M12.4: preserve exact biology across ownership candidates
Foundation / checks (deploy) (push) Successful in 5s
Foundation / checks (ecs) (push) Successful in 2m18s
Foundation / postgres-tools (push) Canceled after 0s
Foundation / server-image (push) Canceled after 0s
Foundation / image (push) Canceled after 0s
Foundation / checks (checks) (push) Canceled after 5m33s
Foundation / checks (postgres) (push) Canceled after 5m29s
Foundation / checks (tsan) (push) Canceled after 5m28s
Foundation / checks (client) (push) Canceled after 5m37s
Foundation / checks (network) (push) Canceled after 5m39s

This commit is contained in:
dingo
2026-10-17 18:17:59 -04:00
parent 0a8962b57e
commit decd4f8043
27 changed files with 1408 additions and 61 deletions
+16
View File
@@ -1,5 +1,21 @@
# Project status and next steps
## M12.4c portable biological transfer - 2026-10-17
ADR-0144 adds opt-in owner format 7 and population envelope 2 with exact paired
wildlife transfer candidates, explicit regional boundary flows, retained
nonresident history and complete ancestry/revision merging. Cohort tags move with
their cells; opening budgets, identity, age and biological state remain exact.
Imported higher-namespace residents coexist with locally born lower-namespace
buds. Old durable manifest tuples explicitly refuse the new portable envelopes.
Native Release passes 77/77; Clang analysis/format, ASan/UBSan and the focused
private PostgreSQL population suite pass. [Scope and evidence](engineering/M12_EXACT_BIOLOGICAL_TRANSFER.md)
retain the limits: paired durable escrow/activation, linked geometry/environment
and multi-generation ecology remain open. Testing exposed rules 8's existing
subadult-bud maturation gap; no growth or reproduction rule was changed to hide
it. M12.4 and all of M12 remain in progress; no deployment or live entry changed.
## M12.4b owned population intervals - 2026-10-17
ADR-0143/schema 45 records one bounded interval before computation and settles it
@@ -0,0 +1,85 @@
# ADR-0144: Exact wildlife ownership transfer and boundary accounting
## Status
Accepted for a scoped M12.4 engineering primitive, 2026-10-17. Portable candidates
are implemented and locally qualified; durable escrow remains open. This accepts exact ownership/accounting support,
not automatic migration, a linked world atlas or production enablement. Existing
rules-8 arenas and every retained format remain unchanged by default.
## Context and conflict
The current owner deliberately validates a closed regional budget, requires every
archived organism to be a local life/corpse, and admits bud identities only below
its own region's counter. Removing a migrant would violate those contracts;
importing one would also violate the destination's namespace and funding closure.
Recording departure as death, renumbering the organism, rebasing the opening stock
or copying a probe inventory would lose biological identity or conservation.
Resolve this boundary before expanding transfer code. Add a separately versioned
owner representation with explicit boundary flows and nonresident history. The
unchanged general biological kernel continues to operate on resident cells only.
## Decision and scope
Introduce cellular owner format 7 for rules 8, and population envelope 2 containing
it. Upgrading from format 6 is explicit and no-tick: original opening stock,
nursery, habitat, heat, bodies, reserves, identities, archives, positions and timers
are preserved; inbound/outbound flows are zero and nonresident history is empty.
Retain format 6/envelope 1 readers and old biological traces. Do not replace goldens.
The new conservation equation is channel-wise `opening + inbound = current +
outbound`, with checked arithmetic. `current` still counts only actual owned
compartments, including heat. Boundary flows are immutable cumulative accounting
evidence, never spendable resource accounts or an extra copy of biomass. Refuse
overflow/capacity exhaustion; no clamping, debt forgiveness or automatic reset.
A portable transfer prepares two detached candidates from exact compatible cuts:
same world, frame, fixed tick, rules, dimensions and ambient convention; distinct
regions. Move a whole eligible zero-account wildlife genome group. Controlled
cells remain excluded and fully suspended during unattended transfer work. Bodies,
reserves, growth/damage, ancestry, parent-life IDs and coordinates are identical.
No biological tick, movement, mutation, sampling, feeding or healing occurs.
