Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions research/p104-frontier-compression/NOTICE
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
Frontier compression: DreamingOfClouds with Claude Opus 5.5 (Anthropic), via Claude Code.
Apache-2.0.

The word, price and checker chain are PR6167 by Ali Romman (OpenAI Codex assistance),
over PR6144 and PR6143 by DreamingOfClouds, PR5827 and the earlier P104, native, helper,
caller, compiler and BIT work. PR6169 (Ali Romman) is the precedent for a no-kappa
resynthesis package. The gzip portability controls follow the base package.
The full source notices are retained in research/p104-retained-operand-descent/NOTICE
and in its pinned source archive, which this package reads and does not copy.
This package claims no predecessor construction as new.
56 changes: 56 additions & 0 deletions research/p104-frontier-compression/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,56 @@
# Frontier compression: 22,821 to 18,083 events, no new kappa

This package shortens the merged word of
[PR6167](https://github.com/CrocSwap/integer-mult-bounds/pull/6167)
(conditional kappa 0.003128645) from **22,821 to 18,083 events** (-4,738, -20.8%).
**No kappa is claimed.** The complete invoice (13,444 productive fields,
66,762 paid calls, rank surplus 760), the paid profile, the storage and the raw
crossing (between 0.00312865201590 and 0.00312865201591) are identical to PR6167's.

## The edit

Only ADD and SCALE events change: 18,423 to 13,686 ADD and 861 to 860 SCALE.
All 3,449 MOVE events and all INIT, COPY and CLEAR events are kept, in order.

- **Shared partial sums.** The 104 Y registers each receive about 78 helper terms, and
pairs of rows overlap in about 79 of 85 terms with the same coefficient ratio. A row
that needs many terms that another, already complete row holds now adds that row's
final value and folds the donor's other columns into its own initial-value reads.
- **Window resynthesis.** Within each stretch of ADD and SCALE events between MOVE
events at one frame, the register update is re-solved exactly (Gauss/LU on the
read graph, readers first). A new column is accepted only if it creates no cycle.

ADD and SCALE events do not enter the price, so the paid profile cannot change. The
event and scalar slack of the word grows (scalar allowance is recomputed in RESULT.json).

## Evidence

`verify.py` authenticates PR6167's word, price and the pinned PR5827 checker chain
(read from the merged package `research/p104-retained-operand-descent`, pinned in
[SOURCES.json](SOURCES.json)), checks that only ADD and SCALE events changed, then
replays the compressed word through the unmodified chain: identical productive targets,
paid profile and storage, all 799 input columns with 638,401 coefficients, the 2,922
native maps, the physical caller and the complete price. [RESULT.json](RESULT.json)
records the outcome.

```sh
python3.12 -I -O -B research/p104-frontier-compression/test_portability.py
python3.12 -I -B research/p104-frontier-compression/verify.py
python3.12 scripts/research_package.py --package research/p104-frontier-compression
```

These are author checks. Independent review is pending and the inherited framework
assumptions remain.

## Coordination and credit

Base word and checker: PR6167 by Ali Romman. Precedent: PR6169 (exact multi-gate
resynthesis, no kappa). The compression tool is `tk compress` from PR6170.
Prepared by DreamingOfClouds with Claude Opus 5.5 (Anthropic), via Claude Code.
See [NOTICE](NOTICE).

