Ultimately of course you're supposed to choose a server that you like and trust. At least here you have that choice. On Twitter or Facebook you don't.

Of course it should have had end to end encryption. It sounds like a massive omission. I found a discussion about adding that to ActivityPub[0] where someone points out that if you don't want server admins able to read messages, you can't store private keys on the server, which sounds to me like it would hurt usability. Makes you wonder how unbreakable the end-to-end encryption of other systems really is. I'm not enough of an encryption guru to say how big of a problem this really is.

[0] https://github.com/w3c/activitypub/issues/225

Also, is the functionality dystopian or the application thereof?

Eventually, you'll reach the point where a duopoly will effectively control who is allowed to be part of the larger network. And this was the original problem people were trying to solve by creating a federated network.

What fresh nonsense is this? There's never been a federated system with anything like a perfect guarantee of connectivity to literally anything. There are always limits, both social and technical.

Where do people get these ideas?

It's the case that a federation tends to erode. Likewise, it's the case that a centralized system tends to collapse.

The question to ask is really one of backing a horse to ride on to be pragmatically useful; winning the technical argument is academic.

The fediverse has a few core principles in mind, and they tend to get centered in Mastodon's marketing. So it does make some sense that it gets attacked for being an imperfect realization of those principles, and for being unprincipled in other respects. But in a head-to-head comparison with other implementations of social media it can still claim to be more principled on those things it tried to pursue.

For this to work it needs to be treated as a common carrier, and not a social network.

Dystopian would be dictating that everyone operates the software they choose to run using their own time and resources in the same exact way.

You can stream illegal content from the firehose on Twitter too btw.

https://www.firstpost.com/tech/news-analysis/some-twitter-us...

In general just following topics or people that interest you is a safer bet on any platform with user generated media.

Unfortunately I seem to have chosen an abandoned or dormant account for my test, so I can't say for certain whether posts will show up in Home, but I would expect so given the other visibility changes. Hopefully someone else with some fedi experience can chime in.

Now, if that increases or decreases the problem space is a philosophical question. I personally do have issues with blocking instances as a whole because one admin disagrees with another admin on unrelated issues.

If I get denied access to other people to organise around, e.g. building our own zen retreat because my admin and their admin disagree on political factions or because one admin nuked the connection to another instance for not being trigger-happy enough against their own users, everyone loses. Such technical possibilities and - in fact - recommendations run against the ideas of a civil society.

I would prefer to use an instance that has no control over who I can follow. As an added benefit, this instance wouldn’t have to fully federate with every instance I follow users from, so wouldn’t add all updates to the federated feed.

This way the instance would only be responsible for user identities but would not have much control over who its users can follow. I still don’t fully understand why Mastodon did not choose this design.

edit: it was called sinblr and rumors are that it was an instance called pawoo that flooded it with seriously illegal content.

Centralized social media platforms also control everything you see, to the point that it can be very difficult to see the full range of posts from accounts in good standing that you follow within their platform, in favor of posts they think would keep you more "engaged". And they are famously opaque on how they do that. If you can guess at what they've done and decide you don't like it, you don't have any recourse. After you've spent time to build a network of people you follow and follow you, they have you locked into their platform, holding your subscriptions hostage against you.

I guess I don't understand what you want. Twitter is too much censorship, right? But we don't know how Twitter censors content, and even if we did, what would you do about it? Mastodon instances censor content, yeah. Why not? It's a person running their own site, they can let whatever they want through it. But they tell you how, when, and why they censor. And if you don't like it, you don't have to put up with it.

Wow. As much as I disliked Twitter deciding who I can and can’t follow (via bans), and what people can and can’t say, at least they put some rigor into it. I don’t think I’d ever want some self anointed little server god king banning so many peers for such soft reasons. I thought it would be csam, armed groups - this has reasons like “Trump fanbase”, “offensive content”, and over and over again “a-holes”.

No thanks.

It was a more civilized time.

Today, server admins ban other servers for random issues. Users do not use the built-in killfile-esque blocking method, but instead go and complain to server admins (who then are incentivized to use massive action against other servers). This obviously is not an environment that will lead to societal harmony.

The end result however is what you'd expect and the experience on those instances are very different from what you see on the more mainstream ones, and most people wouldn't like it.

Some of the "free speech", "low-moderation" instances have self-policing communities. Others are full of neo-nazis. Yet others are run by bigots and ban anyone who objects to the hate and vitriol. They are not all the same.

If you feel harassed by an user, filter that user. If you feel harassed by every single account on an instance, filter that instance for yourself (this option currently does not seem to exist in Mastodon). But do not go crying to our admin and deny me, who happens to be on the same server as you and is semi-ok with some of the folks on that other server the ability to interact with them.

This is creating filter bubbles, your users will at first love the friendlyness and fluff, but it also is exactly what led to the right-wing successes we've seen all over the world in the last few years. Filter bubbles destroy democratic societies.

> Yet others are run by bigots and ban anyone who objects to the hate and vitriol.

You see the dangers of giving admins that level of control? Now you got an instance in which there are no more discussions, only self-assuring and groupthink, and which will slowly become more radicalised. If filtering was strictly an user's issue, other users may have seen the objection to hate and vitriol (which often starts underhanded), and have had another view.

By the way, that is true no matter which political fringe a server admin belongs to: I know a Mastodon instance on which some users become increasingly radicalised against car owners right now, to the point where mass executions are being normalised by joking about them. Guess what happens when you try to become a voice for reason... But are ALL of them just evil, unredeemable beings that need to be banned from talking? Should my Mastodon admin block that instance for spreading hate and vitriol, even though the majority still talks about other things? Obviously, that's a silly idea.

Here's another way to look at it: Our instance has moderation policies. You chose that instance in part because of those moderation policies, just as I did. Those moderation policies are there so we can interact with each other with the peace of mind that we're not going to have, say, child sexual abuse imagery DM'd to us.

Then an instance comes along with a few hundred follow bots that do that. So, after a few reports, our instance admin defederates from the instance.

If you don't like this, you're perfectly welcome to go to another instance, with different federation policies. Thanks to federation, you can do this and still stay in contact with me! It's a win-win!

> You see the dangers of giving admins that level of control?

No, not really. Those people are probably always going to exist. Honestly, much as I hate Nazis, if they're over there doing their thing, and not trying to harass or murder me and my friends, I'm happy just to let them do that. It's just a website; it's not like their kids won't have external influences and be able to figure out that, hey, Nazism is actually bad.

Not saying it's good or bad, just that it isn't a dichotomy between Mastodon restricting things or centralized systems that don't.

For example, on my silly hobby sites I provide all bots a status 200 for all GET and POST requests and just send them to a dummy virtual host to let them play in the sandbox for years on end. If they even once looked at the output they would see my silly ASCII art.

Some may be curious about search engines, but since I only allow HTTP/2.0, only bing can even connect with my site and robots.txt tells them to go away.

In NGinx:

    error_page 404 500 501 502 503 504 =200 /ram/e.txt;