Mastodawn

Erin Kissane Jul 28, 2023

This week, I went over to Bluesky and asked people who'd left Mastodon why they left, and lots of people told me. I grabbed the replies and crunched them and wrote up a summary. I think it's really interesting and often kind of wrenching.

https://erinkissane.com/mastodon-is-easy-and-fun-except-when-it-isnt

#meta

Show thread

Erin Kissane Jul 28, 2023

Rather than trying to head off the unusual unpleasantness about clout-chasers and the ritually/technologically impure, I will just say this:

I wrote this up for fedi people who are actively curious and interested in other people, and I'm not going to worry too much about how it lands for those who aren't.

Show thread

Erin Kissane Jul 28, 2023

The tl;dr (because TL! it's TL) is that, for this group:

- people feel stressed and anxious when they get yelled at for breaking rules and norms they didn't know about

- it's hard to find people and conversations, and specifically hard to follow people across instances

- people want better organic and algorithmic ways to connect with each other

- instance-picking stresses people out, and a lot of the sign-up and settling-in processes are confusing and/or too much work for unknown returns

Show thread

novatorine 🏴🏳️‍⚧️Jul 29, 2023

@kissane I think the best way to make instance-picking more accessible for people is to make it so that instances have a description built right into their representation as a server or whatever, and make it so that those descriptions can have hashtags in them. That way, on joinmastodon.org or whatever, someone could just put in all the hashtags they're interested in and get a list of the instances that share the most amount of those hashtags. Maybe a standardized way to format and present moderation rules and instance values would be great too. Not necessarily something that's like built with software but just kind of a template that everyone can use if they want to to make it easier to scan.

Show thread

Erin Kissane Jul 29, 2023

@anarchopunk_girl Totally agree, I think there needs to be a lot more information and a lot of it needs to be better structured. I also suggested elsewhere that fedi is eventually going to need a reputation system, but culture here is SUPER not ready for that yet, with all the drama.

Show thread

aces

@kissane @anarchopunk_girl

A reputation system gives me alarm bells - China comes to mind. Who is going to set the reputation points, and set the rules? Instance/server-wise, yes it makes sense for each admin to set their own rules - and may the best server win by people voting with their presence on that server. But Fediverse-wise? No society is ever ready, because that's a still a form of domination and control. A co-op is still the best way - and again, servers that like each other will federate with each other - but a standardized reputation system /never/ goes as planned - people will always find a way to become the Judge, Jury and Executioner when deployed on a massive scale.

Instead of systemized thinking, we can look to nature how ecosystems balance themselves organically without invading other ecosystems. Left to her devices, nature will eventually cull out or adapt to an invasive species - if humans don't meddle. Each local environment will regulate if there's good guidelines up front, like you had mentioned in the article. Take for example, where one culture has a certain hand gesture for 'ok', but another culture the same hand gesture is offensive. If we translate that to a global scale, who will say which culture is correct? It comes back to localities - just like natural ecosystems- and that's the fediverse strengths. A reputation system will just further domination and control if someone manages to code it in and normalize it - where again, the minority becomes outliers.

Sorry for the long response - I'm just writing out as these things come to mind, not at all a planned response.

All of that said - a search engine that just shows the most amount of hashtags used isn't a bad idea - rather than crawling the data, it just crawls the hashtags and saves those, instead of the full content.