Urbit Isn’t Alright
Three Things #180: July 27, 2025
It’s been quite a while since I wrote about Urbit. I haven’t followed the project closely over the past year, partly because I’ve been busy with other things, and partly because everything I’ve heard about Urbit recently has been ominous. Last I knew, there was a coup about a year ago, when the board of the Urbit Foundation fired the Executive Director, then most of the staff quit in protest. Apparently, since then, there have been no less than two additional coups, and we’re on the verge of a third one happening. As Curtis Yarvin, the project’s progenitor and current Executive Director-by-fiat said in a recent diatribe discussing current events, these are “real Roman times” for Urbit. This is Urbit’s “year of four emperors.”
I’ve always found Urbit absolutely fascinating, technically as well as socially, a topic I’ve covered here before. I’m friends with a number of the stakeholders, and I like them quite a bit. I’ve met not a small number of interesting, intelligent, motivated people through the Urbit project and community, and it’s one reason I’ve continued to stick around in spite of its dysfunction and general failure to find product market fit or deliver a product that anyone outside the core community seems to care about. I own some Urbit address space, although I’ve always thought of it less as a financial investment and more for novelty and sentimental value. Thus, while Urbit fascinates me and while I feel somewhat socially invested in it, I’ve never been invested to the degree that I was with Ethereum, or Spacemesh, or am today with NEAR and House of Stake.
Nevertheless, I couldn’t help but get caught up lately due to all of this drama, which by the time of you read this should’ve culminated in a galactic vote (ultimate governing authority in Urbit rests with a group of 255 NFTs known as the Galactic Senate) to oust the current Foundation leadership. I read a slew of proposals, counter-proposals, arguments, and recaps of everything that’s going on (this one, in particular, is reasonably complete and accurate AFAICT). As I was getting caught up, I couldn’t help but feel that there are lessons for other DAOs and other open source projects, including NEAR and House of Stake, in the drama that Urbit is currently experiencing. Here are some of those lessons.
Thing #1: Governance Theater 🎭
DAOs have a failure mode that I call governance theater. This is the biggest thing I dislike about DAOs, and about online governance in general, and it’s the number one way I’ve seen these projects fail. Paul Graham refers to a malaise that befalls startups that he calls “playing house.” Governance theater is quite similar to this, and it’s also quite similar to the well-known phenomenon of bikeshedding.
It works something like this: all the DAO participants take their seat at the governance table, metaphorically speaking. Everyone is really excited about governance. We feel that we’re going to do great things together! The governance is going to be awesome. It’s going to be yuge. We’re going to solve all of the governance problems that no one else has been able to solve, because of this or that different approach, this or that innovation, or just because we’re smarter than everyone else.
But before we can get down to the actual business of governance, there are just a few things that we need to work out. First of all, of course, we need to figure out everyone’s role. There has to be someone in charge. We need some committees, and some councils, and maybe an advisory board, and some working groups. For accountability, we need a governing board. We need to recruit people for all of these bodies. We need procedures, and house rules, and of course, mission, vision, and values—all of the things I wrote about recently under the umbrella term North Star.
Then there’s the issue of compensation. Who deserves to be compensated? How much? For what work? Do we compensate people retroactively? Where do the funds come from? How do we manage the budget? Do we adjust based on cost of living and geography? And how are these decisions governed? Do the governors themselves get to compensate themselves? What’s the accountability mechanism there?
We also need some defense mechanisms, of course. We need a security council, with the ability to block proposals, or revert changes, or force through protocol upgrades for security purposes. We need the right set of people on this council, of course, because their role is super duper important. We need delay periods and withdrawal mechanisms if things go really wrong.
The list goes on, and on, and on. The DAO can spend years, and millions of dollars, figuring out how to govern itself, without actually doing any governing! For those who haven’t spent much time in DAOs, this might seem completely absurd: imagine a corporate board making these sorts of decisions in the absence of an actual company with an actual, profitable product. But I’ve seen this happen again, and again, and again in the crypto-governance world.
It happens for two reasons. One, this work is usually funded by crypto foundations who have deep pockets and little to no accountability to anyone, which is a problem: something I’ve written and spoken about in the context of Ethereum. Two, being a real crypto project requires decentralization, and to be credibly decentralized, you do need to be thinking about decentralization from the very beginning. You do need decentralized governance for a project or a protocol to be meaningfully, credibly decentralized: hence the act of putting the governance cart before the product horse.
