ja-inventory: rev 5 concepts and frames for the original bed (model-drafted, Validation-informed), weksa inventory admit, onto:ja-JP:15 #54

Merged
weksa merged 5 commits from weksa-lowering/ja-inventory into main 2026-11-12 18:08:04 +00:00

2026-11-12

Weksa
ja-inventory: map and README name the admit door and the Validation-informed rev 5
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-12 17:48:20 +00:00
Weksa
ja-inventory: onto:ja-JP:15, above and out-group scene entry
Generator: claude-sonnet-5-5 (Weksa Hands, model-drafted scene ontology). Second draft after a reader found the first opposed ja-JP:3; two fresh readers then named nothing.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-12 17:48:20 +00:00
Weksa
ja-inventory: data/original inventory rev 5, the bed's own lexicon (233 concepts and frames)
Generator: claude-sonnet-5-5 (Weksa Hands, model-drafted original-bed lexicon from Train and Validation passages). Validation-informed: numbers on this bed measure lowering, never generalization or convergence. Test never read by any drafter or reader. Admitted through weksa inventory admit; data/daodejing is a revision prefix, proven by the_dao_inventory_is_a_prefix_of_the_original_inventory.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-12 17:48:19 +00:00
Weksa
ja-inventory: delete the keyword lexicon tripwire and the byte-identity test
A substring table judging whether Japanese names a concept is a rules-based reader of meaning; byte identity of data/original and data/daodejing stops being true by design (rev 5 follows).

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-12 17:44:28 +00:00
Weksa
ja-inventory: weksa inventory admit, the one door for a model-drafted revision outside a run
into_revision takes its cap from the caller: the loop keeps MAX_EDIT_CHANGES, the CLI door takes 256 for a draft reviewed whole in its cut.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-12 17:44:10 +00:00