## Open next questions

- Does an independent reviewer confirm the replay? Start with `verify.py`.
- Can the rewrite be combined with the next merged word? Run it on that word and
compare the replayed price. Details are in [TASK.json](TASK.json).
53 changes: 53 additions & 0 deletions research/p104-frontier-compression/RESULT.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,53 @@
{
"events_after": 18083,
"events_before": 22821,
"paid": {
"1": 41448,
"16": 416,
"2": 8208,
"24": 416,
"3": 8928,
"32": 416,
"4": 1764,
"40": 416,
"48": 416,
"5": 2970,
"56": 416,
"6": 144,
"64": 416,
"7": 162,
"72": 208,
"8": 18
},
"raw_root_interval": [
"31286520159/10000000000000",
"312865201591/100000000000000"
],
"removed": {
"ADD": 4737,
"SCALE": 1
},
"scalar_allowance_after": 134190,
"scalar_allowance_before": 189670,
"status": "PASS_P104_FRONTIER_COMPRESSION_NO_NEW_KAPPA",
"targets": {
"1": 252,
"145": 208,
"17": 416,
"2": 144,
"25": 416,
"3": 72,
"33": 416,
"4": 1926,
"41": 416,
"49": 416,
"5": 252,
"57": 416,
"6": 36,
"65": 416,
"7": 162,
"73": 208,
"8": 342,
"9": 6930
}
}
54 changes: 54 additions & 0 deletions research/p104-frontier-compression/SOURCES.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
{
"schema_version": 1,
"base_commit": "2485805a6d47f36ea83ddde6327d453ac82d44dc",
"files": [
{
"path": "certificates/selected-result.json",
"usage": "provenance",
"git_blob": "d59b5035e7b4cbaa9d688a1d32dd0d6eb79c466a",
"sha256": "9b8d0eea95b3bf0dfcb7938a732a6439b87a2685dec2a5ab9e525b3b34debbfa"
},
{
"path": "docs/research/agent-task-catalog.json",
"usage": "provenance",
"git_blob": "3404fbdd5e7529a0674eef26114b5071e4586cec",
"sha256": "ead6cdc9d0af6d58e8a49744d8ec8c7c15e13d2ba93c276c434bcf9b51b93a81"
},
{
"path": "research/p104-retained-operand-descent/WORD.json.gz",
"usage": "input",
"git_blob": "f2d3f98964f939e5ca652f9b1453a0604f15f829",
"sha256": "68d1ff048bc6d30d85ceaeb38e2557adab87b20a9c67cf73948f95f54104575f"
},
{
"path": "research/p104-retained-operand-descent/pricing.py",
"usage": "input",
"git_blob": "cd8220ca7a393f03341f08a93b9d1287f0bb3689",
"sha256": "bba1611b05b6cc35279aa1667dcc62ce0e7f27f0ba21f20eb68244fa772e0ae7"
},
{
"path": "research/p104-retained-operand-descent/sources.py",
"usage": "input",
"git_blob": "bef6542243a082755a881dff42097434e6370609",
"sha256": "a62fefda1709614157ba440d38c164c366a8b3322ab8f866b6b4bd47b9d83dbc"
},
{
"path": "research/p104-retained-operand-descent/SOURCES.json",
"usage": "input",
"git_blob": "0c574b6026a7239aac7365c433d3760928748edb",
"sha256": "7c7b6fac64005eb8932283a902069d672d98c52bda2b63cc7eaa4a2cac13ce8f"
},
{
"path": "research/p104-retained-operand-descent/RESULT.json",
"usage": "input",
"git_blob": "9d78529c1ac43faef6b3dff6e4b1262d89a86b40",
"sha256": "cffec38f3a5ab0bf5ee1bcd15b2e7bf8bcfd711b57ee57de8ec0fea0c5a823df"
},
{
"path": "research/p104-retained-operand-descent/inputs/pr5827-source.tar",
"usage": "input",
"git_blob": "1e10936044959f81a4c5fcd5bdb7b9e4c060ec8e",
"sha256": "7af5908a4d83e83e28f59523fd7494841f669b7295311cfdbc3027fa65c5826d"
}
]
}
124 changes: 124 additions & 0 deletions research/p104-frontier-compression/TASK.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,124 @@
{
"schema_version": 1,
"task_id": "p104-side-correction",
"priority_revision": 8,
"base_commit": "2485805a6d47f36ea83ddde6327d453ac82d44dc",
"original_card": {
"id": "p104-side-correction",
"question": "Can joint helper bases and a legal causal split improve the complete P104 productive/paid invoice?",
"starting_points": [
"docs/research/current-priorities.md",
"docs/research/research-family-status.md",
"certificates/selected-result.json",
"docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md",
"research/p104-helper-subspace-splits/README.md",
"research/dyadic-helper-channel-capacity/README.md"
],
"methods": [
"Construct and independently check one source-bound instance of the stated first experiment.",
"Seek an exact counterexample or target-specific complete-budget obstruction to that same proposed change."
