Hacker News Reader

Comment by young_unixer | Hacker News Reader

I left Mastodon because the big "traditional"/sanctioned instances started banning other instances they didn't like, specifically Gab (instance full of right-wing people that allows basically anything that isn't illegal).

I had high hopes for Mastodon but whatever, this whole social network thing isn't worth the trouble. Now HN and Youtube are the only websites I visit for entertainment.

View on HN

I don't understand this mindset at all. It's not so much "I want to be able to say whatever I want", it's "I want everybody to listen to whatever I say". Go post edgy xenophobic stuff in your dedicated online community if you like it so much but stop acting surprised when other communities want nothing to do with you.

I believe that the issue discussed is more about mods having that control instead of the user itself. Not about the fact that everybody should listen to everything.

I understand that the question is: should the mod define who “the community” the community can connect with, or not?

Exactly. Maybe some heuristics or even deep learning could help retaining control. But then we would succumb once again to the Eldritch Algorithm gods...

Very few people actually want to filter through the unadulterated garbage of the internet on their own. What you want to do on social media is like, read the news, talk to friends, or promote your work. You probably don't want to spend every minute of your online presence blocking new spammers, new trolls, death threats, conspiracy theories, and the like, just to make your feed somewhat usable. So you delegate this task to moderators. On a centralized platform like Twitter or Facebook, you don't get to choose moderators. Facebook decides it doesn't belong on its platform? It's gone. You have no recourse. With Mastodon, you get to a) pick a server with policies that you like in the first place b) you can switch to another server if you're unhappy with the original one.

Server blocks are just an extension of that function. If a spammer creates a new server under their own control and creates a million accounts to send spam from, do you expect moderators from other servers to just click "block" a million times? No, that's why the bulk option exists. There's no way it couldn't.

More or less “I’m not going to pay for your freedom with my privileges”.

De-federation and Gab blocking are mostly for instance operators and app developers to avoid hosting illegal content and avoid being seen as sympathetic to extremists respectively.

What people fear is either could be used as grounds to charge them as child molester or colluding terrorists or to revoke your Apple/Play Store developer accounts for life.

The mindset is "I want to choose who I listen to. I don't want the Mastodon instance to decide for me"

If you don't like listening to some people, that's fine, don't follow them, block them, whatever, but I won't join an instance where I can't listen to other people because the moderator doesn't like their message.

I had a quick look at mastodon a while ago and saw a few more problems with the model, but I wasn't aware instances could block each other. Maybe some day there will be an even more decentralized alternative (Looking at you, indieweb) where it is up to the user whom they block. I'm also not a huge fan of how mastodon is an implementation, more than a protocol.

So long as instances are serving content from other instances (pretty much required for a social network), there will be demand to be able to block other instances. Of course, people could just host their own instance, which is possible with Mastodon, and is indeed what a lot of people already do.

Mastodon uses the ActivityPub protocol, and there are other implementations that use this protocol (Pleroma, for example).

Have you looked into ActivityStreams? It might be the protocol you're looking for!

https://www.w3.org/TR/activitystreams-core/

ActivityStreams is a data format, not a protocol. ActivityPub is the protocol that Mastodon, Pleroma, etc use which shares ActivityStreams data.

I, as a user, am still going to demand that I can join a community where of likeminded people and moderation so I don't have to deal with the overwhelming spam of the internet myself. That's the bit that a number of decentralisation projects don't get - I don't want to have messages by anyone on the whole internet forced into my view (because have you seen how much that sucks?). I want moderation if the day-to-day use of the tool involves coming across a largely random group of people - otherwise it's just a platform for harassment and spamming people with child porn, a la Matrix - and I don't want to perform that moderation 100% by myself.

Then what is the point of federation? You just described a centralized forum/microblogging server.

The point is flexibility and no lock-in. If you buy into the Twitter platform and eventually you're not happy with the way it goes, moving away has a huge cost. If a federated architecture peers can come and go relatively easily.

See email for instance: if you're unhappy with your current provider you can move to a different one or even roll out a new server and you can still interact with the other users.

Simply interacting, yes, but email and messaging cannot replace Twitter. What makes Mastodon viable is the identity, multicast and backlog of toots(tweets?), which does not transfer as easily between instances.

I can pick and choose who my community and moderation team is - that's why I am on the fediverse and not Twitter. I am strongly opposed to having extreme right-wing views (along the lines of "you should be dead for who you are") forced into my conversations? I can pick an instance/moderation team which proactively blocks other instances which refuse to moderate their users according to such basic social norms.

