Just an occasional reminder that disabling replies is the #1 requested feature from Mastodon.

@stefan oh I thought it was already implemented since ActivityPub already has it. Yes, about limiting who can reply.

@7adi Ah, got it. Yeah, I am actually not sure where the hold up is now.

I understood that the Mastodon team has been waiting for GoToSocial to write an FEP, but I did see someone else mention recently that this is already possible to do.

@stefan@stefanbohacek.online see, my question is... you can limit replies and notifications, but the people on other servers can still reply, they're publishing it out onto the fediverse, it just doesn't reach you. Is that ok from a Trust & Safety perspective?

@julian @stefan
The way that Quote Posts work is a good template.
If reply limiting was done I. the same manner, the reply wouldn't reach you nor would it reach other people.

@julian I don't typically deal with abuse online, so maybe not the best person to answer this, but personally, I'd be fine if the reply was hidden from me and only lived on the abuser's server.

I'm guessing third-party servers that implement reply controls also wouldn't see the reply? I guess that should be sufficient.

Obviously you can't prevent people from posting stuff on their own website/blog/etc, but there have to be ways to limit the reach.

@julian@activitypub.space @stefan You can't meaningfully stop other people on the internet from trying to send you messages, but the value is in giving you the option to (premptively or post-hoc) decide which ones you're willing to read and to display to onlookers visiting your post.

Private filtering is nice on its own. Reply controls based on GTS-style interaction policies gain additional value through their adoption in the ecosystem at large.

@stefan occasional reminder that we are very well aware of this and if it’s not implemented it’s because it is incredibly complex and impossible to do right on all aspects due to the federation, and reaching a compromise on how it should work exactly is not easy.
But that’s near the top of the things we want to do and will happen at some point.

@stefan @7adi it's implemented in #GoToSocial. you can set default post's interaction policies in settings, and then even approve interaction requests...

https://docs.gotosocial.org/en/latest/user_guide/settings/#default-interaction-policies

@renchap Completely understand. It's really just frustrating to see the fediverse not being able to live up to its full potential when people leave for platforms where they feel safer.

Well, I trust that you and the rest of the Mastodon team are doing your best. Thank you for that!

@stefan and both are also impossible to implement in a federated network, and I wish everyone finally understood this so we could put this fruitless discussion to rest and move on to things that could actually be improved (like the dismal state of the moderation tools)

@Edent @julian @stefan quote posts don't work, simple as that. Most other AP software implemented them long ago and those softwares don't give a shit about Mastodon's special have-our-cake-and-eat-it-too solution. I turned quotes off, hasn't stopped one Misskey or Pleroma user from quoting me or seeing unauthorised quotes.

All of those limit/approve features, yes that includes blocks, ultimately rely on the good faith of the rest of the network. Whether it's quote approvals, blocks, or any hypothetical reply control, it would only ever amount to muting by a different name.

@stefan agree 100%.

@amberage I think these features mean slightly different things to different people, and my impression is that it is possible to provide at least some control to marginalized people who are most often victims of targeted harassment.

If this truly was impossible, I don't think there's much of a point in sticking around the fediverse if we can't ensure everyone feels safe and welcome.

And I'm sure moderation tools can be improved, but these can only be used after the damage has already been done. There have to be better ways for people to defend themselves before an attack, or before moderators can step in.

@stefan

... suppressing discussion has never been a good approach...

@stefan I have quite a bunch of ideas for moderation that could prevent harassment in the first place, tbh, but chances of Masto devs ever implementing anything like it are about minus 9000%

What can be implemented re: reply controls is, basically, selective muting. A post could indicate "only people XY may reply" (i.e.: followers), fellow vanilla Mastodon servers would respect that, other ActivityPub software may or may not respect that, and bad actors certainly wouldn't. So while it may hide unwanted replies from cooperating parties, it would only ever do so on a good faith basis.

Twitter could do reply controls because Twitter is one company. All user accounts, all posts, all are owned by Twitter. It rules absolutely, for better or worse. That isn't possible with ActivityPub, where each post, each like, each follow, is just servers sending "hey, I did this thing" announcements into the ether and other servers deciding how to respond.

@manankanchu Would you consider a blog that has comments disabled a "suppression of discussion"?

Bottom line: https://stefanbohacek.online/@stefan/115940412454524948

> "But what if I have a strong urge to reply to a stranger?"
> Find a more productive way to spend your time.

@amberage What you described is pretty much how I'd imagine this to work. Obviously you can't prevent people from publishing whatever they want on their website, blog, or social media, but there have to be ways to limit their reach.

Also, have you seen Mastodon's updated roadmap?

> Moderation tools
> Looking at ways to make moderation easier, e.g. shared block lists.

https://joinmastodon.org/roadmap

That sounds promising, I think!

@manankanchu assuming perfectly spherical commenters where there's no power disparity, nobody can shout down or intimidate anyone else, etc, etc. Free Speech only works under ideal conditions, in the real world it does not. Your own country quite rightly has laws suppressing hate speech, for instance.

And besides, abuse isn't discussion.
And besides, nobody owes anyone the right to reply.
And besides, the real point of free speech is *the right to say what you like about the government*, not the right to say what you like to some person and they are obliged to hear you out and platform you!

@amberage

That's the basic misunderstanding that people have about decentralised networks:

They don't get it that once a message leaves your instance, you lost all control about it.

All this "Don't quote, don't reply, quiet public, followers only, opting out of indexing and search machines etc." is merely a recommendation, but cannot be enforced.

I always say: Only post what would do no harm to you if plastered on a public bathroom's wall or taken to the police.

@Edent @julian @stefan

@mina That is a solid advice, sure. But even completely innocent posts can attract mansplaining, tone policing, and outright racism and sexism, and worse.

I don't typically deal with this stuff myself, but on at least two occasions, after posting some pro-trans articles and messages, I'd have bunch of anti-trans losers show up in my replies.

Easy to block, but I just can't imagine dealing with that daily. Or on a bigger scale. I would probably leave for Bluesky myself.

@amberage @Edent @julian