Karma — the Public Ledger
Karma is the read side of Impactium — the Impactivism Public Ledger. It is the record of what every Entity has put into the world, projected so it can be read back, verified, and built on. It is the working name for the Hive-Knowledge Graph.
What Karma does
Karma continuously turns the chain into a form that is fast to query and easy to walk:
- Indexes the chain's event stream into a property graph — Entities, Products, Capsules, Constitution articles as nodes; created, spawned, minted, voted, and residual flows as typed edges.
- Serves that graph through a single typed GraphQL API — the surface people, dashboards, and AI agents read.
- Proves any fact on request — escalating a graph answer to a cryptographic proof against the chain's committed state root.
- Hosts the ecosystem's Chain Resources — the on-chain package and document content, including the
npm install-from-chain gateway.
Fast, not trusted
Karma is a disposable projection. Delete it and it rebuilds byte-for-byte by replaying the chain; nothing of value lives only in it. That is precisely why it can be trusted to move fast and evolve freely: it is never the source of truth, only a convenient, provable view of it. Any consumer that wants certainty can bypass convenience and verify a fact directly against the chain.
The lookup engine
Karma's busiest job — by far the most common query on the network — is the lookup: returning a timestamped snapshot of a record's COA↔COI binding from the perspective of the party asking. Impact value is relational and temporal: what a record is worth depends on who is looking (their alignment to it) and when. Recomputing that on-chain for every read would be far too slow, so Karma indexes the graph to serve perspectival lookups quickly. Making lookups fast is a core reason the indexing/notary layer exists at all.
A lookup comes back in one of two forms:
- Impact-Stamp — a fractal portrait of the result: a visual fingerprint of the recursive impact web behind a record or entity. The most common form is the Sierpiński triangle, the natural picture of the self-similar web-of-webs that lineage and impact chains form.
- Ledger-Report — the data form, which always carries a Scope-Matrix that sets how broad and how deep the lookup reaches across the impact pathways. The scope you can see is bounded by your relationship and permissions.
When a lookup reaches an impact pathway it isn't permitted to follow, it doesn't fail and it doesn't lie: it returns the total impact (Total($IMP)) with limited detail. You always learn the total and the net position; you don't always get to walk into the specifics — the privacy rule, enacted as you traverse.
The questions it answers
Karma is built around the questions the Impactium model exists to answer:
- A member's impact history — every capsule and cost they've created, including the visibly-failed claims (failures are shown, never filtered out).
- The downline walk — an Entity, its entire referral Chain of Impact, and the impact created across that entire lineage. This is the question the referral model exists to answer: show the impact across an entire downline.
- Provenance — for any fact, the exact chain commitment it can be proven against.
Records can be filtered by scope (inward / outward) and by verification tier (self-logged / notarized), so a consumer — especially an AI — can weight what it reads by how it was verified.
A read surface that governs
Karma is increasingly more than a viewer. Because the network weights governance votes by a member's proximity to a subject, and the leading proximity measure is graph-derived, Karma is becoming a genuine input to how Impactium governs itself. Its proximity queries are held to the same standard as everything the chain commits to: deterministic, versioned, and provable.
For AI
Karma is designed to be memory for AI — a typed, walkable, permanent, verifiable record that an agent like Aionima can read as its knowledge of humanity's impact, and prove any entry it relies on. A knowledge graph that cannot silently lose or rewrite an entry is close to an ideal substrate for machine memory, and that is exactly what Karma is built to be.