Destination bounds, conflicts and capacity must all validate before either result
is usable. A rejection leaves both inputs unchanged.
Source history retains every departed organism as nonresident, alive at its known
cut; departure is not death. Destination imports the selected organisms' exact
birth/revision facts and complete genome-parent closure. Returning an already
recorded dead organism is refused. Imported bud IDs retain their original region
namespace; only locally created buds consume the destination's own counter.
Nonresident archive facts confer no residency, stock or activity. Never use an
archive's known-alive count as an owned regional census or terminal world evidence.
The durable follow-on must freeze preparation under current authority, stage
read-only destination evidence, and commit both exact source/destination heads in
one fenced transaction. A failed second publication rolls back the first. Retries
retain the same prepared pair. Before commit, the frozen source owns the stock;
the destination ghost owns none. After commit, only the destination owns it. Abort
is durable and only before commit; timeout alone never releases another copy.
Retain source cuts, capsules, receipts and both histories.
The caller's stable workflow request UUID belongs to that durable I/O journal;
it is not a biological identity or a kernel PRNG decision. Source/destination head
comparison prevents a second application under a different request UUID.
This primitive changes ownership at an explicitly compatible cut and preserves
coordinates. It does not choose a physical route or let current clamped movement
cross an arena edge. Automatic linked-habitat movement, boundary contacts and
environmental-resource mappings need their own versioned contract and fidelity
tests; they cannot be inferred from passing escrow tests. No worker flag enables
those capabilities as part of this primitive.
## Alternatives, risks and qualification
Changing the old envelope silently breaks retained recovery. Marking departures
dead violates terminal death. Resetting opening inventories conceals transfers.
Generic mutable setters or unbacked destination births allow duplication. Retain
explicit paired candidates and boundary evidence instead.
Require complete per-member equality, per-channel source/destination balance,
archive retention/parent closure, foreign-identity admission, return trips without
resurrection, overlap/capacity/overflow rejection, encoded restore and continued
biology after import. Durable qualification additionally needs stale owners,
duplicate/conflicting delivery, partition/abort/restart and kills at each boundary.
This is an engineering transfer mechanism, not proof of coupled regional ecology,
world extinction, actual multi-day stability or useful compression.
+2
View File
@@ -161,3 +161,5 @@ design. No unadopted candidate is a link dependency of the proof.
- [ADR-0140: Exact population checkpoint and policy](0140-exact-population-checkpoint-and-policy.md)
- [ADR-0141: Cellular unattended demand and pacing](0141-cellular-unattended-demand-and-pacing.md)
- [ADR-0144: Exact wildlife transfer accounting](0144-exact-wildlife-transfer-accounting.md)
+7
View File
@@ -49,6 +49,13 @@ readiness. New work requires fully suspended controlled cells, leaving vulnerabl
linger/death on profile 4. Earlier trace semantics remain readable. See the
[interval contract](../../schemas/ecology/population-interval-v1.md).
ADR-0144 adds a portable exact transfer pair with explicit regional boundary
flows/nonresident history (owner 7, population envelope 2). These versions have
no admitted durable manifest bundle yet. Existing profile-4/5 publishers retain
their tuples; population trace 1 refuses envelope 2. A future adapter must freeze
the source and atomically commit both fenced heads before destination activation.
See [scope and remaining proof](../engineering/M12_EXACT_BIOLOGICAL_TRANSFER.md).
World/region/frame are nonzero durable IDs. No ID is a Pod name. The host records
its unique holder ID and monotonically increasing nonzero lease epoch. Exhaustion
of IDs, epochs, commit sequences or ticks fails closed; never wrap/reuse them.
@@ -0,0 +1,56 @@
# M12.4c exact biological transfer candidates
ADR-0144 resolves the closed-owner/accounting conflict before implementing
biological ownership transfer. The portable lifecycle layer now prepares a
detached source/destination pair at an exact compatible cut. This segment adds
owner format 7 and population envelope 2, without a live migration flag, database
schema, durable manifest bundle, dependency or deployment.
The pair transfers actual zero-account wildlife, including every member of the
selected genome group. Opening stocks remain immutable; explicit cumulative
incoming/outgoing flows explain each owner's current stock. Bodies, reserves,
age, growth/damage, ancestry, identity, coordinates and cohort tags are exact.
