Help me build a different kind of platform cooperative
Published on 2025-08-15
In short: I have a vision for a platform cooperative—one that aims to serve regional needs, that empowers communities to own and govern their own data and online services along cooperative principles, served from a distributed data center built from improvised and recycled hardware. Will it work? I don't know, but I'd be remiss if I didn't try, and I can't try without your help. If you are a socially/ecologically minded person who wants to make this a reality (whether you have technical skills or don't), I'd love to hear from you.
What we crave is not trickle-down, faceless profits but reciprocal, face-to-face relationships, which are naturally abundant but made scarce by the anonymity of large-scale economies. We have the power to change that, to develop local, reciprocal economies that serve the community rather than undermine it.
(Robin Wall Kimmerer, The Serviceberry, p. 92)
I was a kid when I first stumbled into setting up a web server. I had no idea what I was doing at the time so it was practically dumb luck, but that didn't change the feeling of awe I felt when I went to school the next day, connected to the WiFi network, punched in my home IP address, dismissed the certificate warnings and saw… the default Nginx installation page. But that wasn't just any default Nginx installation page—that page was coming from my basement, and that was pretty cool.
It didn't last for long though; my parents had mixed feelings about me hosting a website on their network, and I couldn't figure out how to change the default page anyway. I ended up building a real website and uploading it to one of those sketchy, free web hosts that sort of just existed at the time. I think I even had a .tk domain name to go with it to really seal the deal. Then, my website was hosted from… wherever. It didn't really matter, or so I thought at the time.
For the longest time I wasn't really able to run a homelab because I didn't have the level of access to the network I'd have needed to set up port forwarding. Like, a few years back I was living in this place where my router wasn't even physically in my apartment (maybe I'd have been able to figure something out if I had a better relationship with my neighbours, though long-time readers of njms.ca will know that I did not, in fact, have a good relationship with my neighbours)
Empirically measurable problems
A year later and the next place I lived I did have direct access to the router, and I was able to (convince my roommate whose name was on the utility bill to let me) set up a real homelab where I served all my websites for a good two years, and I had that feeling again. There's something viscerally satisfying about running a website from your basement, something that goes beyond a ping time measured in nanoseconds. It serves to deconstruct the cyberlibertarian trope of the internet being this new world, or new frontier—revealing that it's a very real thing you can see and touch if it isn't being hidden from you.
A few months ago, I moved back into the University-ecosystem and now I can't even connect to the enterprise network, let alone run a web server out of my dorm.
How to connect to the UBC Secure wireless network on Linux
So, I made the hard decision to move all my websites onto my VPS in Oregon. But, if you've been paying attention to the esoteric fascist takeover of the United States, you may know what I mean when I say Oregon doesn't feel like the best long-term bet for a transsexual computer programmer (at least, not until it secedes with the rest of the Pacific North West to form an ecotopian society). So, I looked at Hetzner's global map of data centers to find one in a country that wasn't facing an incipient esoteric fascist takeover. Unfortunately, all of Hetzner's data centers are in countries facing incipient esoteric fascist takeovers, so I settled for Germany—the only one I'm aware of that actually has constitutional mechanisms to deal with esoteric fascists.
I've been thinking a lot about that decision. I don't feel good about it. It's not just a political thing, either. It's not just that my average round-trip time to my own website is now in the vicinity of 170ms (!!). It's that having a website doesn't feel real, anymore. I used to be able to physically walk to my server and see what it was doing. If I had an issue, I could carry my monitor and keyboard over, plug it in, and troubleshoot on-site. Now my server isn't "real," in the fullest sense. It's a "virtual" private server. It's more like the concept of a computer, sitting in a data center I'll never see, surrounded by people who don't even realize it exists. That bothers me a lot.
Even while I was running a homelab, I had a nagging feeling that running private digital services was wasteful. It was a lot of work for something that'd only ever benefit one person, and all the while I had a lot of residual capacity on my server that was just sitting idly. There's nothing wrong with self-hosting if you find it fun, but as a response to the social problem of an enclosed digital commons, my self-hosted web services were the suburban McMansion of the digital neighbourhood.
A lot of people have done the self-hosted suburbia discourse, though, so I'm not going to dredge all of it out in this article. But I will say that these two problems are related. They both deal with the alienation inherent to modern high technology. The geopolitics and spiritual rut of hyperscale data centers is a social problem, and the suburbanization of the internet through self-hosting is a technical solution. I think between these two phenomena a better world is possible.
I've already described my vision for a better digital ecosystem:
The People's Data Centre of Sparwood, BC
I want to see hand-made data centers spring up in communities around the world, administrated by volunteers, governed democratically by the communities they serve, operating within limits. Such a data center couldn't run Netflix or Facebook, but a network of recycled computers responsible for simpler tasks and a handful mid- to high-range consumer PCs as required for more complex tasks could absolutely meed the needs of a large town, probably even a big city. You could think of it as a confederation of homelabs forming a decentralized data center, designed to meet local needs.
This is my dream. I've already written about this, though. I've even written about why I fear it'll never happen:
Nobody cares about independent digital infrastructure
I don't doubt there's at least a few people in Sparwood's population of 4000 who'd be interested in this sort of thing, but I haven't found them yet. That basically stops this project dead in its tracks. But the fact that those last two articles seem to be the most well-received articles I've ever written tells me that those people do exist somewhere, and you might be one of them.
So here's my proposal: let's try anyway.
Let's say I live in Sparwood and you live in Toronto. Maybe we can't build the People's Data Center of Sparwood (yet), but we could build a platform cooperative that serves a much larger region—say, North America. The goal is to build networks that serve small communities, but that's a big project and we can start small by covering a big region. Maybe, after we've found our footing and attracted some attention to the project, we find that there's people who want to get involved from Detroit, Cleveland and Thunder Bay. All of a sudden, the world's gotten a lot smaller. Maybe I hear from some people in Portland, Seattle and Vancouver. Here, the project splits in two: one for the Great Lakes, and another for Cascadia. What's left is a confederation of major regional digital ecosystems. The projects share fundamental principles, a code of ethics, and resources and labour as required, but they're autonomous and can focus on serving the needs of their local community.
The goal is to provide a platform to nurture smaller communities, to enable people to administer and govern their own digital community spaces—always striving to distribute more autonomy to local communities, until ultimately every community can provide for itself.
Again, this is a big project, and Sparwood may never literally have its own community-owned data center, but that was always just an ideal. This project being successful on literally any level would be better than what we already have. Even if it's a total flop, maybe it'll inspire someone else to do something better.
So, I guess you could consider this article an invitation. If you share the vision, see the bottom of this page for ways to get in touch; I'd love to hear your ideas. This sort of thing takes labour and resources, and those are going to come from people willing to volunteer their time. I do want to stress, however, that this sort of thing takes more than technical skills. While we would need people who know how to do system administration, we'd also need artists, community facilitators, people to do outreach, and so on. Even if you don't know exactly how you'd fit into a project like this, if you share the vision, there's a place for you.
Respond to this article
If you have thoughts you'd like to share, send me an email!