The issue in nearly all of these cases is that the protocol or project in question doesn’t have product-market fit (PMF) (more on which in a moment). This means that governance feels more preparatory and less urgent, since, without PMF, you don’t have a lot of users or that much at stake. And it also means that you’ll struggle to convince people to participate. Surprise surprise, it turns out that the set of people willing to shout at each other online over trivial decisions for very little money about a project no one else cares about is in fact tiny (more on this in a moment too).
The antidote is pretty simple: invest less in governance and more in building. And I say this as someone who does crypto governance more or less full time! Quite frankly, attempting to set up robust governance for a project that hasn’t found PMF is like shopping for baby clothes together on a first date. It’s wildly premature. And, once the business of governance theater has begun, it’s woefully easy to get lost in the noise and lose sight of this fact. It’s all too easy to forget what you’re supposed to be governing in the first place.
It’s good to think about governance early and often, but it’s a mistake to try to install decentralized governance on top of something the world doesn’t know or care about yet. DAOs don’t build good products; visionary leaders and small, focused, centralized teams do. Decision making should be centralized at this stage.
What does this mean for projects like Urbit that, nevertheless, insist on leaning into decentralization from the get go? I’m still figuring this out, but I think the right approach here is something like a bunch of little fiefdoms, each of which is a product organization focused on a particular product vision and particular execution path, but all of which nevertheless agree on a basic set of standards (i.e., a protocol). Yes, you need to govern the protocol itself, but this can be very straightforward to begin with and we at least know how to govern protocols in a reasonably functional, decentralized fashion. Each little fiefdom is free to pursue its own source of funding, its own product vision, its own go to market strategy, etc.
What does this mean for NEAR and for House of Stake? NEAR has more PMF than Urbit, but less than Apple. Is this the right time to think about decentralized governance at NEAR? I don’t think NEAR has found perfect, complete PMF, but some products have begun to find it. Which means that it’s probably the right time to begin thinking seriously about, and working seriously on, decentralized governance. At the same time, however, we still have a lot of product work cut out for us at NEAR. This is one of our core challenges at HoS: balancing product work against governance work. I reconcile these two ideas in my own head by thinking of HoS itself as a product. I’d like to write more about this topic later, so stay tuned.
Thing #2: Traction 🎢
In his diatribe, Curtis evokes yet another wonderfully simple, useful concept attributed to Paul Graham: a project can either be “default dead” or “default alive.” Urbit is “default dead.” All this means is that it hasn’t yet found PMF, it isn’t profitable, and its runway is limited. Without either additional VC funding or revenue greater than expenditures, it’ll die.
In fact, Urbit almost died recently. There are probably a few times it almost died, but most recently the Urbit Foundation ran out of money last year, which is why the Executive Director at the time was fired. Earlier this year, Curtis convinced a16z to invest $4M in exchange for more address space (the firm is a previous investor), in spite of the fact that the address space market is as dead as the rest of the NFT market. (I don’t know anything about the details of this conversation or the terms of the deal, but I suspect that a16z was investing less in Urbit the technology and more in Curtis’s network and star power, which has been on the rise in the Trump/Vance era.)
I think that Curtis’s assessment that Urbit is “default dead” is correct. In fact, it’s more than correct: it’s also the heart of the issue. Urbit has been around for more than 20 years. It has raised money a bunch of times, and, after all those years and all of that money, it still isn’t profitable: as a company, as a Foundation, or as a network. It still hasn’t shipped a product, or even infrastructure, that the world cares about or that people are willing to spend money on.
At the end of the day, Urbit is infrastructure, and take it from me as someone who’s been building infrastructure for nearly ten straight years: infrastructure is hard. It’s hard to build, it’s hard to explain, it’s hard to sell, it’s hard to monetize.
For an infrastructure project, there are typically two paths to sustainability and profitability, i.e., to switching from “default dead” to “default alive.” The first involves selling to developers, either individual devs (think: Vercel) or else companies (think: Oracle). Urbit, like Linux, doesn’t really have anything to sell, per se. There’s no product. So the alternative, like Linux, is instead to sell memberships, which really amount to a seat at the table in governance. Thousands of companies, large and small, rely on Linux and are happy to each pay a few thousand dollars a year to support the foundation that supports this project. (The Linux Foundation is probably the oldest, simplest, and most well known open source governance model, but there are many others.)