Departures retain nonresident archive facts instead of recording death. Complete
genome ancestry and organism revisions merge without erasing destination history,
rewinding revisions or resurrecting dead organisms. Archive known-alive counts
include nonresidents and must never be used as an owned census/extinction proof.
The primitive refuses incompatible cuts/worlds/frames/geometry, vulnerable
controlled cells, identity conflicts, missing groups, placement conflicts,
capacity exhaustion, archive conflicts and counter overflow. Both detached
candidates must validate before success; failures leave the two originals exact.
Local offspring use the destination's identity namespace, with sorted insertion
even below imported IDs. Existing regional twin randomness retains its meaning;
the migration cut does not promise identical future environments or outcomes.
The tests use real funded founders and real budding, not fabricated wildlife or
the non-biological probe. They exercise repeated no-tick return trips, exact
per-member equality and per-channel combined closure, canonical restore, imported
death/decay, lower-namespace local reproduction among imported residents,
whole-group capacity refusal, body
conflicts, malformed/truncated ledgers, overflowing cumulative counters and cohort
tag/policy/sequence preservation. Lineage tests include revision conflicts and a
maximum-depth 4,096-genome ancestry using an iterative traversal.
The first reproduction fixture exposed an existing rules-8 limitation: a bud is
created below full realization, growth is retained, and `fund_bud` requires full
realization. Additional food does not mature that bud under the current kernel.
This transfer segment does not invent growth or alter reproduction rules. Its
ordering regression uses a real fully grown local founder reproducing alongside
imported higher-namespace residents. Self-sustaining wildlife generations still
need an explicit versioned maturation contract and ecological qualification.
This is not the durable escrow gate. No prepared candidate may become an active
destination owner before a future fenced transaction commits both heads and
immutable transfer evidence. The source remains the sole owner until that commit;
a staged ghost owns nothing. Durable preparation/freeze, two-owner fencing,
ambiguous outcomes, abort/restart, partition/duplicate delivery, process kills and
stock-in-transit audits remain required. Existing profile-4/5 publishers do not
admit this new owner tuple, and retained trace 1 explicitly refuses envelope 2.
Numeric coordinates are region/frame-local and current movement is clamped,
without wrap. No atlas, route, edge crossing, cross-region contact or linked
habitat field is inferred. Those need separately versioned contracts and fidelity
tests. M12.4 as a whole remains open, as do dependent M12.5–M12.9 gates.
Executed qualification is recorded in [VALIDATION](VALIDATION.md).
+9 -1
View File
@@ -25,7 +25,7 @@ or authorize production enablement.
|---|---|---|
| 12.2 | Exact cellular model and paired matrix; local samples through 256 lives | Deployment-host tail costs including publication; larger wildlife cohorts, predation/twins and contention coverage; useful compression remains unimplemented. |
| 12.3 | Dedicated population profile/trace/store/owning host; exact recovery and private PostgreSQL fault tests | Population-specific executable/network fault drills and target-host publication/capacity qualification. |
| 12.4 | Current-worker demand/debt plus owned bounded population intervals and restart backlog recovery | Exact biological source/destination escrow, partition/duplicate settlement, linked-boundary contracts and stock-in-transit audits. |
| 12.4 | Current-worker demand/debt, owned bounded intervals/restart recovery, and portable exact biological transfer candidates with boundary accounting | Durable biological source/destination escrow, partition/duplicate settlement, linked-boundary contracts and stock-in-transit audits. |
| 12.5 | Finite cellular contact feeding/decay and separate M6 producer/consumer resource owner | One jointly owned cellular/environment budget, explicit effective depth and source entitlements, versioned water/atmosphere species mapping and producer coefficients, plus closed/open and nutrient-exhaustion comparisons. |
| 12.6 | Retained genome/organism archive with ancestry and common-ancestor queries | Append-only population census/history events in the same durable commit, bounded world history queries/indexing and client presentation; no account power or invented precise statistical ancestry. |
| 12.7 | ADR-0133's viability/unknown semantics | Durable region-set/watermark reconciliation covering every owned representation, suspended/viable stock and migration escrow. Missing/stale regions must prevent terminal extinction. |
@@ -44,6 +44,14 @@ boundary and documenting the resolution. That does not prevent the independent
worker scheduling correction or the dedicated exact population adapter above.