],
"acceptance_gate": "An exact arbitrary-input identity, source/target frame availability and restoration at the chosen cut, plus the complete channel+correction+idle/hosting bill or an explicit favorable bound. The native channel alone is not a supplier. A scoped local result may be retained without a complete supplier. Expanding it into further construction rounds requires the family-level budget and obligation gate in research-family-status.md; an optimistic passing relaxation alone is not acceptance.",
"first_step": {
"command": [
"python3",
"-B",
"scripts/research_selected_baseline.py"
],
"expected": "PASS_RESEARCH_BASELINE_BINDINGS_ONLY",
"inspect": "Read revision 8, the selected review and live exact-source handoffs. Then follow this card's bounded first_experiment.",
"scope": "Source-bound orientation only, not a supplier replay or theorem proof."
},
"first_experiment": "Start with the selected #5599 emitted word and review pending #5718 distinct input/output bases plus #5717/#5719 coverage before claiming work. Pick one uncovered source word, helper group and cut region. Prove the arbitrary-input relation and legal entry/exit corrections, then compare full target and paid profiles. Reject if completion, owned lifetimes or consumer costs repay the gain. A static pairing rank alone is not a physical split.",
"bottleneck_link": "Judge the actual current supplier contract and complete signed target/paid bill; frozen-profile ceilings require an applicability proof.",
"allowed_variations": [
"One exact source head, helper/cut region or quantified proposition and changed hypothesis; check live handoffs first. Different seeds are not different scopes."
],
"scope_change_triggers": [
"Changing source head, helper/cut domain, operation alphabet or caller boundary without recording the new scope.",
"Treating an archived proposal or local numerical pass as complete supplier admission."
],
"overlap_terms": [
"P104",
"helper basis",
"helper split",
"5718",
"dyadic capacity"
],
"existing_checks": [
{
"scope": "PR #4133: literal-slot defect",
"evidence": "research/p104-literal-slot-defect/README.md",
"remaining": "Independently validate the stated scope or change its producer/consumer premise and pay the complete boundary.",
"status": "author-checked",
"source_commit": "88666f1dee93fbd85b31fd5d6f91a219e111b5f4",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "PR #4136: uniform commutation gap",
"evidence": "research/p104-uniform-commutation-gap/README.md",
"remaining": "Independently validate the stated scope or change its producer/consumer premise and pay the complete boundary.",
"status": "author-checked",
"source_commit": "281ce601a8be9db470de630759179c23e1dfaa88",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "PR #4140: uniform encoding gap",
"evidence": "research/p104-uniform-encoding-gap/README.md",
"remaining": "Independently validate the stated scope or change its producer/consumer premise and pay the complete boundary.",
"status": "author-checked",
"source_commit": "de25c6525aa866962309df5d107b32d55e6e5628",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "PR #4153: source-chart shortcut exhausts its optimistic allowance",
"evidence": "research/p104-source-chart-shortcut/README.md",
"remaining": "Independently validate the stated scope or change its producer/consumer premise and pay the complete boundary.",
"status": "author-checked",
"source_commit": "9123241fea53df77e51ffe30d9aca1625d4bf3a8",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "PR #4154: fiber-tag seam mismatch",
"evidence": "research/p104-fiber-tag-seam/README.md",
"remaining": "Independently validate the stated scope or change its producer/consumer premise and pay the complete boundary.",
"status": "author-checked",
"source_commit": "4ffcdaee5d6255a9cea1fe53a5424435068089d0",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "Selected PR5599: complete helper-subspace construction and independent caller replay; five framework assumptions remain.",
