All 31 files from ideas/passepartout-economics/ promoted to ideas/ root. - Subfolder's passepartout-economics.org (42-line index) renamed to triad-index.org to avoid collision with root-level full doc - index.org removed (redundant — triad-index.org replaces it) - Root-level passepartout-economics.org: stripped file:passepartout-economics/ prefix from all cross-references (now simple file:foo.org links) - compliance-framework-mapping.org: same prefix cleanup - All internal file: links within the economics docs already used simple names (no prefix) — they resolve correctly from ideas/ root
25 lines
1.7 KiB
Org Mode
25 lines
1.7 KiB
Org Mode
:PROPERTIES:
|
|
:ID: efc76898-03f7-57ba-923d-35d65da88bb7
|
|
:END:
|
|
#+title: The Per-Domain Sufficiency Flip
|
|
#+filetags: :passepartout:sufficiency:flip:domain:knowledge:
|
|
|
|
The sufficiency flip is not a single event — it happens independently for each domain, and some domains never flip.
|
|
|
|
**Knowledge types required:**
|
|
- **Structural** (published rules, configs, specs) — LLM translation + ACL2 verification. Flips in days.
|
|
- **Empirical** (what happens when X?) — active sandboxed probing. Flips in days to weeks.
|
|
- **Performance** (latency, throughput) — benchmark harness. Flips in hours.
|
|
- **Transfer** from related domains — ontology alignment. Flips in days.
|
|
- **Tacit** (craft expertise, organizational culture) — requires Phase 3 archivist loop over time. May never fully flip.
|
|
- **Aesthetic** (poetry, creative writing) — never flips. The system is honest about its frontier.
|
|
|
|
**Fastest path to flip any domain:**
|
|
1. Ingest all published text (hours)
|
|
2. Run benchmark harness (hours)
|
|
3. Run active sandboxed probes (automated)
|
|
4. Generate contrastive queries for the 5% uncertain rules (one human session, a few hours)
|
|
5. Start serving real interactions (empirical loop tightens from first interaction)
|
|
|
|
For the Lisp Machine bootstrap, every subdomain is software (the most codifiable domain). The entire bootstrap can flip in days to weeks with one human review session. The [[file:gate-rule-encoding.org][gate rule encoding]] process feeds directly into this: each domain's rules are formally encoded and verified. The [[file:time-estimates.org][time estimates]] for the overall project are derived from the time to flip each subdomain. The [[file:cost-structure.org][cost structure]] shifts from LLM-token-heavy to verification-heavy as more domains flip.
|