We Need a New Paradigm for the Social Web

Earlier this year, Twitter— er, “𝕏”, suddenly killed off access for third-party clients and substantially hiked its API pricing. More recently, Reddit pulled effectively the same stunt, inciting a short-lived rebellion that saw thousands of subreddits shutting their doors. Granted, Reddit’s pitiful 30-day notice was still 30 days longer than Twitter’s, but that doesn’t make the whole affair any less horrendously communicated, on top of just being a rotten idea. What’s clearly observed from both of these companies is the desire to clamp down on their user’s data, to ensure you can only interact with their website on their terms.

It’s hard not to feel a sense of impending doom hanging over every online service we use nowadays: they start out as nice places to be, but over time become increasingly user-hostile until (if you’re lucky) enough people jump ship to make an alternative viable. This is the cycle of enshittification.

Here is how platforms die: First, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die.

Cory Doctorow

For today’s average tech startup, enshittification is in the standard playbook. Initially, the fledgeling startup’s number one priority is acquiring maximum market share while burning through mountains of investor cash. Only once they’ve achieved a dominant market position do they start squeezing their users in a desperate attempt to become profitable.

The key to enshittification is that it only works when your customers can’t leave. One way to do that is to be the only game in town. But for social media, the lock-in is based on network effects. For any website that relies on user-generated content, there’s a certain critical mass of users required for the site to be practically functional. Below that threshold, your site is basically a dead mall. Plus, leaving a social media platform is hard because you can’t take your stuff with you. You can pick up your files and move them from one cloud storage provider to another, but you can’t do the same with the people you follow. And so we arrive at the modern internet: five giant websites filled with screenshots of the other four.

Resilient as they may be, enshittified platforms can fall. Digg was replaced by Reddit. Discord displaced Skype. But this isn’t a solution, just a short reprieve until the usurper inescapably repeats all the sins of its predecessor. This failure mode is an inevitability of the current paradigm. We need a different model for the social web.

Do you know what has never, and will never be enshittified? Web feeds.

Web feeds are a fantastically useful bit of tech that’s tragically obscure nowadays, so I’ll provide a brief overview. A web feed is a standard machine-readable format (such as RSS) that websites like blogs or news sites can choose to publish their content in. A separate website/app called a feed reader or aggregator (I use NetNewsWire and Feedbin, personally) then pulls in content from any feed you’ve subscribed to and presents them in a unified list.

(Yes, this blog has a web feed.)

What I love about web feeds is that they truly put you in control of your experience by decoupling the content you see (the feeds) from the software you use (the reader). There’s no lock-in here: any website with a feed can be accessed by any feed reader, so long as everyone’s following the spec. And that interoperability isn’t just good for user choice, but also for stability. In 2013, Google shut down their feed reader product, Google Reader, forever cementing their reputation as a company that will mercilessly kill off that product you love. Some people certainly gave up on web feeds after that, but for everyone else it was just a minor disruption. Any former Reader user could pack up their subscriptions (there’s a format for that, too) and move them to another feed reader without too much hassle.

Now, from an end user perspective, web feeds are strictly a read-only technology: publishers publish content, consumers consume it. Contrast that with social media which is interactive by nature. But there’s no reason that the same principles of decentralization and interoperability couldn’t be applied to an interactive experience— like, say, Reddit.

Reddit is essentially just a bunch of forums stapled together, so what made it so successful while traditional forums have become more niche? The voting system and nested comments were novel, but not technically difficult to replicate. No, what Reddit offered was consolidation: the aggregation of its communities’ posts into a single feed. Using Reddit was more convenient than visiting dozens of separate forums. We’ve been led to believe that centralization is a requirement for this kind of experience, but this is the exact problem web feeds solve for blog-like content. Imagine instead a network of independently-hosted web forums which all communicate via a common protocol, accessed and aggregated by whatever client you choose. Using a third-party client wouldn’t be a privilege granted and revoked at the whim of Reddit Inc.; it would be an integral part of the system, baked into its very design.

There are, I think, some technical problems that need to be solved to make this social media model truly viable— identity, for instance, seems to be particularly hard to get right under this model. But I think it’s the best shot we’ve got at building a truly sustainable and user-focused social web. New social media projects will always face the challenge of getting past the “dead mall” phase, but I still hope to see more new projects embrace this philosophy gain a healthy niche userbase.

Now, I’ve probably come across as hopelessly optimistic up to this point, so let me bring this back down to earth: I don’t see decentralized platforms ever breaking into the mainstream in a substantial way. It’s been made crystal clear at this point that the general public just does not care about this stuff. That’s why web feeds are hopelessly niche, and why Discord is eating all the forums Reddit didn’t kill. And it’s why Mastadon, a federated, open-source Twitter alternative was absolutely dwarfed by Facebook’s Threads overnight.

As always, relevant xkcd.