I'm with you. Was recently reviewing Urbit again and I think I'm starting to understand it more. You may be interested in it as well.
I'm really fond of their PKI or Public Key Infrastructure
Discussion
I looked at urbit a few years ago. Definitely in this list. The complexity and immaturity of applications is what sent me to nostr. But I haven't written it off.
Urbit interests me because it is the only tech project ever made that has simplicity as part of the core design. Over time, it will only get more simple as well.
It’s just a fundamentally different ethos from anything else out there. Difficult to compare
I guess complexity was the wrong word, because you're right, every layer is very simple. Maybe complicated? But the thing that really turned me off was not knowing if the purely functional paradigm could remain performant at a low level. Evidently not, because they bail out to C code as an optimization. Like any foundationalist paradigm, if a single stone crumbles it casts the rest of the project into doubt.
"the pure function" always comes at the cost of infinite copying of data
just watch a haskell compilation sometimes with a system monitor open to see just how much memory it uses (and then discards)
this is why #golang uses a lot of functional concepts but discards forcing you to avoid pointers, and it's a delicate compromise because pure functions are without race conditions and this is another goal of Go - enabling concurrent processing
for good reason if certain operations happen concurrently on a single piece of data the go runtime kills everything and i am glad they make it that way because races are the most elusive bugs there is
Urbit is a totally new Operating System. So naturally, it will be one step forward, two steps back in terms of performance.
Having said that though:
- Nock can be run on bare metal. The only reason we use C today is bc it’s not running on bare metal. So they don’t “bail out” to C for performance, they just use C as the translation from Nock to machine code if that makes sense. https://x.com/rovnys/status/1766183255097905157?s=46
- They have recently achieved 1 GB / second read speeds of data https://x.com/hastuc_dibtux/status/1765075518637154374?s=46
A stone crumbling is not possible in this project because of the way it was designed. They designed the whole system together as a whole and then worried about performance later. Over time, one team has found a slightly more performant way to design the system, but they are still following the same design. It’s called Plunder. https://x.com/sol_plunder/status/1704554384930058537?s=46
The ethos comment is key. Hard to describe but Urbit is a calmer place to be. Feels more human friendly. Made because things are harder there it’s built in proof of work? It’s a long term project and worthy of the effort. It’s a plant trees for the next generation situation.
*maybe (oh you can delete are fix spelling errors there to….lol)
I stopped using umbrel's containers when they made urbit available (I update manually now).
Urbit seems like shitcoin garbage
At least it's modular shitcoin garbage 👍
Urbit has a native bitcoin wallet. Repeat: a team made a bitcoin wallet with Hoon. And soon an urbit native lightning wallet will be released as well.
The goal is to move the address space off of ethereum and onto an urbit native chain as well. NFTs have very few use cases, but it turns out that one of them is digital land deeds.
What the fuck is a digital land deed?
There is a limited amount of address space in urbit. So it is analogous to land in that sense. You just need something permanent that says this address belongs to this person. There are ways that they can move off of ethereum and use something natively urbit for that. Ethereum was just used to bootstrap it
You're retarded 🤷
There is absolutely no limit to the address space in these shitcoin projects. Call them digital land deeds, pets, or bananas or whatever, put them on whatever chain (including Bitcoin) doesn't change anything.
Absolutely nothing interesting about urbit 😤 Arbitrary ownership schemes have little to do with Bitcoin, this repeated grift has been done many times
Urbit is different from nfts, have you read their docs?
I have, it's garbage written for non-technical people who likely cant differentiate between hosted cloud services vs virtual machines 😅 the grift takes the confusion and expands upon it.
"Stars", "Planets" and "Galaxies" are basically used to describe paying for someone's Amazon instance running a repo filled with random selections from the Awesome-Selfhosted GitHub, then host that (in most cases, free) software in a VM one must pay for with KYC tokens (etheiruym)🫠
Tldr; it's an attempt to re-sell shitty cloud infrastructure using space terms
Not correct. Easy for you to boot up a comet and explore the ecosystem. Reading and doing aren’t the same. Urbit is a long term project and a worthy one.
Then why did they build it on ethereium?🫠
Also not a fan of ethereium either, I get it. I’m not a dev, but practically speaking… it is only used to obtain and reboot(if needed) your unique id. Using L2 you dont even need a wallet. Otherwise you don’t have interact with it. Not an expert on the history, but my understanding is when Urbit was built 10ish years ago it was the best blockchain tool available at the time. It was a pragmatic choice. And for whatever flaws of that blockchain choice, at the community level it does seem to practically help prevent spam and bot problems. Thus, P2P support and trust appears much higher than here (in my several weeks of nostr exposure anyway). I’m here to learn about nostr. There are some advantages here, but my main suggestion if one cares about learnings in the P2P space it would leave a gap in knowledge to gloss over Urbit casually.
nostr isn't p2p, it is just clients and relays. Consider checking out Keet rather than wasting your time (and money) with Urbit, its a p2p messaging/conference application that leverages DHT to make connections, no tokens necessary.
Generally ETH L2s make no sense because ethyrium's entire existence comes from the idea that bitcoin can't scale in layers due to its rigid limitations therefore it was determined that other "chains" to support applications and other usecases must be needed. The term "ETH L2" is an admission that ETH is pointless, yet another half-ass attempt at a narrative shift. It would do you well to avoid anything that uses nonsense jargon like that.
It's not built on Ethereum. They're just using it to associate a Galaxy, Star, and Planet to a unique public key (in this case an Ethereum address). Those aren't just "tiers of cloud hosting," they're uniquely identifiable points that facilitates P2P comms (peer discovery and routing) inside the network.
It's annoying that they've built their PKI on Ethereum. I don't like it as well. If it's possible to move it on Bitcoin I'll be one of the first people to advocate for it.
Urbit is a worthy piece of tech in itself and it's merits shouldn't be understated just because its public key infrastructure is on Ethereum.
Urbit massively simplifies the entire dance of spinning up a VM, installing n+1 packages, and running n+1 Docker containers. It's literally a VM that you can run anywhere that has its entire communications and publishing stack built-in.
You should really read the overview before saying anything about it. It just makes you look massively uninformed.
🙂↔️
I think it would benefit you if you read a little bit more about it. 🙂