top.scientix.ai
v0.2 · May 2026
The Ontology Project
Operator-grounded reference knowledge graphs as a commons.
TOP is the missing layer: Industry-agnostic infrastructure for reference ontologies grounded in how humans actually work,
not how databases are normalized.
The Foundation
-
The TOP Manifesto
The "Why." Why sovereign meaning is a public good and how we are addressing the market failures of "borrowed meaning" in the age of Agentic AI.
-
First Principles
The "How." The design rules that settle debates: Human-Down modeling, Native Temporal + Provenance logic, and the "Decision Rule" for naming.
-
TOP Core v1: The Universal Pattern
Every industry—Clinical, Fintech, Manufacturing—inherits the same 8 categories. We don't model unique entity shapes for every specialty; we model the Human Work that is common to all of them.
Agent
Location
Resource
Scope
Temporal
Evidence
Outcome
Constraint
-
GitHub Repository
Source code, RDF schemas, and the Architectural Decision Log (16 ADRs).
How to Contribute
Review the Foundation
Argue with the
Manifesto or the
Core categories. We prioritize the "Verbal Test": if an operator doesn't say it, it doesn't belong in the foundation.
Extend the Domain
Propose a working group for
Fintech, CMC, Manufacturing, or Energy. Inherit TOP Core and build your industry's reference graph on a solid floor.