Instruments first, then the method that documents them, then the argument they were built inside. Everything below is dated, and everything that has been revised says so on the page.
Three tiers. The instruments carry dated measurements against declared thresholds. The method documents how they were built and when they are retired. The papers and essays are the argument the instruments were built inside, and they are labelled as argument.
Long-form theory, graded claim by claim. High-confidence physics is marked as such; philosophical commitments are marked as commitments. Where the evidence desk finds an unnecessary dependency in FP1's own argument, that note is published inside the paper rather than removed from it.
When you need depth on a specific decision, there is a way in.