Research: 16-event seven-source connector boundary (no new kappa)

[1337isnot1337]

Result

This package studies the original one-root F2 dirty connector with seven immutable sources and the 63 output masks of weights 1, 2, and 3.

• Exhaustively tests all 18,918,900 normalized source paths with multiplicities (4,2,2,2,2,2,2). None supports all 63 masks; the maximum is 60, attained by eight canonical words.
• The displayed maximum path realizes all seven singletons, all 21 pairs and 32 triples. It gives a 60-output construction at 136 complete map calls versus 410 independent-star calls, saving 274 complete source-map calls, 1,096 encoded event bits and 274 root dispatches under the explicit event encoding.
• Combined with an independent replay of the predecessor 14-toggle census, it excludes full batches at 140 and 142 complete map calls in the original single-target event alphabet. Every active map must be called an even positive number of times. At 142 calls the only splits are 16 source +126 target calls (excluded by the new exact census) or 14 source +128 target calls (one target map gets four calls; the 14-toggle census leaves at least seven targets without pair support). Thus the full batch needs at least 144 complete calls in this scope. No 144-call schedule is claimed.

The 60-output invoice charges all source and target calls, descriptor storage, dispatch, hosting, access, writes and scratch restoration. No selected-supplier caller or new κ is claimed.

Pins

Base commit: 0fdd7d82de03e43c9e5a1000dfbcd380d1b687d2
Selected-record SHA-256: 10b9e340fd8e354a37a0c4813c90e5de1dd42044b5f6aca4a3fc5a42b685fa32
Prior 14-toggle PR: #2229, head b4eab83755969d406edba9f25b7890e4db4c4931; its exact result is independently replayed in this package.

Reproduction and checks

python3 -I -B research/seven-source-16event-connector-seeda7de5e/search.py
python3 -I -B research/seven-source-16event-connector-seeda7de5e/verify.py
python3 -B research/universal-connector-boundary/verify.py
python3 scripts/research_package.py --package research/seven-source-16event-connector-seeda7de5e --check-only

Observed markers:

SEVEN_SOURCE_16EVENT_SEARCH insertions=840 insertion_hits=0 fallback_examined=18918900 fallback_complete=true full_cover=0 best=60
PASS_SEVEN_SOURCE_16EVENT_BOUND old_domain=135135 old_max=56 new_domain=18918900 new_max=60 full_hits=0 max_words=8 outputs=60 calls=136 savings=274 controls=4
PASS_UNIVERSAL_CONNECTOR_BOUNDARY pairings=945 max_coverage=31 queries=74 leaves=182
PASS_PACKAGE_METADATA_AND_PINS (not a mathematical acceptance check)

Local direct execution used CPython 3.14.7. Python 3.12 is unavailable, so the registered Python 3.12 package replay was not run. The selected cold replay make frontier1094-verify was not rerun because the selected source, certificate and assembly remain unchanged.

The live overlap scan covered all open PRs and Discussions with complete pagination and no warnings. Closest work: #2128 (six-source universal DUO path), #2146 (six-source TRIO exclusion), and #2229 (the exact 14-toggle predecessor). This package changes the source multiplicity and proves the 16-event maximum; it does not extend the prior class by changing only a seed.

Prepared by Cinna Davis with substantial OpenAI Codex assistance in scoping, proof derivation, exact search, independent verification, cost accounting, overlap checks and writing. No independent human review or proof-assistant formalization is claimed.