Matt Gurney: Can we build a better internet?

What was the absolute best version of the digital world that you can remember? And how can we get that back?

By: Matt Gurney

Have you heard about enshittification? It’s not just a potty word. It’s actually a pretty fascinating concept, and you read about it mostly in tech circles. Enshittification is the process by which something becomes worse over time, instead of better, normally as people try to squeeze more efficiency and revenue value out of it. Through that process of squeezing, the thing becomes enshittified.

If you want a proper definition …

“Enshittification: The gradual deterioration of a service or product brought about by a reduction in the quality of service provided, especially of an online platform, and as a consequence of profit-seeking.”

There are lots of examples. My favourite example? My video doorbell has an annual service fee. Another great example? Cars that now require payments to access certain features, like heated seats. You own the device. But you need to pay a recurring fee to use it. That’s enshittification.

It’s everywhere. And it’s getting worse, especially online. And, perversely, maybe this is a good thing. Maybe it’s going to force us to stop, rethink how we use the digital realm, and, basically, try again. Start over. And get it right this time.

Noah Smith is an American economics writer whose work I enjoy. Smith noted on Twitter recently that we are rapidly getting to the point where we should declare social media a failure. It’s passé to criticize social media on social media, but Smith wasn’t making the usual warmed-over moral argument. He wasn’t saying that it was bad because people are mean there or that they fall down dark rabbit holes and end up believing insane things. Those are problems! But Smith’s concern was the extent to which AI-generated content and bots have simply flooded all the social media channels. Even a responsible user trying to use these platforms for the good is going to find it increasingly difficult to derive any value from them. They’re being rapidly enshittified.

I share his view of the trajectory. I don’t really know anyone who doesn’t. But Smith’s comment led me to ponder what value I actually derive from them — what I would miss if they were gone. I came up with four broad use cases.

One is communicating and sharing info/documents/photos/videos with a small group of family, colleagues and friends. That still works well, but it’s social only on a micro-scale — sometimes one on one. That could easily be shifted to emails or text messages. Social media and messaging apps are only a minor improvement over that, and incur other costs.

The broader use of social media — large, even global scale — offers three uses. One is using it as a news ticker to monitor events in near real time. Another is simple entertainment — talking sports on Twitter and scrolling dog videos is fun. The third is using it as a business-growth tool, a way to share The Line’s work and reach new would-be subscribers.

Smith is right; the cost–benefit ratio in each category is deteriorating, on both sides of the equation. As bots and AI-generated content increasingly dominate the platforms, and this is already obviously happening, I could easily see the ratio becoming untenable.

And then there are the broader, established failures of social media. They’re disinformation superchargers. They’re contributing to polarization by amping up discord and the appearance of discord. They’re taking a real, measurable mental-health toll on people, especially the young. They’re addictive.

So what are we doing? What part of social media, or the internet more broadly, is worth keeping? What was the optimal use of the internet? If we could design it from scratch, knowing all that we know now, what would that look like?

It’s worth noting here that I have somewhat sloppily conflated the entire internet itself with social media. There are certainly enshittification problems with the broader internet; search engines becoming useless and pop-up-riddled websites being unreadable are real problems.

But the issues I actually worry about are largely limited to social media. That’s good! Isolated problems are easier to solve.

As a thought exercise a few days ago, I began to think about how I actually use the internet on a day-to-day basis. And one of the things I quickly realized is that all the stuff that is good about the internet is largely, almost exclusively, an evolution of something that existed before the internet but is now better and more efficient. Renting movies, banking, shopping, staying in touch with friends and family, accessing some government services, swapping family photos, accessing and enjoying music, running an office (especially The Line’s, which is spread across the continent) … all of these things were possible 30 years ago, but are a lot easier today. And as a reservoir for knowledge — how to cook things that are tasty, how to fix things that are broken, how to learn a new language or learn about a new place or people — the internet is a spectacular gift to mankind.