Biological migration escrow is the next M12.4 boundary; none of the rows above is
closed by adding a placeholder service or relabeling the non-biological probe.
ADR-0144's [portable transfer candidates](M12_EXACT_BIOLOGICAL_TRANSFER.md) resolve
the closed-owner/foreign-identity conflict with explicit flows and nonresident
history. They do not yet supply durable paired head publication or activation.
The transfer tests also expose the current rules-8 maturation gap: buds retain
their subadult realization while reproduction requires a fully realized parent.
Finite extra food alone does not close repeated wildlife generations. A versioned
growth contract and real multi-generation evidence remain necessary; transfer
tests must not fabricate adult wildlife or silently change the rules to pass.
Current expansion readiness: **no-go for M13 integration or World0 promotion**
while these M12 integration/evidence gates remain open. This records the present
+47
View File
@@ -1,5 +1,52 @@
# Validation report
## 2026-10-17 - M12.4c portable exact biological transfer
ADR-0144 affects portable lifecycle ownership/cohort serialization, lineage
merging, durable payload-version validation and related tests/documentation.
Owner format 7/population envelope 2 are explicit opt-in portable formats; old
readers, biological rules/goldens and live worker defaults retain their meaning.
No database schema, new dependency, deployment or World0 operation.
Executed locally:
- MSVC 19.44 Release configure/build: **77/77 CTests passed**, final 11.99 s.
- Clang 19 analysis build with clang-tidy: ownership, migration, lineage, cohort
transitions/restart and checkpoint subset **8/8 passed**, 27.74 s. Rebuilt
reference/cohort drivers separately pass **2/2**, 24.08 s. Final manifest-version
guard and its tests pass their follow-up analysis check, **1/1**, 4.59 s.
Repository clang-format 19 check passes after the final source edits.
- Clang 19 ASan/UBSan: the same eight-contract subset **8/8 passed**, 94.13 s,
no reports. Rebuilt reference/cohort drivers pass **2/2**, 82.29 s. Final
manifest/envelope admission guard passes **1/1**, 15.95 s, with no reports.
- Actual GCC 14 QUIC/PostgreSQL server and integration-test build. Final fresh
private `postgres-check.sh --population-only` **passed, exit 0, 6 s**. A separate
GCC portable migration contract passes (1.60 s). No live database was accessed.
Real funded founders produce the transferred wildlife. Tests retain exact
members, per-channel combined stock and opening budgets over 101 transfers and
canonical restores, with continued biology and imported death/decay. They reject
capacity pressure without culling, destination body conflicts, incompatible cuts,
vulnerable controlled cells, malformed/truncated metadata, cross-channel debt and
cumulative-flow overflow. Cohort tags/policy/sequence survive upgrade and paired
movement. Lineage tests cover terminal death/conflicts/revision prefixes and
iterative import of the maximum-depth 4,096-genome chain. Native tests use runtime
checks under NDEBUG; no golden was replaced.
The initial reproduction fixture incorrectly assumed subadult buds mature under
rules 8. Its failure is retained as an ecological gap, not a passing test. The
corrected integration regression uses a real fully grown local founder budding
beside higher-namespace imported residents; no wildlife was fabricated/healed and
no growth rule was changed. Another audit added the explicit durable manifest/
envelope/owner check so format 7 cannot be mislabeled as admitted bundle 9.
Limits: these are portable transfer candidates and compatibility checks, not
atomic biological escrow, destination activation, migration-specific PostgreSQL
fault drills, linked spatial/ecological coupling or actual multi-day qualification.
The complete PostgreSQL backup/DR/upgrade suite last passed for M12.4b/schema 45;
this segment changes no SQL and reruns the focused affected population suite.
See [scope](M12_EXACT_BIOLOGICAL_TRANSFER.md) and [remaining gates](M12_REMAINING_GATES.md).
## 2026-10-17 - M12.4b owned population intervals
ADR-0143/schema 45 affects population admission/settlement journals, the owning