scriptoriumadmin/hypertwist
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Reconcile reference benchmark operator board

Browse source
This commit is contained in:
axiomlogicnexus 2026-05-27 22:54:38 +02:00
parent 25b9d3d785
commit 30c163f52a
3 changed files with 88 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

75
docs/HYPERTWIST_REFERENCE_BENCHMARK_AND_DISCARD_OPERATOR_RECONCILIATION_2026-05-27.md Normal file
View file

@ -0,0 +1,75 @@
# HyperTwist Reference Benchmark And Discard Operator Reconciliation - 2026-05-27
## Status
This document is the operator-level reconciliation note for HyperTwist's
non-live `0R-E` reference, benchmark, oracle, and discard set.
It answers one narrow question:
- does the current canon still need to speak about these rows only through
packet-era `0R-E` shorthand, or do the later `2026-05-27` per-row hierarchy
and discard notes now control operator truth
Result:
- the later per-row notes already control the real operator truth
- the non-live board count does not change
- no new implementation packet is justified from this pass
- the stale gap is only board wording and operator shorthand
## Source basis
- `C:\HyperTwist\docs\HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md`
- `C:\HyperTwist\docs\HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md`
- `C:\HyperTwist\docs\HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md`
- `C:\HyperTwist\docs\REPO_LICENSE_TRACKING.md`
- `C:\HyperTwist\docs\v6_5_deep_manual_pack\HyperTwist\FEATURE_REGISTRY.md`
- `C:\HyperTwist\docs\HYPERTWIST_GUIDED_CLASSIC_CUBE_PEDAGOGY_AND_LESSON_STATE_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_GUIDED_CLASSIC_CUBE_TASK_PLANNER_AND_FOCUS_PLAYBACK_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_BLINDFOLD_MEMO_ATTEMPT_AND_VALIDATION_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_CODEXR_OFF_DOMAIN_XR_COLLABORATION_DISCARD_CONFIRMATION_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_RIFTSKETCH_IMMERSIVE_LIVE_CODING_DISCARD_CONFIRMATION_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_CLASSIC_SOLVER_ORACLE_AND_IMPLEMENTATION_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_CSTIMER_BENCHMARK_ORACLE_AND_REOPEN_CRITERIA_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_LIGHTWEIGHT_CLASSIC_CUBE_SIMULATOR_DUPLICATE_RETIREMENT_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_SMARTCUBE_PRACTICE_REVIEW_AND_SRS_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_LARGE_N_CLASSIC_CUBE_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_CLASSIC_CUBE_STATE_HISTORY_AND_BEGINNER_SOLVER_HIERARCHY_RECONCILIATION_2026-05-27.md`
- `C:\HyperTwist\docs\HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md`
## Current operator truth
The `0R-E` set is still non-live and still outside the active implementation
board.
That part does not change.
What has changed is the precision of the row meanings:
- the pedagogy rows are no longer just generic clean-room-later placeholders
- the solver rows are no longer just generic oracle placeholders
- the discard rows are no longer justified only by packet-era prose
- the smartcube, timer, large-`N`, state/history, and research rows now each
have narrower explicit retained meaning
## Required board posture
The state board should now describe the `13` non-live rows this way:
- keep the live state and reset-lane counts unchanged
- replace packet-era shorthand with the later narrower retained or discard call
- name the surviving retained slice for each benchmark row
- name the discard-confirmed posture for the three discard rows
- do not advertise any default new implementation opening from this section
## Final call
The correct operator posture is:
- preserve the `13` non-live rows as already-closed benchmark, oracle,
clean-room-later, research, comparison, or discard lanes
- treat the later `2026-05-27` row-specific notes as the real authority for
what each row still means
- do not reopen implementation work from this section unless a later first-party
owner gap is explicitly proven

2
docs/HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md
View file

@ -256,7 +256,7 @@ Packet result:
Most important scheduling result:
- the reference, benchmark, reserve, and discard backlog is now closed as a `Phase 0R` evaluation class
- the reference, benchmark, reserve, and discard backlog is now closed as a `Phase 0R` evaluation class, and the later `2026-05-27` per-row hierarchy and discard notes now control operator truth for the surviving non-live rows
- `cs0x7f/cstimer` remains the primary restrictive oracle for competition-grade timer semantics, rolling stats/`DNF` math, session interchange, scramble-registry breadth, and `GAN`/smart-device behavior; reopen only through a dedicated clean-room timer-pattern handoff if a specific gap is proven
- `brownan/Rubiks-Cube-Solver` remains the primary compact solver oracle and clean-room-later heuristic-table benchmark
- `efrantar/rob-twophase` remains a secondary solver oracle for metric and performance comparison

