Mastodon Mastodon a glitched, edited version of a photo of notebooks, laptops, tea cups, and other accoutrements on a table. Photo by Marvin Meyer, via Unsplash.

Technically Good ✨

🕸️ Caught in a Multi-Billion Dollar Web: Why Leaving the Legacy Algorithmic Media is So Hard

A photo of a person's dark silhouette against a tall fire in the night. Image by Mohamed Nohassi on Unsplash.

Recap: Feeding the Fire 🔥

So I just wrote a blog post that I had assumed all of eight people were going to read. Turns out, that wasn't the case at all - it must have hit a nerve, hitting #1 on the BearBlog trending pages1, doing the numbers on Mastodon2, as well as encouraging a few folks to send emails (!), which is by far the coolest way of getting in touch. To think that someone sat down and took the time to grab the email address from the contact page and write a personal note - that's rad.

Now, I'm not saying this all to brag, but to illustrate a point, because: That same post on LinkedIn? Not as much of a success. Crickets, in fact. It got like six likes, zero comments, and nobody reached out. Isn't that odd?

I'll leave it as an exercise for the reader to figure out why a post criticizing and prying apart the workings of algorithmic media may not have been doing well on LinkedIn. 🙂

If you haven't read the article, it essentially outlined the link between human psychology, engagement, and algorithmic media. The tl;dr:

The existing social media platforms - or legacy algorithmic media, as I like to call them - typically have the following hallmarks:

And yet: Even people who know all that, people who don't feel good about it, people who agree that the above is not right and that we should move to alternative platforms, often feel that they can’t leave. Why is that?

Define "Working." Working for Who?

Part of why people are sticking around is that the platforms are still "working." I explicitly want to avoid the phrase "working as intended" here, as I think we need to differentiate between three different definitions of "working" in this context:

  1. "Working" in the sense of "functional in a technical sense." The service has fast servers, few errors, a familiar and maybe even clear-ish UI, and little downtime. When I click a button, I have an expectation of what is going to happen, and most of the time, that expectation ends up being fulfilled. I don't often encounter 520's trying to pull up my home feed, or 403's for something that just vanished one day to the next.
  2. "Working" in the sense of "working for the company." The platform is popular enough that it can retain enough existing users, attract new ones (infinite growth, my friends!), generate income, maybe even be profitable and/or dominate its corner of the market. Or - if not that - maybe it successfully funnels users to another platform owned by the same company that is profitable.
  3. Working for you, its user. The platform has not materially changed its behaviour from when you first signed up. The rug wasn't pulled from under you. The platform did not magically start hiding long-standing, expected features behind paywalls. It is showing you the content you came to see - updates from people and on topics you care about.

The legacy algorithmic media platforms often still “work” in the first two senses: In a technical sense, as well as for their operators (as otherwise they would already have been shut down). However, this can be true while the platform at the same time fails to serve users’ needs.

What's Different Now

Enshittification used to, and should still have, consequences. Yes, even in the digital realm.

When MySpace got bought out and became even more cluttered and added ads, people left and flocked to Facebook, which, at the time, was ad-free and looked sleek and clean. When Internet Explorer kind of just... stopped innovating and adding new helpful features, people switched to Firefox and Chrome.

So why aren't people leaving the algorithmic networks? If you go to a store to buy milk, and suddenly you're told you can only access the milk after the manager makes you stand there and watch as he shoves three big ad poster boards in your face for 30 seconds each, or that there is no milk but you're welcome to purchase this set of tennis balls instead, which is what people who look like you seem to really like these days, so why aren't you buying the tennis balls?, you'd probably go to a different store. And also tell all your friends to avoid that store, because good lord, what a disaster.

So why are we letting social platforms do "every third post is an ad", "we sell your data to our 1,879 partners," and "your feed is now full of promoted content you never signed up to see" to us? Surely technical functionality isn't everything that matters here?

A glitched photo of two locks on a chain link fence. The one on the right says "I regret this already" in all caps.

Image by Bruno BD on Unsplash.

Why People Stay

This is where we need to start talking about Network Effects, and related concepts like Switching Costs and the Collective Action Problem.

