Land the ranking handoff on main so a new session can pick it up - #416
Merged
Merged
Conversation
It was written on the-index (PR #415), which also carries the index, the SessionStart digest trim and merged detector_history revisions -- four unrelated things in one PR. The ranking work is the priority and should not wait behind them, so it lands alone off main. Nothing is built. A design brief with a role-1 claim check, NOT murderboarded, and it says so in its own banner. Five decisions are marked as Tony's. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
One file. It was written on
the-index(#415), which carries four unrelated things — the index, the SessionStart digest trim, mergeddetector_historyrevisions, and this. The ranking work is the priority and shouldn't wait behind them.Nothing is built. A design brief: why F1 cannot separate these detectors (the background-axis winner flips between two blocks of twelve seeds), the four instrumented failure modes including
distractor_hitswhich has never entered a ranking, the chance-coincidence measurements, and the correction that Cossart transfer already works via--score-specso ground-truth measures do survive to another corpus.Five decisions are marked as Tony's with why each can't be defaulted, and four things not to do — headed by not re-baselining the three
test_background_curveasserts, which encode a claim rather than a measurement.Role-1 claim check, not murderboarded, stated in its own banner.
🤖 Generated with Claude Code