Curating my Mastodon feed

Introduction

I like using Mastodon (/ the “ActivityPub universe”) for connecting with people, and I have found many great people I enjoy talking with. I look forward to hearing from people on good and bad days and having somewhere to share my world with.

One challenge is seeing the post I want while seeing a minimum of triggering posts. I think it’s important to be explicit that many things I find triggering and don’t want to see (at least most of the time) aren’t necessarily bad posts - they are, for various reasons, things I don’t want to see.

One good example of things which aren’t inherently bad, but I don’t want to see as much as possible, is politics for countries I don’t live in. Living in Ireland, I live between two very “noisy” countries (the UK and the USA), which speak the same language as mine, so it’s helpful to try and filter most of that to avoid it being too much.

Disclaimers

What I describe here is just what I currently use. I’m not claiming it’s the correct way of using Mastodon.

If you think this post is rather long with a lot of background, you are not long that’s to try to avoid misunderstandings - and since I wanted to get this published, and not just keep editing it.

What I do to curate my feed

tl;dr / summary

• No subscriptions to hashtags or groups.
• Heavy filtering.
• Selective following.
• Blocking/muting people who trigger me.

Overall Strategy

• People do not have a high to my attention/time.

• Just because a topic is important doesn’t mean I should be bombarded with it.

• Some people live in a very different world and have other priorities. Theirs are not wrong, but that doesn’t mean they should be my priorities.

Hashtags / Groups

I don’t subscribe to any hashtags/groups (a.gup.pe, etc.) to show up in my feed. Instead, I may look explicitly at them when I have extra mental bandwidth.

There are two core reasons I don’t subscribe to hashtags/groups:

First, the amount of posts is just too high, which means it generally gets overwhelming at times (especially since I worry about missing posts from people I want to see) regardless of the content of the posts. I know I could use lists, but my brain gets slightly annoyed at seeing the same posts multiple times, so they don’t work great for me.

Second, as everybody can post, I have much less control over what I see. That model is great for many things, such as inclusivity, but it has downsides and significantly increases the risk that triggering topics will slip past my filters.

I would like to be able to follow the groups more regularly to hear more perspectives (which I do enjoy, but that’s a topic for another day), but for my main “goal” of connecting with people, this is a tradeoff I have made.

Filtering

I filter a lot. I do it from things I find very triggering and I absolutely don’t want to see, over things I find annoying, to things I just don’t care about.

The Mastodon filtering system is the best I have experienced on social media yet, and I love using it as it improves the experience. Especially the fact that you can filter things, but they still appear behind a filter name I chose is great for my uses.

Filtering runs the very real risk of false positives, which remove posts you would like to see. Still, the fact that I can expand the posts means I can see some posts when I have the headspace to possibly be triggered, which means “expanding knowledge” and spot-checking if my filter was too broad.

I do expand posts frequently, so I see them even when they match a filter. Because it’s opt-in and I know which filter/topic is blocked, it helps my triggering to be less severe. Most of the time, I will expand some of my filters, but they are still very useful for days when I’m overloaded and don’t have the spare energy to spend. As a side note, I automatically expand Content Warning (CW) posts, as most are not triggering, and the user interface isn’t great for them. I still like content warnings, as they give “early warning” for topics, explicitly show the poster cares about readers, and increase the chance my filters will match keywords.

My use of filters is reasonably simple. If I see posts on topics I don’t want to see enough that it’s worth the trouble to get rid of, I add it to a filter. In some cases, the posts are rare enough that it’s not worth the time of setting up the filter and the risk of false positives. I also rarely proactively filter, so my keywords may sometimes seem random, but they are based on what I run into.

One interesting thing with the Mastodon filtering system is also that you can select where you want filters applied. All my filters are currently set to Home and lists and Public timelines only - specifically not Notifications. That means the filtering won’t suddenly mean I don’t see a reply for conversations I’m participating in. These options may need to change for some filters, but they haven’t yet.

I have included some example filters below. I have hidden them by default since they could be triggering. The rest of the section gives a pretty good idea of what I do, so reading them isn’t essential, but I’m including them for completeness.

Selective Following

I’m fairly selective in following people and don’t automatically follow back. I want to be very explicit that not following back isn’t a moral judgment of the person. Many good people have interesting things to say, but I do not follow the reasons outlined below.

• If I follow a person, my bar for unfollowing is high, so I would like to avoid doing that.

• I want to find more people to connect with, but I also need the bandwidth to interact with the people I follow, so that’s a balance.

• I would like to find more people to connect with, but I also need to have bandwidth to interact with the people I follow so that’s a balance.

• Some people post a lot. I don’t have a problem with it in itself, but it may be more than I can handle.

• Some people post a lot on topics I find triggering. Again, I don’t have a problem with them doing that, but it’s also not feasible for me to be triggered frequently.

All this also means before following people, I normally spend some time browsing their profile, which requires energy I only sometimes have.

Most of the time, when I follow people, it’s because we have interacted in one way or another, and I would like to keep doing that.

Hiding Boosts

Some of the people I follow do boost either too much for me to handle or not topics I find triggering or are generally just not interesting.

Disabling boosts from that person works well in fully seeing their posts but fewer other posts.

Blocking/muting people who trigger me

I block or mute a fair amount of people. Most of those people are people I have never interacted with, but they came into my feed for some reason. By count, most were not actively abusive but just annoying/triggering enough that I don’t want to see their posts. One common case is not using keywords/content warnings that would have allowed me to filter the posts. For most cases, I will write a note in the profile, frequently just a link to the post that caused them to get blocked/muted so I can see it later.

Some people argue that blocking/muting should be a last resort. I don’t see it this way. While I try to do it as little as possible, too many triggering things mean I can’t use Mastodon, so keeping the level down is important. Also, nobody has a right to my attention, just as I don’t have a right to other people’s attention.

The other side of this is that I also generally unblock fairly freely. If I see good posts from a previously blocked/muted person, I’m very likely to unblock them unless I can remember/see they did something really “bad”. On Twitter, my system was mostly mute for mild “don’t want to see” and blocked for severe. That was a hint for care to use when “downgrading”. For various reasons, Mastodon makes a system like this harder, but it still fits my core philosophy in the area.

Mastodon has timed muted, but in the normal user-interface, the maximum is 7 days, which is useful in some cases. It would be nice to set, e.g. 6-12 months, so you could mute people for a long time but still automatically give them a second chance. I may implement some code to do that at some point - but I haven’t yet.

For blocking, it’s also important to note that it’s rare that I feel the need to block/mute people I actively interact with. By rare, I meant that I remember one case of blocking a mutual¹, and I can’t right now remember having to block people I interacted with. Most of the time, for communication, going badly, just disengaging and stopping replying works fine.