Physics–mathematics claim-family and object expansion
Date: 2026-10-25 Status: complete
This is the adversarial expansion below the normalized topic register. It does not treat topic completion as proof that every local corpus object has been admitted, excluded or correctly mapped.
Census
| Measure | Count |
|---|---|
| Inventory objects | 3276 |
| Explicitly mapped in-scope objects | 1706 |
| Typed exclusions | 1570 |
| Unmapped scientific objects requiring scope review | 0 |
| Weak semantic routes requiring independent review | 0 |
| Object × topic × applicable-family cells | 55393 |
| Unique inherited claim cells | 55393 |
| Authored local-exception resolutions | 0 |
| Objects with multiple topic-specific dispositions for one family | 9287 |
| Deduplicated claim-family clusters | 491 |
Reading the result
Each inherited cell is one object-to-topic claim family with exactly one disposition. If an object maps to several topics, those claims remain distinct: an operation can be established on one carrier and obstructed on another without contradiction. The multi-disposition count is therefore a join diagnostic, not an error. A title match is accepted as a strong route; a lead/context-only semantic route remains an explicit review queue. Unmapped scientific objects are not declared irrelevant merely because they lack frontier vocabulary. Typed exclusions remain in the inventory with their reason. The JSON register retains every object hash, source candidate and claim-family cluster.
Current decision
The topic layer is complete. The object layer is complete because 0 objects lack a route and 0 mapped objects still have weak semantic routes requiring independent review.