I work on problems with networks because I am an ordinary person and I care about ordinary people. Sometimes caring about ordinary people means finding new ropes to pull on and throwing your weight into it.

For those of us inside the US but outside the VIP lounge of American wealth and power, all our common goods and civil protections are on the chopping block: clean air, safe food, vaccine access, basic science, medical research, reproductive healthcare, gender-affirming healthcare, due process, equality, freedom of movement, of association, of religion, of speech. Freedom from the arbitrary exercise of raw power. All the things. As usual, the instability and suffering the US is exporting to previously colonized parts of the world is orders of magnitude worse than what we experience here as a result of our election of a pack of venal and disordered oligarchs. But people here are suffering too, and it’s going to get worse.

In times of crisis, the (sound) advice from old movement heads is to do what you’re already good at, right where you are. What I do as a person on this earth is to try to understand and explain tricky things in useful and unflashy ways. Where I am is here, with you.

Maybe you can see where this is going.

Knowledge in networks

I’ve been immersed in network problems for the past couple of years: Culture problems and their relation to structure and shape, knowledge ecosystem damage and its relation to…structure and shape. The way we’ve been building holding pens and large context colliders when we need connected refuges. But where I’m from, “professionally,” is working on what the networks should contain: context, knowledge, sense, and ground truth. Those are the things I fell back on with the Covid Tracking Project, five years ago last month—because when things are truly bad, you will not make it without knowing what is happening around you. This is the mode I’m falling back on now.

My various modes and problems are connected, but it’s taken me a few months to gnaw through their disparate machinery to understand how. (I am, in this figure, a lightly charred cartoon rat with a sparking wire in its mouth. Go, rat, go.)

Last month, I gave a remote talk at the first ATmosphere Conf. There’s a recording, but I gave the talk remotely because I was sick as hell—so for real, watch that at your own risk. (My citations and notes are here.)

In the talk, I focused on two things I wanted to seed into conversations around the nascent ATProto ecosystem: The first was an attempt to lure more network developers who don’t come from especially vulnerable communities into thinking with more people in those communities—whether through what I talked about as “collaborative threat modeling” or through a more explicit design justice framework. I got to pull in Ursula’s Franklin’s always-illuminating clarity about fear there, and I hope it was helpful to the folks who assembled in Seattle and on the livestream.

The second part of the talk centered on an idea I’ve been wrestling with for a while, which is my sense of vernacular institutions. And that’s the part I want to get into in more depth. (Dan Hon’s notes on this part of my talk are excellent—thank you, Dan.)

Vernacular institutions

Vernacular institutions as I am thinking of them have three essential qualities:

  • They emerge from highly specific local conditions.
  • They prioritize needs on the ground over those of centralized powers or abstractions.
  • They are more useful than they are legible. (Some of that usefulness is maintained under the protection of illegibility.)†

Mutual aid organizations are necessarily vernacular institutions—unless or until they’re captured by private or non-local interests. Black Panther food programs and medical clinics, including the clinic that’s still running were vernacular institutions; so were Freedom Schools and local ACT UP groups. On the fediverse, I think IFTAS is a vernacular institution—as are the servers that make up the fediverse as a whole. Blacksky is an emblematic vernacular institution in the nascent ATmosphere, and I highly recommend founder Rudy Fraser’s ATmosphere talk.

I think existing institutional structures—especially but not only corporate ones—serve our global communication networks so poorly because they’re the wrong shape. I’ve written here about the necessity of finding new ways to build networks that offer both shelter and connection. I’ve written elsewhere about the value of governance according to local norms in connection with my research with fediverse server teams. And I’ve talked about the tracking work I did with hundreds of volunteer crisis workers, and which was itself a somewhat illegible blend of journalism and mutual aid. This orientation made our work extremely useful to news organizations (because we weren’t competing with them) and highly deniable as real journalism (because we weren’t competing with them). It cuts both ways.

The red thread running through all my grappling and hollering is the necessity of institutions that emerge from local conditions and serve the needs of ordinary people while remaining largely illegible to centralized power.

I’ll write more about the character and uses of vernacular institutions when I get my head a little further above water. But right now—having written my way to the definition of vernacular institutions by thinking about networks—I’m working with old friends and a couple of new ones to found a new and extremely vernacular institution for our moment. At Unbreaking, we’ll be spending the foreseeable future dredging up the knowledge from which we can build a durable and coherent sense of what is happening to us—collectively, and at scale. (I would like the record to reflect that I tried to get other people to just go do something like this without me, but here we are, beep boop.)

You can sign up to be notified when we soft-launch, and when we start bringing more people into the work. Once we’re launched, my co-founders and I will be writing about the thinking behind our work and the way we’re doing it so that other people working along similar lines can use it to strengthen their own projects. I’ll put the networkiest bits on this site, and link out to the rest.

