Method, the Decision Test, and the Death Test

Borrow power without surrendering sovereignty

We use servers, models, networks, hardware, payments, and open-source capabilities the world already does well, while important dependencies should have clear interfaces, replaceability, portability, auditability, and exit paths.

Never outsource the core

Capabilities that determine who you are, what you possess, who has authority, and how authorization, revocation, recovery, migration, and exit work are core and cannot be wholly controlled by an irreplaceable third party.

Mature commodity capabilities outside the core need not be rebuilt for the pride of doing everything ourselves.

Protocol before product

Core interoperability changes should define a public contract before Hoop OS or another implementation builds them; a product cannot use private code to present its own behavior as the protocol.

The four sovereignty tests

Missing any test means partial sovereignty, and we must state what is missing, the destination, and the first step.

• Source: the source of truth lives on your machine.
• Migrate: it enters the export and can be reborn on another machine.
• Audit: you can see who changed it.
• Unplug: it remains when any one company is removed.

The honest current state: Company Death Test

• L1 passes: the current export contract covers local data lanes, providing an implemented basis for retaining and reading local data apart from company services.
• L2 passes: the local agent service lives on the Hoop node itself and does not use a company account as its local runtime root.
• L3 does not pass: the identity root has not yet been returned to the owner; an owner root key, device add/revoke, recovery, and same-identity migration remain to be implemented and drilled.
• L4–L8 are not represented here as passed; each level needs its own implementation and evidence.

The seventeen-question Decision Test

• What power does it add for a person?
• What freedom does it restore?
• What new risk does it create?
• Can you see and understand that risk?
• Can you refuse?
• Can you revoke?
• Can you recover?
• Can you migrate?
• After leaving HoopGram, can you still possess what is yours?
• Do we truly need to build it ourselves?
• Is there a better power we can borrow?
• After borrowing, can we still replace the provider?
• Did convenience hand sovereignty to an irreplaceable third party?
• Are we solving the real problem or a symptom?
• Can failure recover safely?
• Does this make people more autonomous or easier for HoopGram to control?
• In the end, does the person still have a choice?

What we do not do

• We do not make Hoop Protocol an appendage of one company, account, cloud, model, domain, bootstrap, programming language, or private codebase.
• We do not profit primarily from artificial migration difficulty, closed formats, hidden interfaces, or locked identity.
• We do not default to blockchain or global consensus for problems that do not require them.
• We do not grant unlimited authority because an agent is intelligent or treat authentication as unlimited authorization.
• We do not manipulate your choices “for your own good” or equate dangerous capability with capability that must be taken away.

← Back to /who← All posts