TALAMANA · THE COMPLETE MAP · ABOUT · THE METHOD · SOURCES · OPEN DATA

How Talamana is made

Every idea on the map says what kind of claim it is, how settled it is, and what it rests on.

ONE IDEA, ONE CARD

Each of the 287 ideas is a single teachable thing. Its card carries a plain-language description, the evidence that shows you have it, a crit question, an approximate age, and its sources.

EVERY LINK HAS A WRITTEN REASON

The map holds 614 links between ideas. Every link says in one line why it exists. 396 are strong links: they name a genuine foundation. 218 are helpful links: they offer optional enrichment and are never a compulsory step. There is no jump from zero to mastery. There is a path, and it is walkable.

TWO LABELS ON EVERY CARD

The first label says what kind of claim it is. FACTUAL — a checkable claim; the sources are listed on the card. POSITIONAL — the Lab’s own view, stated with the strongest case against it and what would change our mind.

The second says how settled it is. DURABLE — it will read the same in five years. EVOLVING — true today; we check it every edition and the card shows the date. CONTESTED — experts disagree, and the card says where. THE LAB’S HELD POSITION — a position we hold and argue for. THE LAB’S METHOD — a way of working we recommend; it works, and it is not the only way that works.

Our test for a FACTUAL · DURABLE label: if a reasonable expert could accept all the evidence and still reject the idea, it is not a fact — it is a position, and we label it as one. Cards that state a legal position show the month the law was checked.

THE TWO-ANCHOR RULE

A FACTUAL idea carries at least two independent sources — frameworks, verified courses, peer-reviewed work, or published essays. A POSITIONAL idea is anchored to the published text it comes from. Today the map holds 175 factual ideas and 112 positional ones, each marked. Most curricula hide their opinions inside their content. This one publishes its bias honestly.

JUDGMENT NEEDS A CRIT

Generic AI literacy has no category for what design education actually certifies: judgment. 54 ideas on the map are typed JUDGMENT. They can only be evidenced in a studio crit. Each carries a crit question addressed to a reviewer, not to a quiz engine.

TWO POSITIONS GOVERN THE WHOLE MAP

Ethics is threaded, not boxed. There is no terminal ethics module. The Ethics & Provenance strand wires into every other strand.

Students direct AI; they do not build it. The map’s ceiling is directing, judging, and choosing AI well — the architect steering the instrument, not engineering it.

AGES

The age rings show when an idea can first be met, not who is allowed to learn it. An on-ramp band of 22 ideas for ages 11 to 15 gives the map a foundation in school-level probability, data, computing and media literacy; those ideas adapt text from the Marble Skill Taxonomy (CC BY-SA 4.0) and each carries its attribution.

REVISED EVERY EDITION

Talamana is revised every edition from feedback by teachers, practitioners and students, and not a standard or a peer-reviewed text. Evolving cards show the date they were last checked. Corrections are welcome; the sources page says where every claim comes from.

Open the map · Logika · RBDS AI Lab, India · revised every edition.