Network Effects

In short, network effects mean that the value of a product or network is intrinsically tied to how many other people use it. It's a term borrowed from economics. Think about the first telephone: It was effectively useless. You can entirely install the one lonely telephone in existence in your living room - at that point, it's an expensive piece of home decor. Who are you going to talk to? Network effects drive value, or, as Forbes puts it 3, network effects describe:

the benefit of participation and how something can become more appealing and usable as more people become involved. In other words, value increases as usage increases.

You can see non-digital examples of this with rideshare companies - if there's not enough drivers, calling your rideshare will take longer. You get annoyed, maybe call a taxi instead next time. The more people feel that way, the fewer fares a rideshare driver is able to pick up, the gig becomes less lucrative, fewer drivers offer their services, riders have to wait longer, and round and round we go.

It works similarly for social networks. A social platform’s value depends on who else is there. Family groups, community pages, local businesses, clubs, and niche hobbyists congregate on the same network. If your people aren’t there, that platform becomes far less useful to you.

In the end, this all comes back to human psychology - the "interpersonal, tribal impulse to connect with others." But, layered on top of individual psychology, network effects then make it so that, even if you find that a network or a platform isn't serving you well anymore, you don't just pack up and leave. “I can’t switch away from [insert enshittified platform here] because all of my friends are there, and they don't want to switch.” This successfully stifles much individual momentum, and makes collective migration extremely hard.

Unfortunately for us, it is not just the network effects that keep people locked in - Switching costs also play a role.

Switching Costs

Especially when it comes to platforms on which you are well-connected - where the network effects are highest, and "drive the most value" for you - there is a "cost" associated with leaving that network behind.

This switching cost, sometimes also referred to as a switching barrier, is not necessarily monetary (although it can be, if you are selling services or goods over the network, or access employment opportunities). This kind of "cost" is figurative, costing you something as a human being. For example, leaving the network could mean you are now being disadvantaged in one or more ways: By losing access to social contacts, events, interest groups, and simple coordination with friends and family.

The larger the network effect of a platform, the harder it is to leave it. This gets dialled up to eleven when the platform in question does not let you export your data (think posts, images, connections) and take them with you to a different platform. Which, surprise, the enshittified platforms aren't designed to do.

Depending on whether you are a user simply trying to go about your day, or someone trying to "hack" a "business model," you either find this very aggravating and petty, or you view it as "one powerful way" to keep customers "loyal".

The Collective Action Problem

But let's say you are one of those folks that just - takes the plunge. You cancel your facebook account, axe your twitter, the works. You sign up for the Fediverse. Maybe you even find some folks there that have interesting accounts, and you start following them.

The problem? Unless you get all of your buddies from the other networks - the networks you kicked to the curb - to switch over too, you will not be able to entirely let go of the other networks. This is where the collective action problem becomes pertinent: Even if you yourself escape the networks that no longer serve you, you still have to make your social contacts do the same thing if you don't want to be pulled back into it.

Exporting your network from these walled-garden-type platforms is basically impossible. So now you are faced with rebuilding your network elsewhere, which is an entirely different beast. This takes effort, and it requires convincing others to move too. Unless those folks are themselves interested in moving, you can't make them, as much as you'd like to. The full article seems to no longer be available, but it does indeed seem to be easier at times to move country than switch social media.

Basically, the network enshittifies, and then enshittifies some more, but it retains its "social value" - and with it, the network effects, and high switching costs. This makes users tolerate all sorts of objectively bad products: What's one more ad, what's just a liiiiittle more algorithmic meddling, what does it really matter if more horrific posts slip through due to shoddy moderation, and they've got a "premium" track now that removes ads and shows you slightly more of the content you actually want to see.... Sure, you've got to pay for that using your money and your data, but it's slightly better, it gives you some reprieve from all the ads and promoted content, and - all your friends are there....

Here's a great quote from Cory Doctorow:

Social media bosses know all this. They play a game where they try to enshittify things right up to the point where the costs they're imposing on you (with ads, boosted content, undermoderation, overmoderation, AI slop, etc) is just a little less than the switching costs you'd have to bear if you left.

