diff --git a/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md b/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md index 56fb5f79..19037b03 100644 --- a/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md +++ b/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md @@ -861,3 +861,10 @@ Where two preparers disagreed, the chosen one is named. on the bare subject and on `.>` alike (one handler, both forms), for events and for views, with the test that a bare-subject event reaches the handler; an event a module declares it consumes and nobody handles is said loud in the journal with the subject, as today, and counted in the module's health as a finding. P3, P9, P7. +- **H-A-167 (a re-pin moves the fingerprint; the pin is always the newest release, 17:40)** Met in the review + of H-A-165: a re-pin that leaves the fingerprint as it is leaves a build done at that fingerprint against the older + pin as the build the mesh runs, never built again against the release. The pins are part of what a build is made + of (G §47, ADR 0573 §4): the version line's fingerprint includes them, so a re-pin moves the importer's fingerprint, + the importer becomes a mover (H §745) and the builder builds a new key against the release; a pin is the newest + released version on the branch, an accepted series' older version never standing over a newer release; a failed + re-pin is not acknowledged, so the package-published is delivered again. P3, P5, P7.