Foundational manifesto
A binding document. Not marketing, not team copy: the moral constitution under which Basis Network is built. Modifiable only through an open procedure via the Constituent Boulé, with a public consultation period of at least sixty days.
Preamble
It is the problem no human society has fully solved. The Greek assemblies, the Roman republic, the Renaissance mercantile republics, modern constitutions, contemporary digital platforms — all offered coordination mechanisms, and all, without exception, eroded along the same path: the asymmetry between observer and observed, between decider and decided, between who can verify and who must trust.
Basis Network is not the solution to that problem. It is a computational infrastructure that narrows its margins in a bounded domain: transactions, data, and computations that can be formally represented. It does not replace human deliberation, it does not substitute institutions, it does not abolish politics. It does something more modest and more demanding:
it makes verifiable, in the mathematical sense of the word, what until now was only trustworthy in the sociological sense.
This manifesto declares the principles under which that infrastructure is built, the consequences that follow from them, and the commitments undertaken by those of us who sustain it.
The nine principles
Each ternary groups principles that hold each other up: truth and its verification, organization and its form, operational ethics. None of them is decorative. Each one is instantiated in code, in institutional architecture, or in economic mechanics.
First ternary
A claim about the state of the world is valid when it can be verified by whoever doubts it, not when it has been issued by whoever has the authority to issue it. In Basis: no state change is accepted without reproducible cryptographic proof, and no proof is considered valid because of who signed it.
The truth of a system does not reside in its implementation but in the mathematical formalization that the implementation tries to embody. When code and specification diverge, the code is the error. It is the only known way to scale a computational system in complexity without accumulating silent debt of correctness.
The false dilemma of the previous era forced a choice between protective opacity and surveillant transparency. Modern cryptography dissolves it: one can prove that a state transition is valid without revealing its content, prove membership without naming identity, audit a computation without observing the data. The opaque and the correct coexist in production.
Second ternary
Each subnet defines its own rules: who enters, what is charged, who it talks to. That autonomy is real, not decorative, and is exercised without asking permission from a central coordinator. When two subnets choose to cooperate, they do so through cryptographic treaties verifiable by both. The first time institutional diversity coexists with universal verifiability without an instance that ranks one over the other.
A network that runs multiple languages, hosts multiple privacy models, and permits multiple economic policies per subnet is not a confused network: it is one that has accepted that no single design is optimal for all uses, that the errors of one niche do not contaminate the rest, and that evolution requires variation. Homogeneity is efficient in the short term and fragile in the long term.
We do not decentralize for aesthetics. We do it because concentrated systems, however well intentioned their operator, generate asymmetries of power that end up capturing the operator itself. Decentralization is an upper bound on the corruption possible: it limits what any actor — including ourselves — can do wrong.
Third ternary
As artificial intelligence runs on top of our infrastructure, we take on a commitment no centralized AI system has taken on: the execution of a model leaves a verifiable cryptographic trace of which input produced which output under which weights at which nodes. Weights can be private; data can be encrypted; but traceability of inference is a right of the subject being inferred upon, not a concession from the operator who infers.
The protocol is non-rival common property, under the custody of an independent non-profit foundation. Private companies can capture value in the adjacent layers: services, integrations, proprietary products, hosting. What no entity can do, whatever its present or future shareholder composition, is modify the protocol in its favor, introduce operator privileges, or install backdoors.
Cryptographic verification does not resolve political legitimacy, does not compensate for digital exclusion, does not eliminate plutocratic capture of the vote, does not guarantee collective wisdom, does not produce material justice. Basis is a layer of verifiable coordination; the deep questions of human coexistence are still played out, as always, in the deliberative spaces where speech is exercised.
Ideological axis
The twentieth-century cleavage organized itself around ownership of the means of production. The twenty-first-century cleavage is organizing around who can verify what, against whom. That question cuts orthogonally to the classical left-right axis, and defines two radically distinct ways of inhabiting the computational world.
Asymmetric by design
Symmetric by design
Basis deliberately sits in the symmetric quadrant. Not by aesthetics or dogma: because it is the only configuration that simultaneously fulfills the principles of verification, pluralism, and public good declared above.
Operational commitment
What follows are public obligations. The separation between the protocol layer and the services layer is not rhetoric: it materializes in the legal separation between the foundation that custodies the protocol and the companies that operate on top of it.
Publish and keep open every formal specification of the protocol, so that any technically qualified person can verify the correspondence between mathematics and code.
Keep the protocol under a permissive open-source license and under the custody of an independent foundation whose governance cannot be unilaterally controlled by any commercial entity, including the company currently leading development.
Not introduce protocol-level privileges reserved for any commercial entity, neither in its present version nor in future ones, regardless of whether the entity changes ownership, merges, or is acquired.
Publicly document any relevant conflict of interest between protocol decisions and the commercial interests of the companies that build on top of it, and submit those decisions to the foundation's governance procedure.
Explicitly document every trade-off between privacy, auditability, efficiency, and cost, without presenting as emergent properties what are deliberate decisions with consequences.
Publicly acknowledge design errors when we discover them, even when doing so is costly, because trust in a critical infrastructure is built on honesty about its failures, not on their concealment.
Cultivate a community of operators and users that is geographically, linguistically, and institutionally diverse, because a network decentralized in its code but homogeneous in its community is just centralization with more steps.
Modification
Through the open procedure described in the Governance Charter: constitutional decision of the Foundation with three-quarters of the board, plus a public consultation period of at least sixty days through the Plaza Común. The authority of this text rests on its correspondence with the code and with the institutional architecture that instantiates it.
Its relevance will be measured, years from now, by the distance between what it promised and what we actually did.