Hi Evgenii, all,
Thanks for putting v0.1 and the change log together.
To answer the point at the end: I have no objection to the row 4
construction. Nicolás’s explicit cause value, as carried into sections 1.2
and 4, answers my representation question in 0076. The row compares the
declared cause with state, with confinement-failed-during-check as the
cause value. Classifying that comparison as consistency is right. The
qualification in 1.2 also matters: checking those declarations does not
establish that every failed confinement control was disclosed. No reversion
is needed on my account.
Section 5.4 also addresses my two wording requests. It makes clear that a
negative witness must be an actual fail from the relevant declared check,
binds the control to the same checker revision and relevant configuration
and constraints, and requires the prior-run reference to establish that
same binding. Keeping “nothing” available, without requiring a separate
negative witness for every check, preserves the scope I intended.
That closes my row 4 representation question and my two §5.4 wording
requests. O2 still concerns aligning the release corpus with that text, and
O5 concerns the second reader’s re-run of the historical 272-member report.
Those remain separate from this confirmation.
Roel