(And in case you’re thinking I’ve missed an obvious one here, yes, something like The Line itself could have existed 30 years ago as a newsletter or magazine — or zine — but is much easier, for producers and consumers, today.)

So I do not hate the internet. It did a lot of good. But social media isn’t really an evolution of anything we had before. I’ve never really bought the concept of social media just being a digital version of the public square. Human beings are social creatures, sure, but the meat computers inside our skulls were optimized to handle vastly smaller communities, at much slower speeds, and without searchable archives of everything previously uttered. We’ve all had dumb ideas and dark jokes that didn’t land, but the ability to broadcast it at the speed of light to a global-scale audience didn’t exist 20 years ago.

That was a good thing! When Smith said that social media has failed largely because it will be overrun by useless slop content, I think he’s right. But I think it already failed a more basic test. On balance, it’s made our lives worse.

A few months ago, I interviewed Andrew MacDougall for an episode of On The Line. He proposed switching to a user-fee model for social media. Not only would this disincentivize the companies from pursuing user addiction as a business model, but it would start pushing us off the platforms that are causing us harm. Price signals work; I don’t use social media a lot, but I’d use it less if I was billed by the minute.

I’m increasingly convinced that Andrew has got the problem correctly identified, even if figuring out a practical solution would be hard — imagine the lobbying, good Lord.

Still. He’s right. And I’m honestly open to the prospect of going even further.

Should we maybe call a mulligan on this internet thing and start over?

Easier said than done. No kidding. But think about it. The internet transformed our world in only a couple of decades. Many of those transformations were amazing. Some of them have been really bad. And it isn’t impossible, in theory, for us to actually ask ourselves what an Optimized Internet for Western Civilization (TM) would look like, and then build it.

The best rebuttal to my suggestion is simply to wonder which Canadian heritage minister would get to decide what the New Internet looked like and how we were allowed to use it. I’ll grant that point! But I still think it’s worth thinking about, even if political pragmatism forces us to stay in the realm of the thought experiment. What is an optimal internet? How could we design a digital realm, deliberately, that maximizes the real benefits to society, while minimizing either the accidental downsides or the opportunities for deliberate manipulation and sabotage by enemies foreign or domestic? Can we keep the benefits while minimizing polarization, online harassment and other crimes, and the real and growing risk of cyberattacks on critical infrastructure?

And if we can, like, maybe we should!?

History shows a pretty consistent pattern when something new is invented and becomes popular. It becomes ubiquitous … and then, later, we realize what the downsides were, and the process of optimizing it begins. What will that process look like for the internet, and for social media specifically? We haven’t really begun to have serious conversations about that yet. People are talking about it, but mostly insiders, and only at a theoretical level.

That’s fine. That’s where it starts. But maybe it’s time to move it a bit more out into the open. Let’s all start thinking about what that kind of internet would look like, even if that includes some kind of deliberate and phased retreat to something like an evolved version of the internet as it existed before social media. The New Internet might look a lot more like the internet of 2005 than of 2025, but with 4K streaming services and decent online gaming.

Would that be bad?

This would be hard. But I think Smith is right — the internet of right now, particularly social media, is headed toward failure. It’s not too soon to start wondering about how we could avoid that failure, or, if necessary, start again after it.

---

The Line is entirely reader and advertiser funded — no federal subsidy for us! If you value our work, have already subscribed, and still worry about what will happen when the conventional media finishes collapsing, please make a donation today. Please note: a donation is not a subscription, and will not grant access to paywalled content. It’s just a way of thanking us for what we do. If you’re looking to subscribe and get full access, it’s that other blue button!

---

The Line is Canada’s last, best hope for irreverent commentary. We reject bullshit. We love lively writing. Please consider supporting us by subscribing. Please follow us on social media! Facebook x 2: On The Line Podcast here, and The Line Podcast here. Instagram. Also: TikTok. BlueSky. LinkedIn. Matt’s Twitter. The Line’s Twitter. Jen’s Twitter. Contact us by email: lineeditor@protonmail.com.