liam-fitzgerald/shrub.txt

## shrub.txt
Urbit, as it lives and breathes

Segment 1: Bootstrapping
Urbit should be the easiest place to write any application. Excluding applications that require high performance, shrubbery already makes this very close to true. It needs a little more love and sanding down rough edges, but it’s extremely close
Result: The namespace is alive

Segment 2a: The “maximalism” in namespace maximalism
The most annoying part of working with Urbit is getting data in and out of the system. The solution to this is simple: Put everything inside the system. If all the data you could possibly want is bound somewhere, then a significant number of things we currently consider “applications” simply become transformations over the namespace. Moreover, this opens up the possibility of no-code shrubs because if the data already exists, then the necessity of marshaling shenanigans is a lot lower. We should leverage the network topology and provide stars incentives to republish data from the outside into their namespace.
Metabolized: Zapier, IFTTT, bloomberg terminal etc.

Segment 2b: Thinking with names
Concurrently with 2a, AI efforts should be ramped up. See: vector database overlay. This allows for actually useful, personalized AI efforts. More importantly, with aggressive use of finetuning, so that the AI becomes accustomed to the user, which should markedly improve sense-made per watt. If the data formats are universal, then urbit AI doesn’t need to be all that intelligent to be a better experience than ChatGPT/claude or whatever. Intelligence = compute + context. We won’t win on compute (nobody but openAI will) but we can win on context.
Metabolized: Every AI wrapper company

Segment 3b: Markets
Now that we have the world’s data inside the namespace, and cognition chomping at the bit to digest it, we can start to build out markets. Shrubbery has been designed so that primitives will coalesce naturally around popular libraries and ecosystems, but these “libraries and ecosystems” (actually they’re ontologies, but that’s another discussion). The big primitive here is namespace markets. A namespace market is simply a prediction market for the advancing of the namespace (or some series of namespace advances). The Urbit Foundation should stay out of the business of globally defining ontologies, simply providing enough in the kelvined distribution for the bootstrapping of arbitrary ontologies. A namespace market is the fully generalizable form of any market for information.
Metabolised: The entire financial economy, google etc.

Segment 3a: And Caesar wept, for there were no more names to bind
We are getting close to running out of low hanging fruit to stick in our namespace. So now begins the jihad: “Shrubs are the only acceptable form of computation”. The parts of the runtime that are not shrub-like should be redesigned to be shrub like. Redesigning the runtime to be shrubbish allows for arbitrary effects, outside of what the current closed set of vanes gives you. We can use this shrubbish-runtime to run urbit (or at least something that speaks the shrubbery protocol (nb: determinism is still a hard requirement)) literally everywhere. First phones and computers, then smart air conditioning, self-driving cars.
Metabolised: The entire tech industry

Segment 4: Serial Experiments Urbit
Having totally bound all of human existence into the namespace, we have successfully:
a) woken up the collective intelligence of man
b) merged urbit with reality itself
Metabolised: The entire will of life itself.
	Urbit, as it lives and breathes

	Segment 1: Bootstrapping
	Urbit should be the easiest place to write any application. Excluding applications that require high performance, shrubbery already makes this very close to true. It needs a little more love and sanding down rough edges, but it’s extremely close
	Result: The namespace is alive

	Segment 2a: The “maximalism” in namespace maximalism
	The most annoying part of working with Urbit is getting data in and out of the system. The solution to this is simple: Put everything inside the system. If all the data you could possibly want is bound somewhere, then a significant number of things we currently consider “applications” simply become transformations over the namespace. Moreover, this opens up the possibility of no-code shrubs because if the data already exists, then the necessity of marshaling shenanigans is a lot lower. We should leverage the network topology and provide stars incentives to republish data from the outside into their namespace.
	Metabolized: Zapier, IFTTT, bloomberg terminal etc.

	Segment 2b: Thinking with names
	Concurrently with 2a, AI efforts should be ramped up. See: vector database overlay. This allows for actually useful, personalized AI efforts. More importantly, with aggressive use of finetuning, so that the AI becomes accustomed to the user, which should markedly improve sense-made per watt. If the data formats are universal, then urbit AI doesn’t need to be all that intelligent to be a better experience than ChatGPT/claude or whatever. Intelligence = compute + context. We won’t win on compute (nobody but openAI will) but we can win on context.
	Metabolized: Every AI wrapper company

	Segment 3b: Markets
	Now that we have the world’s data inside the namespace, and cognition chomping at the bit to digest it, we can start to build out markets. Shrubbery has been designed so that primitives will coalesce naturally around popular libraries and ecosystems, but these “libraries and ecosystems” (actually they’re ontologies, but that’s another discussion). The big primitive here is namespace markets. A namespace market is simply a prediction market for the advancing of the namespace (or some series of namespace advances). The Urbit Foundation should stay out of the business of globally defining ontologies, simply providing enough in the kelvined distribution for the bootstrapping of arbitrary ontologies. A namespace market is the fully generalizable form of any market for information.
	Metabolised: The entire financial economy, google etc.

	Segment 3a: And Caesar wept, for there were no more names to bind
	We are getting close to running out of low hanging fruit to stick in our namespace. So now begins the jihad: “Shrubs are the only acceptable form of computation”. The parts of the runtime that are not shrub-like should be redesigned to be shrub like. Redesigning the runtime to be shrubbish allows for arbitrary effects, outside of what the current closed set of vanes gives you. We can use this shrubbish-runtime to run urbit (or at least something that speaks the shrubbery protocol (nb: determinism is still a hard requirement)) literally everywhere. First phones and computers, then smart air conditioning, self-driving cars.
	Metabolised: The entire tech industry

	Segment 4: Serial Experiments Urbit
	Having totally bound all of human existence into the namespace, we have successfully:
	a) woken up the collective intelligence of man
	b) merged urbit with reality itself
	Metabolised: The entire will of life itself.