The Network Serves Us

The recent growth in goodwill around Bluesky, a social network that has long been under the radar, has been pleasing to see.

Given the utter chaos of the recent political climate, it has been an amazing reprieve to not have to only focus on the messy political climate, which honestly doesn’t have very much to offer right now.

In recent months, it has felt like our social platforms just haven’t been meeting our needs. There are many reasons for this, but there has been a sharp disconnect between audience, platform, and experience. One bad social experience after another has left an opening for another player. And that player has started to emerge.

Somehow, over the last couple of weeks, Bluesky has brought all the elements together to create a platform that, as of right now, seems to have solved nearly all of these problems, to the point where Ryan Broderick suggested it basically won the “new Twitter” battle this week. It has topped 15 million users—and unlike Threads, it did so without piggybacking on another existing network.

And it did it by putting the user in control.

What’s interesting is that so much of its growth is happening essentially through organic tools. Sure, there are algorithms, but what is fascinating is how they don’t feel incredibly visible. The company’s decision to create “starter packs” that users can share with one another has created an absolute explosion in followers in a short amount of time, that, combined with the accelerant of the election putting pressure on legacy social networks to change direction, suggest that what many users wanted all along was not extremely high levels of engagement, but an experience they more or less feel in control of.

That, to be clear, is not a given. For years, nearly every major social network has taken steps, in the name of its business model, to minimize the impact of basic digital tools like links, self-promotion, and even the ability to talk about desired topics. And these decisions seem to be driven by business calculus, not the needs of the user. And it just leads to awful experiences.

Over the weekend, for example, I tried posting on Threads about mini PCs, just making the point that, hey, the new Mac Mini is cool, but that you could get a relatively nice AMD mini PC for the same price that has some advantages. You know, the kind of thing I would post.

Suddenly, I found myself fending off maybe two dozen replies from people who felt like I was attacking their thing, rather than simply pointing out the nature of the market. The conversations got heated. They weren’t friendly. And by the end of the night, I was steaming. I deleted the thread, blocked the people trying to start arguments, and decided that Threads, a platform I went to tentatively, and only after a similar conflict, wasn’t for me.

I have been able to revive old posts I liked on Twitter and put them on Bluesky, where they were also successful.

The reason my interactions went south? Blame Threads’ highly tuned algorithm. Essentially, you talk about a hot-button topic, and within a matter of minutes, it puts you directly in front of anyone who has shown interest in the topic. They don’t know you. You don’t know them. Written out, the result sounds like a Quake deathmatch from 1997, except with text. But Threads is going to make you interact anyway. It’s not natural, and it reflects the work of a company that has invested too much in the algorithm to do it any other way.

Bluesky, by contrast, does not reward opinion-broadcasters. You have to put in the work to be engaging. You have to share replies and do the heavy lifting of building an audience. But, with the addition of starter packs, it works—and super-well, too.

It should be noted that the fediverse, to its credit, also works like this, where a successful following is built on a ground game. If you’ve done the work, you can have a deeply engaged audience. But unfortunately, other social networks have decided that it’s not the best way to build a network and make money from it.

I think to earlier this year, when I wrote about how absurd it is to get a link to fly on LinkedIn, and how that contrasts with a network that makes linking a first-class citizen. Experience-wise, the difference is night and day. And I think Bluesky should take steps to lean into that.

Honestly, if Bluesky was smart, they would use the base of the AT Protocol to build out other types of networks, and charge for some of the services they offer, but leave the primary platform alone. I could see it now:

• An Atproto version of LinkedIn, where people could engage professionally and look for jobs without all the arbitrary tracking and forced showboating that LinkedIn pushes.
• An Atproto version of Facebook Marketplace, built with simple, effective search and a modest, simple cost.
• Creator tools that could help people who are trying to make money off of their social presence do so more effectively. Think Patreon or Substack, but built to work most effectively with Atproto.

The problem that most of Bluesky’s predecessors have faced is that they have turned the money-making spigot too far over to the right, which has led them in the direction of algorithms and personalization that harm the experience. (Sometimes, the handle just straight-up flies off, which one might argue has been happening to Threads—despite, ironically, not having advertising yet.) Bluesky, by building a ground-up protocol, has pushed things in the other direction. It is not trying to heavily personalize or build algorithms. As a user, that’s your job. It is instead focusing on building a strong central identity, and a good user experience that the user ultimately controls.

The real play is to build a really strong underlying system that can be used for any kind of social network that you can think of, and using those ideas to fund the main social network, which effectively promotes everything else thanks to its broad scale. They put in all the hard work up front. Now they benefit from having an easier time going forward.

I don’t know if that’s what the team at Bluesky is building, but there is a contingent of users out there that is clearly getting sick of being pushed around by algorithms that could use a company that actively chooses to push against screwing with the experience and the algorithms in favor of building a strong ecosystem that benefits from the halo effect of running the best social network out there.

The network should serve the users, not the other way around. It is shocking that it has taken this long for someone to figure it out.