Publications / 001

We publish the path,
not only the result.

What the laboratory learns, including what did not work. Corrections are published beside the claim, not in place of it.

01 / What we publish

Accountable
to evidence.

A laboratory that reports only its successes is reporting only half of what it learned. Negative results and retractions are part of the record, not an embarrassment to be removed from it.

Research notesExperiment reportsTechnical reportsSystem cardsOpen problemsNegative resultsRetractions
In preparation

Atlas: When the Model Card Was Wrong

A documented retraction and the experiment that changed the laboratory’s central conclusion.

Publication in preparation — not yet released.

Request the publication ↗︎
02 / Laboratory ledger

What changed,
and why?

The ledger records:

  1. 01Hypotheses proposed
  2. 02Experiments executed
  3. 03Beliefs revised
  4. 04Results reproduced
  5. 05Claims retracted
  6. 06Invariants preserved

A laboratory should not hide the history of how it learned.

The public laboratory ledger is being prepared.

Request access to the laboratory ledger ↗︎