It's easier to imagine the end of the world than the end of Instagram
Published on 2024-10-09
In early 2021, I published an article on my website where I talked about dark patterns spreading in the UI of the fediverse.
The Fediverse only solves half the problem.
A week later, I published a second article codifying a design philosophy that I called "ethical anti-design".
Ethical anti-design, or designing products that people can't get addicted to.
This was actually very well-rehearsed. At the time, I was finishing my senior year of high school and I was in a program where I'd have the opportunity to dedicate two hours that I would've otherwise spent in the classroom working independently on a project I got to choose. I decided that I wanted to make an app, and over the next five months I'd spend most of my free time working on what'd go on to be the side project I cherish the most: Resin, the social media client you won't get addicted to.
nat/resin: Pixelfed client employing ethical anti-design - git.nats.solutions
But while it was well-rehearsed—it was literally a high-school project I was turning in to my professor—I didn't expect anyone to notice. I didn't think anyone actually read my website, but at least one person did, and they shared it to Hacker News in what was the first and hopefully only time something I published landed in that sludge pit of a website.
A number of people seemed to think it was interesting. A lot of people didn't like the name "ethical anti-design". A lot of people reached out to me to tell me that they didn't think I should be calling it "ethical anti-design". But they kept talking about it. I definitely let it go to my head a little, and I'm trying to be a bit careful in talking about it now for that reason—this was the first and for all intensive purposes last time more than one person had ever shown serious interest in something I was working on, and it quickly took over my life.
I was really emotionally invested in the success of Pixelfed, like many at the time. This was way before the release of the official Pixelfed client; there were a few you could find around, but they were all missing key features. I was uniquely endowed with not needing to pay for rent or food or school and a limited social life, and I was able to plow ahead, building this thing that I really hoped would change the way people think about how we interact with social media. Pixelfed was Instagram for the Fediverse, and as someone who spent all their time in a space where Twitter and Facebook were old people websites, I saw Pixelfed as something I could more viably convince my friends to join. But, of course, it needed an app first.
By the end of the semester, I finished what I had liberally¹ called v1.0-beta and handed in my final report. At the time, I was pretty gung ho about continuing to work on it, and I did try, but needless to say, life got ahead of me. A few months later and I was leaving home for university. Another year had passed before I seriously sat down to work on it again. By that point, Expo—a hilariously bloated part of the tool chain I used to build Resin—had gone through like seven major version changes for some reason. The latest version required I download an app that targeted an Android API version my phone didn't support. That was pretty much the final nail in the coffin for me.
It's not like it would have been impossible to fix though. Before I'd even realized this, I had opened what now feels like a very forlorn issue on my page for it on Codeberg:
#1 - Ditch Expo - njms/resin - Codeberg.org
But by this point, the passion I had for Resin—and by extension Pixelfed—was pretty much gone. I think that's mainly what I want to talk about in this article. All that was just exposition; I've wanted to do a "follow-up" on my first two articles basically since I published them. Here's what I've learned in the years since:
I must have thought it was easier to imagine the end of the world than the end of Instagram. I'm a bit embarrassed that I couldn't see it at the time, but I suppose I was a teenager. We can all be forgiven for being a little naive when we were teenagers. That doesn't extend to the people developing Pixelfed and projects like it; I just feel like in hindsight, we were working on solving very different problems.
I first started using Instagram when I was a literal child. Way before I was legally allowed to enter a contract, and way before Facebook and the independent Instagram company before it were legally allowed to collect my personal information. And way back then, Instagram was how I shared my life with my friends outside of school.
There was something really beautiful about that, and as it turns out, very lucrative as well. Instagram seemingly pioneered a lot of social engineering techniques that'd be developed further in Vine and Snapchat, and finally perfected in TikTok. When I was a kid, Instagram felt homey. When I revisited the platform earlier this year, it felt more like an IV bag by the time I'd gotten used to it again.
I don't blame people for looking at Instagram today and wanting to see it succeeded by some new, community-led photo blogging alternative, but I fear the people who are—myself included—aren't looking far enough back in its history. I can't help but feel like the reason why Instagram grew into the monster it is today, rather than, say, any other competing social media platforms that have all since tried and failed to rip it off to the same effect, is because the concept of Instagram itself is prone to this sort of thing.
When I wrote "The fediverse only solves half the problem," I seemed to think that fediverse developers were recreating these dark patterns by mistake. I certainly don't think they were actively trying to make their websites addictive, but today I think the problem is more systemic. If you're trying to make a truly "post-Facebook" social media platform and you start with "it's like Instagram, but for the Fediverse," then you've already lost.
But that's the thing: I don't think dansup, Eugene and friends are trying to create a "truly post-Facebook" social media platform; I was wrong for thinking so. It's a lot more obvious to me now that they're reformers. And for what it's worth, I hope they succeed in making something better. Anything better than what we already have. But these days, I'm having a much harder time getting myself excited about social media reform.
Ethical anti-design is, as so many people pointed out to me, an oxymoron. It is fundamentally a very contradictory design philosophy. It's uncomfortable to interact with by design. That awkwardness is what lead me to explain Resin as more of an art project than an app I'd reasonably expect anyone to want to use. I mean, its main conceit was that instead of having infinite scroll, you'd have to press and hold a button for a second to load more pictures. Resin never interrogated why posts need to be structured in an apparently infinite gallery for the interface to be intelligible, it just treated the dark patterns as bugs to be patched. And that's what ultimately made Resin a reformist project itself.
There's a reason why old-guard web users still insist on prefixing their Reddit links with "old.". Unlike Instagram, which feels a lot more like a new idea that would quickly evolve into the dominant mode of production of social media, Reddit was different. Reddit feels more like the way the Internet used to be, only evolved in a lazy attempt to make it fit into the new way people share content online. It is, at its core, more like the old web than the new
That's not to give them too much credit either. Reddit is what it is today; "old.reddit.com" is a fossil of how things used to be, and today I can't visit it without being fed the most vile and hateful garbage I will inevitably see all week.
So I guess in the end, I stand exactly where I stood in 2021. The Fediverse presents an opportunity to radically re-imagine the way social media could be², and in general it seems like we haven't taken it. I thought I knew how, I tried really hard, I failed, and I learned quite a bit from it.
I'd really love to see more experimental social networks in the world. Maybe one day I'll have the bandwidth to try.
Footnotes
¹ Resin was missing a lot of things I promised myself I'd implement. One of the things I really wanted to see was support for direct messages, but as I learned the hard way, at the time, a lot of stuff that Pixelfed ostensibly supported wasn't actually exposed through its API. Not to mention, as Pixelfed was itself in beta, things were changing constantly and it got really hard to keep up.
The reason Pixelfed didn't have a feature-complete mobile client wasn't because nobody had gotten around to it as I had thought—it didn't have a feature-complete client because building one was practically impossible at the time. It wouldn't be another year or so before the official client was well underway and fixing those issues was prioritized.
² It's more than worth mentioning that ActivityPub itself was guided by the "dominant mode of production" of social media of the time it was specified—namely, that of all the W3C's biggest sponsors. But, that's not to say we can't use the latitude it gives us to think more outside the box.
Respond to this article
If you have thoughts you'd like to share, send me an email!