Ladies and gentlemen, I am just a caveman. Your modern scientists found me frozen in a snow bank. They thawed me out, brought me back to life, and then Curtis Yarvin gave me an Urbit galaxy.
I'm not intensely technical nor am I particularly business-savvy. But I thought I'd give you the shape of things as perceived by my primitive brain.
And anyway, Urbit doesn't really make sense from either a technical or business perspective. It only makes sense if you're a dreamer.
In the disillusionment of the post-dotcom-bubble, post-financial crisis, this project by this more-or-less retired blogger, was leaked. Something called "Urbit".
It was like the very last gasp of the very weird, extremely idealistic techno-libertarian era. The era that gave us open source software and Linux.
Everything about Urbit was different, not different for difference sake (with the possible exception of the Loobian type), but someone had picked up everything and thought about it. Every one of Chesterton's fences were gone with reasons attached. Real Year Zero vibes to it.
The project has been around for over a decade, so it's easy to lose track of how incredibly novel it all was at the time. And still is.
Code was written with punctuation marks that were given one syllable names, so you could easily read your code out loud (or subvocalize it). Versioning worked backwards toward perfection. 32-bit addresses you could pronounce. The corporate arm was named after a story by a South American author about multi-century conspiracy to create a new reality based upon the principles of Berkeleian idealism. This was the perfect name for the size of the ambition involved.
Notably, it attracted the sorts of people who were intrigued by this. Very smart, very odd people who wanted to build something huge and strange and improbable. The kinds of people who heard about a company called "Tlön" and immediately got it without anyone needing to explain it, or else were curious enough to go read a short story.
Of course it can't all be dreams. A software project needs to be written by software people, who will need actual currency, not dreams, to pay their very real rent. So at some point you do need money coming in. But for a hothouse flower like Urbit in its infancy, the amount of money was really "as little as we can get away with and still make payroll."
There have been many mistakes made. Spinning off the Urbit Foundation from Tlon needed to happen eventually. But it didn't need to happen when it did.
And Josh had many positive qualities. But he was not a dreamer.
It would take a special kind of person to turn down the money being shoved at Urbit during the peak NFT craze. Unfortunately Urbit had nobody willing to do that. This ended up being a (fatal?) disaster for Urbit. It attracted exactly the sorts of people the project didn't need. The kind who dreamed, but it was the wrong kind of dream. The kind who ended up pushing toward monetization that Urbit wasn't ready for, who then abandoned it when they decided their bags weren't getting pumped fast enough. And lots of good developers got lost in the shuffle.
(Really someone should have looked at the valuations for address space a few years ago and said, "This is too low for what it should be eventually, but way too high for what it is now". And then sold a bit and used the money to ride out the inevitable crash. But everyone figured if Bored Apes were getting that kind of valuation, surely Urbit address space, with an actual use, was priced correctly. Rather than saying, "Wait, no, everything is overpriced, including us." I assure you, as a caveman, I was not thinking that. I also had bags, the line was going up, and I wasn't complaining.)
So as this all pertains to the recent UF board dust-up, I am on the side of the dreamers. The people who are thinking, "What's going to put this project on the path to devour the world in ten to fifty years?"
I want to believe in Curtis. Almost every part of this weird project came from him mind. There's no bigger dream than to have said, twenty years ago, I think I'm going to recreate computing from the ground up.
But I don't think Curtis in 2025 has the same juice as the version of himself who invented a programming language that looked like dialup modem line noise and named it after a Wallace Stevens poem.
But, as mentioned previously, a software project needs software developers. Even if Curtis was exactly the same person he was when Urbit was leaked all those years ago, he has clearly lost the developers. Maybe they ought to believe in him, but they don't.
To the extent Curtis even writes code any more, he can't write it all. And I don't think Urbit can afford to lose many more developers.
The people on his side seem blasé about losing all the people who have been working on this for years. Maybe they're right on this. Maybe they can just go to JiffyCoder™ and get a dozen or so Hoon/C developers off the shelf to keep the lights on and move things forward. And maybe these devs will not also get tired of the way Curtis does things, as the current crop clearly has.
Or maybe not.
Or maybe Tlön will end up taking over core development, in which case what's the point of the Urbit Foundation anyway?
I have instead voted for the new board. I know these people. They are true OGs. There is no doubt they are motivated by doing what's best for the project. And, crucially, the have the backing of the developers.
I don't know if they have put much thought to what this project is going to look like in 50 years, but I am bearish on the project existing in 50 weeks if there's an acrimonious split that leads to the project getting forked. One Urbit is plenty please.
For Urbit to actually take over the world, at some point it needs to be not dreams, but something people actually use to do things. The ideal time for this to happen is when the core software is fully-baked enough to show what Martian computing can do that Terrestrial computing can't. It was not ready when third-parties first started building on it a few years ago, and after a lot of needless distraction it's still not there yet.
I haven't spent a lot of time applying my primitive brain to either of the projects the two boards have put forward. Maybe one or maybe the other will make the line go back up a bit more, but I don't think anyone's going to look at either of them and say, "Martian computing has arrived!"
But I know that Martian computing will never arrive if the project dies, and I think the new board has the best chance of keeping the dream alive.
The sad fact, and obvious opportunity, is that there still is nothing as ambitious with as much potential as Urbit out there.
Despite all the unpleasantness of the last year, despite the financial implosion that was narrowly avoided, the project is in way better shape than it was a few years ago. Someone estimated at Assembly 2021 that Urbit was running about the speed of a 56k baud modem, which felt about right at the time. It's massively improved over that.
I think the number of booted-up planets is past its peak. But this makes it easy to underrate how much life there is on the network. I don't think it's decline is irreversible. We can build on what is there now.
It just has to survive.