Note that the Linux Foundation does not directly maintain or market products to those companies. Instead, a global community of independent, hobbyist developers, mostly employees of member organizations or unpaid volunteers, builds and maintains those products. The products—the Linux Foundation manages dozens in addition to Linux itself, such as Node.js, PyTorch, and RISC-V—are fantastically useful, and they’re free to use.
Here, again, Urbit has a problem. Urbit is infrastructure that you can build useful things on top of, like Linux. But there are scant useful things built on top of Urbit today. I could fill a lot of space here speculating on why that’s the case, but my simple diagnosis is that Urbit doesn’t have great products today because the Urbit community has not, historically, focused on building great products. Instead, it’s been focused on building infrastructure, because Urbit infrastructure is still relatively immature. It’s slow, it’s limited in terms of its capabilities, and it’s difficult to build on. (Tlon Messenger is the one obvious exception. It’s pretty good, but it’s not on par with Whatsapp or Discord just yet.)
Which brings us back to the original quandary: Urbit isn’t going to be profitable anytime soon, which means it’s going to have to continue to raise money for the foreseeable future. But as Curtis points out, in order to do this, it’s going to need to demonstrate “traction,” in the form of those acronyms that investors love such as MAUs. That’s going to require a product strategy, even if Urbit isn’t quite ready for the big leagues yet.
I actually do think this is feasible. Urbit isn’t quite there yet, true, but there are ways to “fake the backend” on the parts that aren’t quite there yet. The real question is what’s the right application to build, and why: why is Urbit the right platform for it, why can’t it be built elsewhere, and why are Urbiters the right people to do it? Curtis answers the first question: he wants to build a private, anonymous social network for elites called Moses (see the above linked diatribe for his description). He sort of answers the second question too, if not the third.
Would this work? I have no idea. If anyone but Curtis Yarvin were pitching this idea, I’d probably laugh. He might just be able to pull it off. It’s probably worth trying. (Note: the opposing camp has a competing vision called Neoscape. It’s kinda cool too, but it’s more of an operating system and less of an app, which is harder to sell.)
What does this mean for NEAR, and for HoS? Unlike Urbit, NEAR is “default alive,” at least for the foreseeable future. This is true not because of PMF or profitability per se, but for the same reason that other large crypto communities and foundations are “default alive”: a large war chest, a well-managed treasury, and the ability to mint money, something Urbit hasn’t figured out how to do yet. But, actually, we shouldn’t rest on our laurels. Like all cryptocurrencies, $NEAR has experienced a lot of volatility, and the NEAR Foundation and House of Stake treasuries might not last forever. That’s why it’s so important that we, too, invest in product, in finding PMF, and in profitability sooner rather than later.
Thing #3: Professional Leadership 💼
Something I’ve noticed about crypto projects since joining the Ethereum Foundation in 2017, and something I’ve never totally understood, is their allergy to professionalism. Or perhaps it’s the other way around: maybe professionals are turned off by crypto, so crypto projects are forced to make do with what they can, which often means inexperienced people.
This was actually most severe at the Ethereum Foundation. As far as I could tell, when I was there there weren’t really any “professionals,” at least in the way that we usually use that term. There was no one who had run a foundation, or even a large team before. The people handling marketing weren’t experienced marketers, and the people in charge of events didn’t have prior experience running events. The software developers knew how to write code, of course, but many of them were “basement hackers” that had never done proper software engineering or shipped production software before. (This explains a lot about the dysfunction at the Ethereum Foundation.)
I’ve seen the same thing at multiple other projects and in multiple communities since then. It’s hard to put your finger on precisely why this is the case, as it’s due to a unique combination of factors. First and foremost there’s the contrarian nature of the space and its progenitors. The cypherpunks deeply distrusted institutions, and many crypto people today take this to include things like universities, consulting companies, auditors, lawyers, financial institutions, and big tech companies. There’s a feeling that professionals are, by definition, drinking the mainstream kool-aid and are therefore unable to think outside the box or come up with the sort of creative solutions that we’re famous for, and indeed that crypto requires.
This also results in a perverse form of “anti-credentialism,” where having a big school or a famous company on your resume not only doesn’t help you, in fact it actively works against you. To be clear, this isn’t speculation: I saw multiple candidates with stellar resumes rejected from places like EF on the grounds that, “We don’t hire people from [Harvard/McKinsey/JPMorgan], they’re arrogant and they’re too expensive.”
