Structured Wisdom

01 · Working Premise

The record comes before the synthesis

Structured Wisdom is a working inquiry into how firsthand human observation becomes a durable record, how provenance and uncertainty survive transformation, and what happens when those records are retrieved, combined, and synthesized by machines.

Machine synthesis is downstream of records. It can reorganize, compare, compress, infer, and generate from what is available. It does not create the missing primary observation.

The concern here is the continuity of the record: what was observed, why it mattered to the observer, what was inferred, what remains uncertain, and whether source lineage survives reuse.

The working rule is simple. Use established standards where they already solve the problem. Add structure only when it preserves meaning, provenance, or useful context that would otherwise be lost.

This is not a completed framework. The page records propositions, source connections, questions, and revisions as the work develops.

02 · Current Model

A working anatomy of a human record

A useful record may contain several different kinds of information that are easily collapsed together after the fact.

Situation → Observation → Salience → Interpretation → Boundary

Situation The conditions in which the record was produced.
Observation What was directly seen, heard, measured, encountered, or documented.
Salience What the observer considered worth recording, and where relevant, why.
Interpretation The meaning, explanation, or inference applied to the observation.
Boundary What the record cannot establish, what remains unknown, and where its evidentiary limits end.

The sequence is an analytical model, not a required writing template.

Human surface A readable record that preserves context and can stand on its own without machine interpretation.

Editorial surface Identifiers, source notes, distinctions, relationships, status, and revision information used to maintain the record.

Machine surface Metadata, structured data, and provenance relationships that allow the record to be found, exchanged, connected, and interpreted downstream.

The surfaces should derive from the same underlying record rather than becoming separately maintained versions of it.

03 · Research Cards

Current propositions

Each card records one proposition, what it appears to establish, what it does not establish, and the question it opens.

Card 001 Working · 2026-09-20

Synthesis does not create primary reality.

A synthesis is downstream of records. It can compare, compress, reorganize, infer, and generate from what is available. It cannot retroactively supply an observation that was never made, recorded, sourced, or preserved.

What this establishes

Primary records remain a distinct evidentiary layer even when later synthesis is useful.

What this does not establish

A primary record is not automatically accurate, complete, objective, or reliable simply because it is primary.

Question opened

What minimum provenance should survive when a record is quoted, summarized, embedded, extracted, or synthesized?

Card 002 Working · 2026-09-20

Human salience is part of the record.

Observation is selective. A person notices something, ignores something else, chooses what to record, and places it in context. That selection can contain information about why the record exists and what mattered at the moment it was made.

What this establishes

A durable record may need to preserve not only what was observed, but why the observer considered an element worth recording.

What this does not establish

Salience is not proof that something is objectively important, representative, or true.

Question opened

Can salience be preserved explicitly without turning ordinary human writing into a form?

Card 003 Working · 2026-09-20

Provenance should survive transformation.

A record may be copied, summarized, translated, indexed, embedded, quoted, or synthesized. Each transformation can increase usefulness while also increasing distance from the source.

What this establishes

Source lineage is part of the information. It is not merely administrative decoration.

What this does not establish

Provenance does not establish truth. It establishes traceability.

Question opened

Which existing standards are sufficient for preserving that lineage, and where do they stop short of the human record?

Implementation note: these cards are stored once in the PHP record at the top of this file. The same source data produces both the visible human record and the page's JSON-LD.

04 · Existing Standards

Use what already exists

The project should not invent vocabulary merely to make the project appear complete. Existing standards are the starting point.

DCMI Metadata Terms
A mature vocabulary for describing resources, including identity, authorship, dates, relationships, formats, subjects, and other basic metadata.
W3C PROV-O
A provenance model for representing relationships among entities, activities, agents, derivations, and responsibility.
W3C / OGC SOSA–SSN
A vocabulary for observations, procedures, features of interest, properties, results, sampling, and the systems involved in observing them.

Inclusion here means the standard is relevant to the investigation. It does not mean that Structured Wisdom has adopted every part of it.

05 · Open Questions

Questions still doing useful work

Q01 What must be preserved in a human-readable field record so that later synthesis can distinguish observation from interpretation?
Q02 How should human salience be recorded without making ordinary writing mechanical or formulaic?
Q03 What provenance should survive quotation, extraction, summarization, embedding, translation, and synthesis?
Q04 When does an established standard already solve the problem, and when is a new field or relationship actually justified?
Q05 What information should remain deliberately unstructured?

06 · Working Record

The project should show its own development

20 Sep 2026 Public notebook opened. Initial working premise, five-part record model, three research cards, standards references, and open questions established.

Next test

Take one firsthand historical record and one operational business record. Preserve the original human-readable account, map each to existing metadata and provenance standards, and document where meaning survives cleanly and where context is lost.

07 · About This Record

An independent working notebook

Structured Wisdom is maintained by Jeffrey Ziegler as a place to record developing work on human observation, information structure, provenance, retrieval, and machine synthesis.

It is not a standards body, product announcement, or finished methodology. Entries are published when they become useful to the investigation and revised when the record changes.

Professional background: ziegler.co