Blog Urbit is at a crossroads
Jeremy Tunnell at
I am not going to sugar coat it: Urbit has had a tough few years of governance.
To make a heroic effort to summarize the last few years (dates are approximate):
- Curtis Yarvin, the founder of Urbit, left the project and divested of his galaxy holdings sometime around 2018. He still doesn't have any voting control over Urbit.
- The Urbit Foundation (UF) has been under the leadership of Josh Lehman since it's founding around 2021. Josh was appointed by Galen, who is the CEO of Tlon corporation which is the original custodian of Urbit.
- Urbit adoption and progress stalled about two years into Josh's tenure. The strategy of grand partnerships with external venture funded companies fell through, apparently because Urbit was not able to deliver the feature set needed and also because Josh mismanaged the relationships.
- As the UF stalled, the board continued to back Josh's turnaround plan that never actually worked.
- In late 2024, on a Friday, the board discovered suddenly that the UF was out of money. Josh's financial projections apparently missed some really important things, and He did not realize the dire financial situation of the foundation. Over the weekend, the board fired Josh, and his response was that on Monday he was going to lay off everyone and quit anyway.
- To the surprise of the Galaxy holders, the board immediately and without consultation, appointed Curtis to come in and execute a turnaround plan for the UF, and also Curtis brought on an ED named Topher. At the same time, one of the board members, Ryan Kneer, chipped in to oversee this turnaround.
- Ryan and Topher didn't get along, and Topher left after not receiving a contract, amid, I'm told, a request for an outrageous salary.
- In early 2025, the UF raised $4M in investment from Andreesen Horowitz, greatly, but not exclusively, because of Curtis's personal relationships there.
- In early 2025 Ryan and Curtis worked together to execute Curtis' turnaround plan, and they didn't get along either. Nor did Curtis get along with the new CTO at the UF, who resigned. Complaints about Curtis's leadership basically amounted to him being absent, distracted, and not a very good executive or manager.
- In June 2025, Ryan fired Curtis and took control of the turnaround plan.
- About a week later, the board fired Ryan, and brought back Curtis, and also hired someone to act as Curtis' COO.
- Ryan, in response, has submitted a motion to the Senate to recall the board and replace them with five well known developers and community members. It is assumed but not written into the motion that the board will then reappoint Ryan as Executive Director.
Urbit is at a major crossroads. This vote will decide which direction Urbit goes in, both technically and culturally. This is a call to all Galaxy owners to please educate yourself about what is happening and vote before the deadline on July 26.
Now, to be clear, I am not a journalist, and I am not unbiased. I am a Galaxy holder since 2017, and I have mostly opposed the current board and specifically Josh and Curtis' leadership.
Below is my argument that Galaxy holders should vote FOR the proposal, and while it will not be unbiased, I will attempt to be fair to all parties.
Here are the things that I believe to be true, and I think even those on both sides might grudgingly admit are true.
- The current Urbit Foundation board has been nothing but a complete failure of governance and oversight. They stood behind Josh when they shouldn't have, they failed in their oversight and were surprised that the UF ran out of money, they failed to communicate to the Galaxy holders about what was happening, they executed what so far has been a very rough hail mary turnaround plan, and generally, they seem to keep making wild and surprising decisions without consultation with any of the stakeholders.
- At least one of the board members has expressed his desire not to be on the board anymore.
- Nobody really likes Curtis or wants to work for him. He doesn't get along with Ryan; he didn't get along with the CTO, both current and past; a majority of the employees of the UF don't want to work for him. Nor does he get along with Galen at Tlon, or the vast majority of Galaxy holders.
- Curtis doesn't want to work at the UF full time, and he doesn't want to do any development on Urbit anymore. He's quite busy, to put it mildly, being MAGA's house philosopher and attending dinner parties and salons and doing interviews.
- Urbit isn't really ready for prime time, and what is really needed is some really large architectural changes to enable true app development, but a lot of people are afraid there aren't enough resources to get to the end of that development, So therefore there are a couple different alternate stopgap plans.
So here we are, today, trying to fix the board situation and choosing between two of these plans.
Ryan's plan is called Neoscape, and it proposes to continue working on the core improvements that are needed to get Urbit to primetime, but also as a stopgap, work on a "product" that features lots of blockchain integration. It also proposes the release of an $Urbit coin on the Hyperliquid exchange, with this blockchain product targeting its users. You can see the presentation for yourself, but I think the plan is plausible and fairly low risk and at least continues to fund development on core.
Curtis' proposal is to essentially put the core work to the side for a while and focus on being a "startup" with a "product". His plan is to create various applications leveraging the utility of the Urbit namespace, separate from the Urbit server. For example, he wants to create an anonymous chat application where, he imagines, Venture Capitalists (Presumably leveraging his social connections) would all gather to speak freely behind Urbit's anonymous identities. He also wants to launch an $Urbit coin, but on a centralized exchange (Presumably leveraging his social connections).
Before I comment on Curtis's proposal, I need to make it clear that he's a self styled ideas guy, and he has an idea for every moment of the day. Lots of ideas. I don't actually know what his specific proposal and plan is...there are just ideas scattered between hours long youtube diatribes and 10,000 word manifestos posted in various places.