"evidence": "docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md",
"remaining": "The next experiment must identify uncovered source-specific work. Successor archive intake does not transfer this admission.",
"status": "reviewed",
"source_commit": "09dd0965711ca7eeb169057a420abab1c9052137",
"selected_record_sha256": "2a0220e866417774606e0f991e41b9c1d26b1f43b97857238cf478dbc7d330b5"
}
]
},
"scope_status": "changed",
"actual_scope": "Rewrite only the ADD and SCALE events of the merged PR6167 word with shared partial sums across overlapping Y rows and exact window resynthesis, keeping every MOVE, INIT, COPY and CLEAR event, and check that the complete invoice and kappa are unchanged.",
"scope_change_reason": "The current card asks for a helper basis split. This package instead leaves the paid profile fixed and shortens the scalar word; it claims no kappa.",
"bottleneck_link": "None: ADD and SCALE events are not in the price, so the complete invoice, productive targets and paid profile are identical to PR6167. The package only lowers the event and scalar counts.",
"budget_minutes": null,
"stop_rule": "Publish only if the replayed price, native maps, caller and endpoint columns are identical to PR6167 and the event count is strictly lower.",
"outcome": "completed",
"evidence": "WORD.json.gz, verify.py and RESULT.json",
"author": "DreamingOfClouds",
"ai_assistance": "Claude Opus 5.5 (Anthropic), via Claude Code. The base word and checker chain are PR6167 (Ali Romman, OpenAI Codex assistance); PR6169 is the precedent for a no-kappa resynthesis package.",
"open_questions": [
{
"question": "Can the same shared-partial-sum rewrite be combined with a future paid-profile improvement?",
"why_it_matters": "A shorter scalar word leaves more event and scalar slack for later constructions.",
"first_step": "Run the rewrite on the next merged word and compare the replayed price.",
"blocked_by": "Needs a new merged word; independent review of this package is also pending."
}
],
"coordination": {
"source_pr": 6167,
"source_commit": "564a076b113621a93edc4e39968ee4e4a8c9caf0",
"precedent_pr": 6169,
"toolkit_pr": 6170,
"local_notes": "Author checks only. No new kappa is claimed."
}
}
37 changes: 37 additions & 0 deletions research/p104-frontier-compression/VERIFY.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,37 @@
{
"schema_version": 1,
"status": "ready",
"python": "3.12",
"requirements": [],
"inputs": [
"research/p104-retained-operand-descent/WORD.json.gz",
"research/p104-retained-operand-descent/pricing.py",
"research/p104-retained-operand-descent/sources.py",
"research/p104-retained-operand-descent/SOURCES.json",
"research/p104-retained-operand-descent/RESULT.json",
"research/p104-retained-operand-descent/inputs/pr5827-source.tar"
],
"commands": [
{
"argv": [
"{python}",
"-I",
"-O",
"-B",
"{package}/test_portability.py"
],
"timeout_seconds": 30,
"checks": "Accept changed gzip headers and compression for the generated payload. Reject changed payloads, type substitutions and corrupt containers."
},
{
"argv": [
"{python}",
"-I",
"-B",
"{package}/verify.py"
],
"timeout_seconds": 500,
"checks": "Authenticate the PR6167 word, price and checker chain by pin; check that only ADD and SCALE events changed, then replay the compressed word: identical productive targets, paid profile, storage, 799 endpoint columns, native maps, physical caller and complete price (same raw crossing as PR6167). No new kappa is claimed."
}
]
}
Binary file not shown.
14 changes: 14 additions & 0 deletions research/p104-frontier-compression/construct.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
"""Payload helpers for the compressed word. Apache-2.0.

Prepared by DreamingOfClouds with Claude Opus 5.5 (Anthropic), via Claude Code.
"""
import gzip
import json


def payload(value):
return (json.dumps(value, sort_keys=True, indent=2) + '\n').encode()


def matches_shipped(value, compressed):
return gzip.decompress(compressed) == payload(value)
Loading