Over the last two weeks, I've been running an experiment with my online use just to see how it went: While I have been using Buffer and Freedom to automate and limit my Twitter exposure for about a year and a half, what I've done over the last two weeks is use Twitter purely for sharing Mere O content. (Ed. note: I refuse to indulge the petulant space billionaire and call it X.) Everything else I might have done on Twitter I've instead simply not done or I put it into my email newsletter. It's actually gone quite well.
It turns out that, contrary to what years upon years of social media conditions us to think, you actually don't have to post things every time the thought occurs to you to post things. You can just... not.
If it was something I wanted to share, I do it in the newsletter—and I have several days to think about it and decide if that really needs to be shared in the newsletter or not. Knowing that you only have one broadcast social message a week as opposed to an infinite number of them changes how you think about that message. And with the newsletter I know who my audience is, I control my audience to a large degree, and I have much more control over the format and context in which whatever I want to share is seen.
One question this has raised for me: There's a fairly clear demarcation between the pre-social media internet and the post-social media internet. The good with the former was that the internet was more human and smaller scale. The bad was that if you wanted a personalized web presence, you needed a relatively high degree of tech knowledge to pull it off.
The social media age flipped that: The human was replaced by the algorithm and the scale became quite industrial. But, on the other hand, tools like Facebook, Twitter, Tumblr, and Instagram made it quite easy for anyone to access the internet and create a personalized experience for themselves which also made it easy for them to find like-minded people and to surface content they enjoyed or found interesting. The metaphor doesn't work perfectly, but in some ways the first era was Linux for a desktop and the second was Apple's iOS mobile operating system: Linux affords the user far more control over their experience, but is only accessible to people with the expertise to use it (or the time and ability to learn how). Apple is infinitely more accessible than Linux, and yet it is a walled garden where you play by Apple's rules. In a similar way, Facebook made it far easier for you to have an online profile and to surface content you wanted to look at. But it came at the cost of subjecting yourself to Facebook's algorithm.
Here's my question: The value of social media networks is basically their social graph: The larger a social media platform is, the more likely it is that you will have lots of friends on it and that you'll connect with those friends. And that complex web of connected people is the real value proposition for legacy social media. The problem is that as people become more autonomous and fractured, the social graph loses some of its value because the people they were connected to ten years ago on Facebook might not be the people they want to follow today. Moreover, as digital networks and platforms multiply, it becomes harder to create a sufficiently large social graph to offer that value to users anyway. Tellingly, the only social media platform to really explode in the past several years has been TikTok, which isn't actually social media at all. It's an entertainment app with a powerful algorithm.
Moreover, as social media audiences splinter, the social graphs that will unify us in the online world will almost inevitably be political for reasons we've explained in the past. It's hard for me to see a path for, say, Bluesky to achieve anything like the power and status of Twitter just because it's coming along 15 years later in a time when society is so fragmented and atomized that it is virtually impossible to obtain the kind of reach that was once normal for Twitter, Facebook, and perhaps even LinkedIn. The days of universal social media networks may well be at an end. So as social media becomes more politically coded and as the dominant social network in media declines, I can't help wondering if we're moving into a new era of the internet.
At my most optimistic, I wonder: Can we use technology from that older era—primarily RSS + email—and combine them with new tools meant to make those technologies more accessible and easy to use? Use tools like Feedly, Buttondown, and Substack to fulfill the ease-of-use problem that social media solved for people. But, crucially, those tools must either be owned by independent producers simply trying to make a living, not build a tech unicorn, or they need to be owned by a company that really does understand its role as an information platform, not a social media network.
In other words, the easy technologies we're using are not algorithmically driven with the goal of maximizing individual engagement. They are, for lack of a better term, more human—you're either subscribed to an RSS feed or you aren't; you are subscribed to an email list or you aren't. With RSS and email, we mostly take the algorithms out of the equation altogether—with some wiggle room around the spam box in people's inboxes. So could this be a way to have a human internet and an accessible internet?
Probably not. Things online never go that well. But could this time lead to a radically different internet and perhaps a less algorithmically defined internet? Perhaps. And after the last 10 or so years of online life, I can live with "perhaps."