If you're thinking: I know I should leave facebook/instagram/twitter, but I just can't - it's not just you. This isn’t about individual choices or weak resolve. It’s structural, and it's by design.

So... What do we do?

This all sounds very depressing. It sounds like there are all sorts of effects at work - from individual human psychology to stickiness engineered by multi-billion dollar companies whose entire raison d'etre is to keep your eyeballs on the platform for longer - that make it nigh on impossible for us to leave those enshittified networks behind.

But! Here's the good news. There are things we can do.

A glitched image of a blank page saying "TO DO". Image by Annie Spratt on Unsplash.

Sign up for a Non-Shitty Alternative

You can use network effects to your advantage. You (yes, you) can be the start of a flywheel effect in the other direction. Be the first among your friend group, your social circles, your family, who creates an account on a platform that does not lock you in.

Outside of a few minutes of time, it costs nothing to set up an account on Mastodon, Pixelfed, whichever flavour of the Fediverse you prefer. Do that! Set up an account! Play around with it. It's truly not rocket science, even though the way some legacy media write about it, it occasionally sounds like it is. You do not need to have an advanced comp sci degree or be a seasoned ham operator to be able to navigate the Fediverse. Trust me.

Follow a couple of accounts there that interest you. Punch in some hashtags you think you'll enjoy, and follow those too. Follow people who post under those hashtags. Follow people retooted/reposted by the people you follow. It will feel slower in the beginning, because there is no algorithm, and at this point, the legacy algorithmic networks have conditioned its users to view the constant barrage of content, and the constant hits of endorphins, as "normal." But do the above, and bam, you now have an active feed of only things you elected to see and nothing else.

Stick with the slowness. You'll see. At some point (very soon) you'll wonder how you ever put up with feeds that were 85% made up of content you did not actually sign up to see.

After that, if you are having a good experience, talk to people about the platform you just joined! Tell them how much better this is than the platforms that make pushing content they want you to see (not what you want to see) their entire reason of being. No ads. No tracking. No surveillance. No "promoted posts." You can see content from other federated platforms. Invite a friend or two. If it's slow to start, well, so be it. But you can be the start. You can be the beginning of a new network effect!

Demand Interoperability

One of the largest roadblocks, and a massive component of the switching costs of leaving a platform, is the fact that you cannot take all your stuff (posts, images, connections) and simply port them to a different platform. That's, of course, by design. These fire exits were intentionally left out.

It does not have to be this way. If you want to migrate your mastodon account to a different server? You can. You can export your follows and followers, and your mute and block lists, as a plain CSV, and simply upload them on your new account. You can also redirect your old account to your new one, so that visitors get redirected, and you can move your followers over (!). 4

There is another aspect to interoperability, one that is also implemented in the Fediverse. And let me preface this by saying that I don't like equating Fediverse platforms with the legacy algorithmic media they most look like, but just to illustrate how cool interop is, I'm going to use some equivalencies here:

Let's say your buddy prefers the layout and functionality of Facebook. You're more the microblogging type. So you set up an account on Mastodon or Pleroma, and you buddy goes with Friendica. You can follow each other. You can see each other's content and interact with it.

None of this is technologically impossible. It's just that multi-billion dollar corporations have spent decades trying to tell us that it is.

So how do you "demand" interoperability? Keep submitting feature requests on platforms to add interoperability. They may not listen to you; that's ok. Find organizations that push for interop and standardization. Follow them, follow their work. And, of course, find interoperable platforms, create accounts there, use them, and share your experience with others!

... and More

Here's a couple additional suggestions to let simmer in your brain:

Questions? Thoughts? Feel free to get in touch!

Recent Posts

  1. That's a first!↩

  2. Mastodon-style. So, like... fifteen boosts and quotes. HECK YEAH 🤘🏼↩

  3. In a very 2022 moment, which is when this article was written, Forbes also claims that " blockchain applications like cryptocurrencies are a natural element of the new global network." I'll just let that sit there :)↩

  4. If this sounds like magic, it's because a lot of companies have spent years telling us this isn't possible. (And sometimes it isn't, like when the origin instance is down for an extended time, but those are edge cases.)↩

#algorithmic media #blog #enshittification #social media