Atlas Reference💾 Database Structure
Database Structure
A fascinating claim is easy to remember. But can we still find its source—and what we actually concluded about it?
Suppose you're reading an argument in the Atlas. A passage seems puzzling. You follow a proposition number, find a discussion of its meaning, and then discover a separate investigation comparing the claim with external evidence.
Now imagine returning to that argument months later. Can you retrace the steps? Has the conclusion changed? Was the source actually reviewed—or merely cited?
Those are the questions behind the Database Structure section. Its less glamorous but essential job is to make our research traceable: from the original words, through the analysis, to the limitations and unanswered questions.
What exists today: the Atlas publishes linked Markdown pages through VitePress, maintained in GitHub. Its registries work as organized records and indexes—not as a separate searchable SQL database, an automated verification engine, or a fully standardized machine-readable knowledge graph. The distinction matters because the way we present the research should never promise more infrastructure or certainty than we have.
The Main Research Collections
Think of the Atlas as a research library with different rooms. In one, we ask whether two passages can both be true. In another, we compare claims with outside evidence. Elsewhere, we document what a human editor has actually reviewed.
Each room has its own purpose. A record in one collection cannot silently stand in for an answer supplied by another.
| Collection | What it records | Where to explore it |
|---|---|---|
| Propositional Logic Registry | Individual assertions extracted from The Urantia Book, along with source locations, formal analysis, and interpretive tensions | Master Proposition Index |
| Analytical Audit Register | Which parts of the logic analysis have undergone documented methodological review, and which still require it | Analytical Audit Register |
| External Verification Registry (EVR) | Externally assessable claims, their evidence, current classifications, limitations, and source-verification audits | Evidence & Verification |
| Human Editorial Review | What the human editor personally examined, questioned, endorsed, or left unresolved; distinct from AI source checks | Human Review pilot |
| Cultural Reception Registry | Documented or disputed claims about the book's influence on people and creative works | Cultural Reception & Influence |
| Explanatory Atlas sections | The book's proposed geography, civilizations, beings, consciousness concepts, and diagrams | Universe Geography · Visual Maps & Diagrams |
These are related collections, not interchangeable kinds of evidence. A proposition's internal consistency, an EVR's external classification, and a human editor's sign-off answer different questions.
How a Claim Is Traced
Here's where the organization becomes useful. Imagine finding a sentence in Paper 5 and wanting to know what the Atlas has done with it.
The trail begins with the canonical passage—the actual source. A registered proposition points to one formally identified claim. Another record may then examine how a claim fares against independent evidence, while a human-review log documents which source material the editor personally checked.
These are related steps, not automatic endorsements of one another. The identifiers below show how to follow them without getting lost.
| Layer | Example | What it means |
|---|---|---|
| Canonical reference | 5:4.5 | Paper 5, section 4, paragraph 5 of the source text |
| Permanent proposition ID | UB-005-P338 | A proposition registered from that passage; the P number identifies an Atlas record, not a canonical paragraph |
| External-verification record | UB-EVR-022 | A separate assessment that examines this and other propositions from 5:4.5 against external sources |
| Cultural-reception record | UB-CR-001 | An example of a different record type concerning a reported cultural connection; not part of the 5:4.5 research trail |
These identifiers serve different purposes. The EVR and cultural-reception records do not automatically correspond one-to-one with propositions. One investigation may address several propositions, and a proposition may have no external-verification record.
For example, EVR-022 — Comparative Claims about World Religions explicitly connects its assessment to propositions UB-005-P338–P349 and to the canonical passage 5:4.5. Its human-review record separately documents the editor's involvement and pending decision.
You can also inspect the Paper 9 proposition analysis to see permanent proposition identifiers and distinctions among reconciled and unresolved tensions.
How We Distinguish Findings from Review Status
A detailed page can look finished long before every question has been settled. That's why we distinguish having a record, analyzing a claim, checking a source, and obtaining human or specialist review.
