
New here as well, so there's questions... Ok, mods can block entire instances, great. But can't abusers simply spin up their own private instances and repeatedly abuse people? Or is it not that simple to do?
New here as well, so there's questions... Ok, mods can block entire instances, great. But can't abusers simply spin up their own private instances and repeatedly abuse people? Or is it not that simple to do?
@vodamark @taylorlorenz Seems like it'd be a lot of work for a single toot, but may be feasible. @mattsheffield any idea?
Tracy HallIn concept, yes, but it isn't *that* trivial (nor free) ... some earlier admins have told stories of similar action, but I doubt it would be effective at the moment.
Ms. Penny Oaken@vodamark @TJ @taylorlorenz a few years back there was a group that was spinning up distinct *domain names* and pointing them at the same server in an effort to evade a block; I never caught the back half of that incident so I don’t know how exactly they’d mitigate it, but it would likely generally parallel user ban evasion cold wars.
@PennyOaken @vodamark @taylorlorenz I think the bottom line here is that Mastodon, unlike Twitter, is a community-based enterprise, and unlike some Twitter Safety deus ex machina, we rely on each other, and our instances, to manage spam and trolls - perhaps some holes in the short term, but in the long term far more effective. Yes?
@TJ @vodamark @taylorlorenz Yes. Twitter’s “Only People Mentioned Can Reply”/“Only Followers Can Reply” was:
1: developed to mitigate noise/spam/dog piling/harassment/protests without dedicating policy (& human labour for actioning) to AUP violations;
2: needed a central technological access control system to enforce it.
If it were written into the specification … it might not be honored by some instances, but that could be a defederation condition.
@TJ @vodamark @taylorlorenz ActivityPub (the protocol the Fediverse / mastodon is built on) has “Only Followers Can See” for individual items, which - combined with the ability to restrict your followers to only the accounts you approve - serves a similar anti-noise, anti-harassment purpose.
One-(or Few)-to-Many, Replies Restricted (“broadcast” style), though …
One-(or Few)-to-Many, Replies Restricted (“broadcast” style), though …
Schuyler D@PennyOaken @TJ @vodamark @taylorlorenz
I love the idea of defederation conditions that are "fast"-- ie if you're not respecting X then you're clearly not respecting the basic tenets of the platform.
It makes test suites more important for alt implementations but it also solves renegade bots and more
huxley(fur) / Laelia 🔜 Furcationland DeadDog@PennyOaken @TJ @vodamark @taylorlorenz "mute this thread" could be done on the receiving side right now
@huxley @TJ @vodamark @taylorlorenz my instance and/or client definitely has “Mute this conversation”.
One of the use cases for restricting replies is to counter & prevent misinformation from piggybacking on a conversation or announcement with a wide distribution / reach, so in such a case “Mute this …” would be insufficient.
pont@PennyOaken @TJ @vodamark @taylorlorenz
Couldn't the instance you are on filter replies that are not from people you mentioned, or followed ?
At least then the OP doesn't have to filter through the junk themselves.
robryk@PennyOaken @TJ @vodamark @taylorlorenz
> 2: needed a central technological access control system to enforce it.
Not really. We already can kinda have it in some way:
a) the canonical view of the thread is served by your instance, so it can refuse to publish some replies there (using whatever internal logic it wishes to use),
b) when a reply is sent to the OP's followers, if the OP's followers list is not public, it's the OP's instance that does that forwarding (and can choose to refuse to do so based on any internal logic it uses).
If OP's instance is Mastodon/Pleroma/Akkoma and blocks (suspends) the domain of the replier, both of these things will happen. I don't know which kinds of other blocks (incl. in other instance software) will currently cause which subset of them to happen, but would hope that OP blocking the replier would also cause both to happen.
This obviously doesn't prevent the replier from sending that message to e.g. people explicitly mentioned. Alas, they could send such a message as a straight-up new message instead of a reply, and no reply-blocking would help with that (point (a) above already deals with visibility of that as a reply when viewing the OP's post).
lord pthenq1@PennyOaken @TJ @vodamark @taylorlorenz
Mastodon is good like it is. If you don't like to be questioned, answered, etc, please close your accounts or block everybody.
Mastodon has the tools so you don't have to depend if a dictator to tell what to say, what to think, etc.
There were versions of mastodon that did that. They were defederated.
If mastodon does not provide the features you like, it is better to migrate to another networks.
@PennyOaken @TJ @vodamark @taylorlorenz
Mastodon has far more moderators than the birdsite.
3 o 4 moderators average by instance, times 3000. Besides everyone have powerful moderation tools.
Everybody can block an user or entire instances and they are just gone.
@pthenq1 @TJ @vodamark @taylorlorenz There are a few use cases where it makes sense for someone to make publicly readable posts which have comments turned off. Example: the leader of a political party or country makes an announcement. Terrorists use the replies to deliver a bomb threat. The bomb threat causes a panic, leading to humans being injured or dying in a crush. Countering & preventing terrorist amplification is legitimate.
@pthenq1 @TJ @vodamark @taylorlorenz
Everything open will be exploited for fun, politics, and profit;
In an active crisis there is no real-time knowledge, only real-time information.
Mastodon should be about knowledge. As more people join, its exploitability for (mis)information scales exponentially, and infrastructure which addresses that should be planned for.
Thanks for coming to my ModX Talk
@PennyOaken @TJ @vodamark @taylorlorenz
Perhaps. There're tools to avoid that already in place. We were being attacked for all kind of trolls. They could not yet solve mastodon.
We will see. Even in your scenario it can be easily muted.
@PennyOaken @TJ @vodamark @taylorlorenz
For example, nobody can search in mastodon. And tags can be muted.
It is open, but no uncontrolled.
Compuguy 🐱@TJ @taylorlorenz Yes they can do that. Though they need to be comfortable with Linux, Docker, and possibly working with a cloud hosting service like Digital Ocean, Linode, Vultr, AWS, Azure, etc....
@TJ @taylorlorenz Looks like someone has created a nearly one click deploy for Mastodon: https://mstdnandchill.fun/@sparkle/109543067894079228
🌟𝗈𝗈𝗁 𝗌𝗁𝗂𝗇𝗒🌟 (@[email protected])
@[email protected] there's also fediverse.express that will auto install it along with a few other platforms. I tried it out on a few different hosts and it works perfectly.
@compuguy @vodamark @taylorlorenz Still seems like a lot of work for your average troll. And, deployed at scale, I'd imagine it'd be spotted relatively quickly.
@TJ @vodamark @taylorlorenz That’s true. I’m pretty sure I’ve black-holed at least one personal instance…
Mike MasinI have a new, shiny instance. Fourth Estate, aka @jeff hosts it for a very reasonable monthly fee. I'm on the smallest plan, 1-5 users. So I didn't have to tinker with the internals. Moving my account was easy, too. DM if you want deets.
Josh Soref@vodamark Eventually if this attack becomes a problem, I expect instances will stop automatically accepting content from instances they've never seen before. It'll make it harder for nascent instances to join the fediverse, but it's likely it'll happen.
One cooldown approach to new instances could be functionally a "shadow-ban" for a cooldown window w/ only some users (who probably opt into a firehose) seeing content for the window. Giving time to 👍👎 before the instance is shown to normal users