And of course, all my other network things have been happening, too.

Little [network] things, ceaselessly happening*

Starting at the end of last year, the essays and talks and research I’d been working on started resulting in new collaborations and conversations and projects, and it felt like overnight I went from working out in some kind of modestly haunted lighthouse and into the heart of a bustling market town. This feeling lasted until January 21, 2025, when the contours of the new US administration became visible and concrete. Focused work got un-focused for a month or so. Then it came back.

I’ve joined the board of A New Social, a new nonprofit coming out of Ryan Barrett’s work on Bridgy Fed and Anuj Ahooja’s thinking about the perils and joys of bridging and interconnecting decentralized networks. My work with Ryan and Anuj has so far mostly consisted of working through variously sticky questions closely related to the kind I wrote about in “Bridges & Scruples”. Ryan and Anuj are trying to accomplish some culturally and technically tricky things, and I appreciate their commitment to getting them as right as possible. (We’re also all in agreement that we could really use some institutional support for collaborative threat modeling for new networks and the connections between them. If this sounds like an appealing project you can help set up and fund, you know where to find me.)

I also joined IFTAS’s advisory board at the end of 2024, focusing mainly on various kinds of documentation, and I’ll have more to share about those things later this spring. And I’m working a little bit with the wonderful crew at Bonfire Networks as well, currently on planning something we’ll also be able to dig into later this spring.

The first half of a long and crunchy conversation with Ethan Zuckerman and Mike Sugarman at Reimagining the Internet went live as podcast episode last week, with one more to come, sometime soon. I got to do two sessions at Sabrina Hersi Issa’s wonderful Rights x Tech convenings, one last week on networked knowledge projects with John Mills from Watch Duty and one back in early March on care and grief and failure-states with Malkia Devich-Cyril of the Radical Loss Project. Recordings aren’t available, but if technology for justice and human rights on your radar, I highly recommend getting with Rights x Tech.

At the end of last year, I recorded an also-crunchy episode of Mike McCue’s Dot Social podcast with my fediverse research collaborator Darius Kazemi, and I don’t think I mentioned that here.

There’s a chunky open-source meta-documentation project I’ve been working on that is making its way toward the surface of the internet as well, but I’ll save that for another post.

Things on hold

Not everything is going forward. I’m still devoted to the idea of making it easier to find a good home on the fediverse, but my own work on that has been on a hold in response to the intense levels of stress many fediverse server teams have been experiencing. A handful of teams were able to speak with me about their work and thinking, and I am so grateful to them for making it possible. Many more, though, have been consumed by other more pressing needs—some related to the crisis of US governance and others not. I’m extremely interested in not becoming a drain on the mental resources of over-stretched server teams, so I’m holding on this project until a few more teams have brain-space to speak with me.

Another beloved plate I’ve decided to fully set down for now is the interview series I’ve been looking forward to getting going on wreckage/salvage. I love interviewing, and it gives me a lot of fuel. It’s time- and brain-consuming to do it warmly and well, and while I’m launching a new organization and meeting my other commitments, I can’t do it justice, so that’s on a shelf, waiting for a better day.

Thank you

These last few months have been brutal for a lot of us. And let’s be real, the work I do here at wreckage/salvage has felt like a bit of a long shot from the beginning! That shot feels longer, now. But, you know, I’m going to keep going—and that I can keep going is thanks to those of you who’ve thrown in with me. Thank you. (If you want more of my particular work to exist in the world, this is where to help that happen.)

And the remit here remains the same: I said last summer at XOXO that whatever happened in the US in the November elections, rebuilding our networks of knowledge and community was going to be essential. I believe it just as much now.

I also said—like a dope who’d never heard of foreshadowing—that I didn’t think that we could do something like the COVID Tracking Project with the internet we had in the summer of 2024. Maybe the internet of 2025 is different enough to support a new and equally unlikely communal knowledge-building project. I guess we’re about to find out.

Notes

(*) A long time ago, I moved cross-country and enrolled in grad school specifically to study the work of (formerly??) "lost" modernist poet and novelist Hope Mirrlees and the band of queer-coded scholars and writers she hung out with. Her Paris is always running as a background process for me, and this essay, titled for a line in that poem, was good to see.

(†) I am obliged to nod to Ivan Illich’s uses of “vernacular,” but my own thinking comes more from Architecture Without Architects, Christopher Alexander, and James C. Scott (CIA running-dog accusations notwithstanding).

The featured image for this post is a detail from Bruegel the Elder’s wonderful The Gloomy Day (Early Spring), image courtesy Wikimedia.