In my opinion, Curtis' proposal is, at minimum, highly speculative, and reliant on the assurances of a guy with a massive ego that he's really well networked and that his network is just going to eat all this up. At worst it's totally delusional, for example the idea that venture capitalists are going to say anything controversial on a beta platform with assurances that it will forever remain anonymous. Or perhaps also delusional is Curtis' belief that his followers and buddies who follow him are highly motivated to engage with a platform that has nothing to do with his philosophy or, dare I say, his political shitposts.
He also wants to put core development into a separate organization that is not the UF, which feels pretty backwards to me.
I don't buy it. Again, I invite you to peruse his plan; if you can find it. Here is an excerpt from a recent post from Curtis:
The solution, in my opinion, is for the core devs to create a new organization, which the UF should fund. This new organization should make all of its own technical decisions. Money creates power, but the strings between the UF and the core guild should be as loose as possible. But money is power and has to be managed. The correct way to do this, for now, is to pick one person, employed by the UF, trusted by the core guild, and let him manage the flow. Indeed, it is the mission of the UF to fund core development. The mission is NOT traction. Traction is only a way to fund core development. It is only a means to an end. Unfortunately, at least in my opinion, it is the only such means—so we need to hit it as hard as possible. Core development needs to be funded. But it can only be funded by (a) raising money and (b) finding product-market fit. If the core devs run the UF, they will run it into the ground—as they did last time, because they are not startup guys and the UF has to be a startup. They need to eat. But they cannot be in charge of killing what they eat, for a simple reason—they just aren’t killers. Leave the jungle to the jungle creatures. Actually I am only about 40% a jungle creature. I know how terrifying the real ones are. This is a fundamental issue of political science—not office politics, but politics. A startup is a monarchy. The CEO does not have to earn his title every day. His legitimacy is formal, and absolute—subject to the accountability of the board. He just has to win. An open-source project is an aristocracy. Rank is informal, and earned—it is prestige. Prestige comes from contributions. Old contributions don’t really count. No one can rest on their laurels. In the special cases in which the project is a monarchy (looking at you, Linus Torvalds), this is the exception that proves the rules. The king is also a lord—he is only first among equals. And he earns this position every day. Perhaps the developers would have accepted me if my way of being a monarch was like Linus’s—to also earn my rank constantly, through contributions. Actually I haven’t written a line of code since I came back. There are lots of reasons for this, but it’s how it is. An involved boss might be tolerated—a distant boss, just by existing, is demonstrating a lack of respect. A distant boss is “sea gulling”—he flies in, shits all over everything, and flies out. For a technical aristocrat, it’s unendurable. And Urbit needs technical aristocrats. My very existence in this project is a bad smell to all the best people. It damages their sense of accomplishment, which is vital to the whole economy of open source. In fact, of all the reasons I left in 2018, this was the main one. That strategy worked even better than I expected. Perhaps it worked too well. Some have suggested that the UF should not be run like a startup—rather, it should spin off a startup. This is true, probably, in the long run. In the short run, we just don’t need this level of structural complexity. Everything is default dead. Once Moses gets some traction and feels more real—or once it fails, but the next traction experiment succeeds—it should be its own company. And these spinoffs shouldn’t result in duplicate effort. We had this spinoff idea—and pulled back from it when it became apparent that the first result of the spinoff would be for one pool of money to be used to develop two different Urbit front ends. Because the core devs are not jungle creatures, they should not play the jungle game. They should not be finding end-user product-market fit. They should be building the Arvo stack and the software needed to run it. They should have funding GUARANTEES and grant discretion. They should not be building front ends. The business people should feed them, ask them politely for certain features they need, and be ready to build and contribute those features themselves if no one in core thinks they’re needed. If there is another front-end product-market fit project—not a problem. Urbit is a free country. It should raise its own funding. It should also reuse as much of the same technical infrastructure as possible, so that users are not confused. Onboarding flow, social strategy, the claims contract, are things that cannot be done unilaterally if there are two competitors in this area. Tlon is different enough that confusion is not much of a problem. Two different interpretations of Butyl on two different platforms is.
My pitch for your votes
What I want for Urbit is for the talented devs to work on Urbit-as-a-server....toward the whole point of the project. I want a competent board, good governance, UF employees who are not there for a nice paycheck or checked out. I want an end to the drama. I want the project to finally be free of Curtis' political philosophy, that I might add has contributed to the drama that's happening right now.
I want Urbit to focus on developing a personal server that enables you to take your data and privacy back from the big corporations. I want Urbit to aim steadily at delivering a delightful developer experience. I don't want to grow or raise money with gimmicks, I want to grow and raise money by actually delivering the utility that we've promised.
Voting FOR this proposal will replace the failure of a board with a new one and enlarge it. It will make governance clearly better. It is also a vote to break with Curtis for good and set a course for Urbit free of the drama and politics that are attached to him. Voting FOR this proposal is a vote to seriously consider Ryan's plan, but the new board could decide to go in a different direction.
Voting AGAINST this proposal is voting against most of the developers, most of the community, and instead to install an ED who few want to work with, is steeped in drama, cannot or will not communicate effectively, and who has admitted he doesn't even want to be here. It's also a vote to keep a board that has demonstrated its incompetence over and over and also has expressed a desire not to be there and an unwillingness or inability to communicate effectively.
Further, as of this writing, the proposal is stalled at 19 for/19 against, with against made up of mostly a16z votes, current board members, and Curtis' minions.