Each status answers a different question. None should be used as a shortcut for the others.
Registered means an entry has an identifiable place in a registry. It does not automatically mean its conclusions have been independently verified.
Analytically assessed means a textual proposition or tension has received a documented logical analysis. A provisional reconciliation is not proof that the book's statement is factually true.
Source-audited means the research record documents what outside material was checked and what limits apply. Some audits examine relevant excerpts or passages rather than complete books or research papers.
Human-reviewed requires an explicit record of what the human editor personally examined and decided. An AI-assisted source check does not amount to the editor's approval.
Expert peer-reviewed is a separate standard. The presence of scholarly references on an Atlas page should never be mistaken for an independent expert endorsement of the Atlas's conclusions.
See the Logic Methodology, EVR Source-Verification Standard, and Human Editorial Review record for the actual procedures and current limits.
Stable IDs, Links, and Revisions
A research project should be able to change its mind without losing its memory.
A question about the meaning of responsible inquiry: What does it take to change our understanding without quietly rewriting the path that brought us here? A stable record lets readers revisit an earlier conclusion, see why it was revised, and judge the reasoning for themselves. In that sense, the Atlas's research trail can help tell an honest story of inquiry. This is a reflection on accountability and interpretation—not evidence that any underlying Urantia claim is true, or a promise that the Atlas already has complete automated revision tracking.
If a source is corrected or an interpretation improves, readers still need to know which claim is being discussed and why an earlier assessment changed. That's why stable identifiers, source links, and visible qualifications matter.
The Atlas's organizational principles are:
- Preserve a permanent record identifier. Once a proposition or investigation receives an ID, later changes should not silently reuse it for a different claim.
- Keep the canonical passage identifiable. Readers should be able to distinguish a quotation from the Atlas's paraphrase, formalization, or commentary.
- Preserve interpretive qualifications. Technical terms, speaker or narrator, quantifier scope, and the level of description matter when testing contradictions.
- Link related but separate inquiries. A logic analysis may direct readers to an EVR, a related Atlas essay, or a review note without treating them as equivalent assessments.
- Show what remains unverified. Classifications and conclusions should be revisable in light of improved evidence and analysis. Changes should be documented instead of silently presented as long-established findings.
Our current registries, source notes, and editorial logs implement parts of these principles. The Atlas does not yet guarantee uniform metadata fields, automated link integrity, or completed audits on every record.
What Remains to Be Built
Imagine being able to begin with a single word, find every proposition that uses it, compare different interpretations, and immediately see which conclusions have been reviewed and which remain open. That would make a demanding investigation easier to navigate.
It's an ambition, not a feature we're claiming is already built. The name Database Structure points toward a future, more standardized and cross-searchable system connecting claims, terms, sources, and review records.
Possible additions include consistent machine-readable metadata, automated integrity checks, richer cross-registry queries, and visual maps of propositional relationships. Each requires proper design, testing, and documentation.
For now, follow the work we can actually show: the Master Proposition Index, EVR Index, and Cultural Reception Registry.
🎭 Celestial Intermission — The Database Administrator
Fictional office comedy, courtesy of the Reversion Directors.
MORTAL: "I've developed a database for the entire universe."
CLERK: "Excellent. Is it fully cross-referenced?"
MORTAL: "Not yet."
CLERK: "Have all the claims been verified?"
MORTAL: "Not yet."
CLERK: "And has everyone agreed on the terminology?"
MORTAL: "Absolutely not."
CLERK: "Congratulations. You have successfully invented a research project."
The Reversion Directors stamp the paperwork: WORK IN PROGRESS, ETERNALLY.
This exchange is fictional satire, not a canonical quotation or a statement about actual celestial administration.
A trustworthy archive is not the one that never changes. It is the one that lets readers see what changed, why it changed, and which questions still deserve attention.
Organize the claims. Preserve the sources. Keep the conclusions open to correction.