n.siegfried
d092db3920
Wave 7 opens with the largest engineering item in the plan, and the wave file is explicit that it ships alone. This is the measurement that should precede it - the same role tokens.md played for T3.2, produced for the same reason: the estimate in the plan came from a read of the symptom, not a count of the work. Four numbers, and where they come from: page-stylesheet selectors colliding 21 (9 of them the sequence editor) script top-level names colliding 9 (all of them one feature done twice) markup ids colliding 0 cross-frame call sites 28 across 8 scripts and 1 page The zero is the largest piece of good news available: 96 and 127 ids and not one shared, so every getElementById in both files survives the merge untouched. The nine script collisions are misleading in the other direction. They are not nine names for nine things - they are usage analytics and the feedback panel, each implemented twice. The merge is a de-duplication, not a rename, and behind nine names sit two parallel implementations. Also recorded: every cross-frame call added by waves 5 and 6 is a shim over the boundary, is commented as such, and is DELETED by T7.1 rather than migrated. T5.5 already proved the SOP-borne propagation path needs no boundary crossing at all, which is X4 resolved rather than outstanding. And three questions no count can answer, which T7.1 has to settle: whether the creator gets an app bar back, whether the sequence editor becomes one component or two, and what happens to body.embedded when "framed" stops being a state. Wave 7's implementation is NOT started. This commit is measurement only, and nothing in html/ or server/ changed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>