While I share the cypherpunks’ mistrust of institutions, including universities and many big companies, I personally think this behavior is silly and immature. Just because Harvard and Google tend to attract mainstream thinkers and do engage in woke brainwashing doesn’t mean that there aren’t talented, contrarian thinkers and builders in these institutions.
The reality is that the founders of many of these projects (professor coins aside) are themselves outsiders and dropouts, as exemplified by Vitalik himself. In many cases they’re people who could not, or would not, take traditional jobs at traditional companies. They tend to hire people like themselves. But I believe there’s more to the story: also a naivete and lack of self confidence. I think this sort of founder feels uncomfortable hiring people who are older or more experienced than they are, or who have better credentials.
Finally, there’s also just a youthful, rebellious, overconfidence at work. Crypto people tend to feel that working within the existing system is a waste of time, which is why we’re instead building a parallel, better alternative. There’s some truth to this, but taken to the extreme it leads to a strong, unfortunate tendency to throw the talent and experience baby out with the trad bathwater. In other words, crypto folks like to think and build from scratch; what we often refer to as “first principles thinking.” I like to call this “blank slatism”: a naive belief that you can jettison and effectively ignore the lessons of the past because “this time it’s different” and “we’re just smarter.”
This time isn’t different, and we aren’t smarter in the ways that matter, 99.9% of the time. That’s the lesson of history. And to feel otherwise is hubris plain and simple. This is, incidentally, why crypto has been speedrunning the lessons of traditional finance and making so many of the same mistakes. It’s the biggest risk the industry faces, in my opinion, and it’s entirely self-inflicted.
Urbit suffers from these same problems. It’s trying to do too many things at once: reinventing, as it does, the entire application stack, rather than fixing things one problem at a time. Of course, the entire Urbit thesis is that you can’t fix things in piecemeal fashion. You have to start at the bottom and reimagine everything. I’m not completely sold on this idea, but I also don’t think it’s crazy.
But, to get there, you still need professionals. You need people who are world class at what they do. You need people who have led big, successful teams before. People who have raised money, built products, hired teams of A players, and sold companies. Urbit has dreamers, motivated people, good people, smart people, hard working people. But beyond the core technology it has scant few professionals. And I think it’s going to keep going in circles until it manages to attract some real professionals to help turn it around.
This will be hard for a bunch of reasons. To call it straight, these include: being crypto adjacent; being nearly impossible to describe or explain; Urbit’s reputation as a half built, irrelevant project; Curtis’s reputation; and a lack of funds to pay for top talent, which doesn’t come cheap. But worse projects with less money have recruited better people. It’ll take a founder’s passion and resource magnetism to find the right person or people, the right new leadership to lead a transition, but I firmly believe that person (or group) is out there. We should all be trying to help recruit them, by any means necessary.
What about NEAR? I can say with confidence that fortunately NEAR Foundation does have a lot of professionals on staff. It’s refreshing. It’s by far the most functional, professional team I’ve been a part of in this industry (although the bar is admittedly, depressingly quite low). But the broader ecosystem does lack both talent and professionalism, and we’ve had trouble attracting both. This is something I think about constantly in the context of House of Stake. HoS is supposed to contribute to ecosystem growth. If one of the main things holding the ecosystem back is talent, then attracting talent and closing that gap should be one of its main goals.
But, as with Urbit, there’s a bigger issue at play here. The best people in the world want to work on projects that are interesting and challenging, but also growing, and socially valuable. They want to build products people love using. They want to be part of teams that are values-aligned and mission-driven. Crypto is not perceived as such today. To the extent that there is growth, it’s mostly “intrinsic”: meme coins and casinos, which destroy social value.
Of course, there are exceptions, most obviously stablecoins, which have already begun to change the world in a positive way. I think NEAR, and perhaps Urbit, could do worse than doubling down on the few interesting, exciting, high potential, socially valuable use cases that we have found, such as building more and better stablecoins. That’s a mission that could attract talent and strong leadership.
This is just one random example, and it’s not immediately obvious how it fits into the Urbit roadmap, but the Urbit community is smart and resourceful. I’m certain that, if they put their minds to it, they could answer this question or come up with other, better ideas. I’m not holding my breath, but nor am I ready to give up on Urbit just yet.