24
docs/HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md
View file

@ -362,19 +362,19 @@ Count: `13`
| Repo | License | Prior bucket | Prior action | Current live state | Reset lane | Next step |
| --- | --- | --- | --- | --- | --- | --- |
| `alinen/cube` | `no explicit license visible` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a no-license clean-room planner benchmark for task segmentation, focus cues, and stepwise teaching flow; do not incorporate source directly. |
| `ambisinister/blindsolve` | `no explicit license visible` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a no-license BLD memo benchmark for hide/reveal, letter-pair rehearsal, and feedback cadence; keep it narrow and clean-room only. |
| `aMonteSl/CodeXR` | `GPL-3.0-only` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Discard from the active retained set; its XR collaboration value is off-domain and already superseded by stronger retained rows. |
| `AviKaufman/Rubix-cube-trainer` | `All Rights Reserved` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as the primary all-rights-reserved guided-pedagogy benchmark for clean-room lesson-state and stepwise solve flow design. |
| `brianpeiris/RiftSketch` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Discard from the active retained set; immersive live-coding is off topic to HyperTwist's retained product scope. |
| `brownan/Rubiks-Cube-Solver` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as the primary compact solver oracle and clean-room-later heuristic/search benchmark; use for correctness, not donor code. |
| `cs0x7f/cstimer` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as the primary restrictive oracle for competition-grade inspection or multi-phase timing semantics, rolling stats/average or `DNF` math, session/import-export expectations, scramble-registry breadth, and `GAN`/smart-device behavior; reopen only through a dedicated clean-room timer-pattern handoff if a specific first-party gap is proven. |
| `efrantar/rob-twophase` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a secondary solver oracle for metric, pruning, and performance comparison; do not treat as donor code. |
| `MathewKJ2048/Rubiks-cube-simulator` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Discard from the active retained set as an inferior duplicate of stronger pedagogy, planner, and history benchmarks. |
| `alinen/cube` | `no explicit license visible` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` guided task-planner hierarchy clarification are closed. Retain only as a no-license clean-room planner benchmark for task segmentation, focus cues, queued move demonstration, and stepwise teaching flow; do not incorporate source directly. |
| `ambisinister/blindsolve` | `no explicit license visible` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` blindfold memo hierarchy clarification are closed. Retain only as a no-license clean-room BLD memo benchmark for conceal/reveal validation, letter-pair rehearsal, and feedback cadence; keep it narrow and clean-room only. |
| `aMonteSl/CodeXR` | `GPL-3.0-only` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` `CodeXR` discard-confirmation note are closed. Discard from the active retained set; its XR collaboration value is off-domain and already superseded by stronger retained rows. |
| `AviKaufman/Rubix-cube-trainer` | `All Rights Reserved` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` guided pedagogy hierarchy clarification are closed. Retain only as the primary all-rights-reserved clean-room guided-pedagogy benchmark for named beginner milestones, oversatisfied-step explanation, bounded step help, and manual-turn/tutorial coherence. |
| `brianpeiris/RiftSketch` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` `RiftSketch` discard-confirmation note are closed. Discard from the active retained set; its immersive live-coding value is off-topic and already superseded by stronger retained browser/XR, capture, and export rows. |
| `brownan/Rubiks-Cube-Solver` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` classic solver hierarchy clarification are closed. Retain only as the primary compact solver correctness and heuristic-table oracle; use for correctness and clean-room planning, not donor code. |
| `cs0x7f/cstimer` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` `cstimer` oracle clarification are closed. Retain only as the primary restrictive timer oracle for competition-grade inspection or multi-phase timing semantics, rolling stats or `DNF` math, session interchange, scramble breadth, and `GAN`/smart-device behavior; reopen only through a dedicated clean-room timer-pattern handoff if a specific first-party gap is proven. |
| `efrantar/rob-twophase` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` classic solver hierarchy clarification are closed. Retain only as the secondary metric, pruning, robot-execution, and performance comparator beside `brownan`; do not treat as donor code. |
| `MathewKJ2048/Rubiks-cube-simulator` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` duplicate-retirement note are closed. Discard from the active retained set as an inferior duplicate of stronger pedagogy, planner, and history benchmarks. |
| `poliva/cubedex` | `no explicit license visible` | Locked Strategic Donor | `integrate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 1R` plus the `2026-05-27` smartcube practice/review/SRS hierarchy clarification are closed. Retain only as restrictive comparison context for smartcube-aware practice-shell behavior, review/SRS UX, recognition-versus-execution timing presentation, and local stats/history comparison; do not treat it as a live owner or a default clean-room next row. |
| `ShellPuppy/RCube` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a large-`N` centers/edges/parity benchmark and later clean-room research input, not as an active product donor. |
| `vwcwong/CubeSim` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a readable state/history and beginner-solver decomposition benchmark for later clean-room use. |
| `yakupbilen/drl-rubiks-cube` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as an MIT research benchmark for `54`-sticker transition encoding, `ADI` state generation, batched neural `A*` search, and offline solver experimentation; do not treat it as a live camera/perception owner or a near-term product donor. |
| `ShellPuppy/RCube` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` large-`N` hierarchy clarification are closed. Retain only as a large-`N` centers/edges/parity and virtual-rotation benchmark for later clean-room research; do not treat it as an active product donor. |
| `vwcwong/CubeSim` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` classic state-history hierarchy clarification are closed. Retain only as a readable state-history and beginner-solver decomposition benchmark for later clean-room use. |
| `yakupbilen/drl-rubiks-cube` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 0R-E` plus the `2026-05-27` learned-search research clarification are closed. Retain only as an MIT research benchmark for `54`-sticker transition encoding, `ADI` state generation, batched neural `A*` search, and offline solver experimentation; do not treat it as a live camera or perception owner or a near-term product donor. |
## Implementation rule going forward