We attempted to write a hunch as a procedure. Within six weeks the procedure had a name, a metric, and a dashboard, and the hunch was gone. This note recovers the shape of the failure: how formalisation eats the thing it was meant to protect, and why we keep the ruin rather than the repair.
Status · PublishedThis note is in the ledger. The full transcript is not yet cleared.
Editorial status · Abstract and research roots published. Full body forthcoming.
Research roots · tacit knowledge; knowledge-how; Goodhart-style measurement failure.
Our first mistake, kept for the record and for the shape of the failure. Reviewer 2's verdict: Publish the failure, not the method.
← Back to the ledger