Thoughts on Mastodon

Over the holidays I finally took my time to dig into Masodon. It has been a bit of a journey, both to understand how it works, and how to use it. From the start it feels like twitter, but as you use it you see the differences and realise how different they are. Regardless I do now feel at home on Mastodon, and the lack of an algorithm feeding posts it thinks I will like is most refreshing.

What is Mastodon?

The main selling point of Mastodon is that it is a decentralised social network. Think of it as combination of Twitter and Email. Just as Twitter, Mastodon gives you a timeline of posts by people you follow. However, it is not a centralised system, instead you chose your Mastodon instance. That way your username looks as an email address. Mine for example is @[email protected], and my posts can be seen by any other instance.

This is called a federated solution. You work towards your instance, that shares its data with other instances. In practice this minimises the dependency on a single centralised entity. If you wish to change instance due to a lack of trust, it not being maintained or closed down you can just export your data and import it on a new instance. That could be on a personal instance or one of the larger ones that is already existing.

For now it seems to be mostly tech people whom have migrated from Twitter, but there are others as well. However, since each instance usually have a focus, this might be somewhat due to the my choice of instance to join, and the people I follow.

This design of Mastodon puts a big responsibility on its users to follow accounts with diverse opinions to not end up in a bubble. There is a risk that if you only follow people with opinions like your own, they will be further deepened, regardless if they are right or wrong. It is up to the user to ensure that views are challenged in a good way and that everyone uses their critical thinking to not be fooled by malicious actors.

Main Differences from Twitter

The main things that confused me was the likeness of twitter, while Mastodon is different. Some things that seem similar works quite different.

There is no algorithm. You see the posts authored (or boosted) by the pepole you follow. If you want to find new people to follow you can also see the timeline of your instance and instances configured by its admin. However this timeline is sorted chronologically, and all posts of those instances are shown.
The difference between Boosts, Favourites, and Bookmarks.
- Boosts are retweets. On Mastodon they are used to further spread a post. This allows your followers (regardless of instance) to see the original post. Note also that it is not possible to quote boost. If you want to add something to the discussion, you need to do so as a reply.
- Favourites shares some features with likes. They can be seen on an original post, but does not push the post to your followers. Also note that there are no algorithm pushing posts with many likes. Therefore the effect of favouriteing is limited.
- Bookmarks are a way to save a post so that you can find it again later. These are seen by neither the author of the post nor your followers.
The search does not search the full text. Instead only displaynames, tags, and usernames are searched. In addition, searches can only find things the instance is aware of. For example that could be posts that are federated to the users of the instance or the authors of those posts. To find a user of another instance you need their full username, including the address to the instance.
Each post is tagged with a language. This makes it possible to use 1 account to communicate in different languages. Your followers can easily filter your posts to the languages they speak, and does not see the posts in other languages.