Fact of the matter is that I don't really want to talk to most people on the internet, and I don't want to see what they have to say about me every time I want to see what my friends are up to. I want to talk to my friends, maybe have our wider communities able to chime in, and occasionally discover new people through that. It's not my job to convince random assholes on the internet that I deserve to exist, and it's not useful in any way to see their messages. Blocking extremist free speech instances which promote harassment as a normal part of their operation is... a feature, not a bug.

That is a fair point, but what happens when the moderation team deviates from your beliefs? What happens if your instance is declared not-safe? You're effectively exiled from the fediverse, people you follow will never see your toots again.

You can move to another instance.

Then I move instance (probably well before my instance is declared not safe, tbh). It's a feature in Mastodon, assuming my moderation team hasn't decided to disable it - basically, I send a protocol message to my followers saying "I'm over here" and they automatically follow me over there. In a future p2p protocol that's designed by people who actually realise that people exist who don't want everyone on the internet to have a direct line to their inbox (aka none of the current ones), I could simply move moderation team and keep my identity.

It's incredibly unlikely that tomorrow, my instance pushes the needle so far that everyone blocks it immediately. More likely a series of changes in the moderation team gradually pushes things that way and I can change instance before things get bad enough that anyone would block it - and I'd do that because it wouldn't be a community I want to be part of any more, rather than any particular fear about being blocked.

I had no idea account migration landed. Your description actually sounds quite reasonable. I guess I was too shocked and burnt by the instance blocking incident. Maybe I should give Mastodon another try.

Just need to find an instance that doesn't block...

Finding an instance that doesn't block other instances, but also actually moderates its users and thus doesn't get blocked, is going to be pretty hard - and also a rather harassment-filled experience unless you fit in with the Gab crowd, I imagine. You could always run your own instance.

Note that the majority of instances that are "blocked" are actually soft-blocked by most instances, meaning you can still talk to people on them if you follow them, you're just not going to find posts from their users otherwise.

"where it is up to the user whom they block" does not imply that you have to manually select who to block, rather it could as well mean that you are free to select block-lists (with the ability to fork them if you so wish) of your own choice rather than have the admin of your network force them on you.

This changes nothing, one blocklist will get a lot of users, end up as a "recommended" setting, then all those hurt that their bigoted views aren't more popular will moaning about "free speech" again.

I'm happy let those who want to run a community that has similar values to mine block obnoxious content for me. Far better that than Facebook or Google being the sole arbiters.

> This changes nothing

The change is that it empowers the users.

> one blocklist will get a lot of users, end up as a "recommended" setting

This is fine as long as the users select said list out of their own will.

> then all those hurt that their bigoted views aren't more popular will moaning about "free speech" again.

Have you considered the possibility that a lot of non-bigoted views are currently blocked due to trigger-happy admins?

> Have you considered the possibility that a lot of non-bigoted views are currently blocked due to trigger-happy admins?

Exactly who is on Gab who I want to listen to? If they're not a bigot and not ok with bigotry, why are they on there?

I am not talking about Gab. I am talking about the other instances that were silenced in their crusade.

That was a feature added later, mostly to address instance operators' fear for police intervention for enabling access to Japanese content(they say loli but even Japanese in 30s to 40s are legit mistaken even in court as underage, so not much to do with age, IMO more of specific makeup style or lack thereof)

There's also pawoo.net, which was operated by Pixiv (switched hands).

Pawoo is the go-to instance for (lolicon) artists banned from Twitter, and despite being the largest Mastodon instance is blocked by almost all instances.

The fear came from some European laws forbidding under-aged illustrations (typing this makes me die inside), so instances serving pictures from Pawoo may get into trouble.

https://mastodon.social/@Gargron/2657448

I kind of don't like how "lolicon" is being used -- because pedophilistic orientation don't seem to correlate all that well to how likely people is banned from Twitter. There's a real pedo guy on it, who I've seen public about how arts don't substitute stuffs, who posts his... arts a bit jagged, and doing it for years, then on the other hands there are artists getting banned nth times this year posting an abstracted person with clean and continuous lines.

My theory is there's a set of curve parameters that evaluate to "legality" in a sigmoid response that has less to do with ages or even if it's depicting a human or an animal or not, like a picture of desert hills looks pornographic sometimes or how sumo wrestlers charging evoke no sexual emotion. That human curve scoring yet to be discovered is, to me, looks like how legality is determined worldwide on and off internet, so calling those drawings as loli or people doing those curves as lolicon is inaccurate in my opinion.

Actually It's better than traditional centralized system like Twitter. If it's centralized , you can't do anything after banned. On decentralized SNS, you can join the problematic instance or start a your own instance that not banning.

Reply on news.ycombinator.com