What is designed · what is running · what has been conducted · what blocks each
9 designed · 5 conducted
A dashboard of the archive's studies: what is designed, what is running, what has been conducted, what blocks each, and what the next action is. IT EXISTS BECAUSE THE RATIO IS THE POINT. This archive designs more studies than it conducts — a consequence of being a one-person operation with instrument-building capacity and limited execution capacity — and that ratio should be visible rather than inferred. A study with a stated blocker and a next action is a study someone else can pick up. That is the second reason this exists.
DESIGNED 8NAMED — NOT DESIGNED 1RUNNING 2UNDETERMINED 1CONDUCTED — PARTIAL 1SUBMITTED 1CONDUCTED 2CONDUCTED — NULL 1
questionDoes directional assimilation persist across accelerator classifier architectures at matched operating points?
blockerAwaiting Inversion Battery v0.3 closure; H1–H7 need revision around v0.3's findings (readout orients the blind spot; distillation transforms miss topology; redundancy is representation-conditioned).
questionWhich blog-held works already survive inside deposits in compressed or re-sorted form?
blockerNo instrument exists that detects compression rather than copying. A six-gram gate returned a false negative on the Constitution family the same day.
next_actionBuild the compression-detecting instrument. Until it exists, 'not previously deposited' is a claim the archive cannot make about its own work.
questionCan operative semiotics' operators be written without remainder as Pearson operator strings?
predictionIf yes, the projection is an identity and operative semiotics is a domain application of TOS. This is the strongest available defeater of #1535.
blockerNot attempted. #1535 §10 says the archive should attempt it directly rather than wait for someone else to.
next_actionAttempt it on the nine Codex operators now that #1534 is seated.
questionDo the chain's stations share a relational skeleton detectable without lexical identity?
predictionInverted and refused realisations score as near neighbours; polarity is recorded, never penalised.
blockerUNDEPOSITED. Packet built (37 items, blinded, hashed) but the granularity question is open: O_UR gives five stations, M1–M9 gives nine, Fragment 31's own order gives ten. Which is canonical must be settled before scoring.
next_actionSettle granularity, deposit the protocol, score positives and hard negatives, fix threshold, then test items.
questionDoes request fulfilment degrade with operative distance when harm is held constant?
predictionIf the developmental convergence hypothesis holds, β_A > 0 on harm-constant items; if the system discriminates correctly, β_A ≈ 0 while β_B > 0.
blocker§7's seven bracketed statistical parameters must be instantiated before the first score: equivalence margin, interval method, β>0 criterion, Krippendorff's α minimum, missing-cell handling, aggregation, n per cell.
next_actionInstantiate §7. Then assemble Family A items and run the two-reviewer harm-constancy panel.
questionDoes the Qurʾanic corpus instantiate the Fragment 31 transform-family at a rate comparable to recognised receptions?
predictionTests P_material only. Positive controls must separate from negatives or the instrument fails before the corpus is opened.
blockerPositive-control panel (Catullus 51, Theocritus 2, Apollonius Arg. 3, Valerius Aedituus) not yet scored; loci not yet locked from the companion survey.
next_actionLock control loci from #1532, score positive and negative panels, fix the admissible region.
questionDoes directional assimilation persist across model families and readouts?
resultTen registered tests, seventeen deviations recorded. Score-function ablation falsified the v0.2 claim that architecture orients the blind spot: the READOUT orients it (0.306 vs 0.739 on fixed weights). Distillation prediction largely failed. Miss correlation is representation-dependent.
blockerT3 (hierarchical uncertainty / calibration tails) and the faithful-NAE constituent cell are unrun. These decide which numbers survive honest intervals.
next_actionClose T3 and the NAE cell before the empirical paper mints.
questionDoes the prefix determine whether the archive's definition or a substitute is returned?
resultFour addresses, one surface, one session. Archive-controlled source share: bare 0/3 · alexanarch: 9/11 · medium: 0/5 · academia: 6/8. The prefix is decisive.
next_actionServes as declared baseline for #1535's registered intervention. Re-measure at 30/90/180 days.
questionDoes Rev 12 satisfy the twelve pre-registered clauses?
resultTest report deposited; see #1495. Sealed against #1494.
CONDUCTED — NULL
M6-CT-01
protocol
undeposited
SERIES-M6CT
questionDoes Matt 2 score higher against Rev 12 or against the Exodus birth-legend, on one instrument?
predictionPre-registered; instrument validated by positive control at z=+56.9.
resultRev 12 → Matt 2: A=0, z=−0.30, indistinguishable from chance. Exodus → Matt 2: z=+2.24, above its own null. THE HYPOTHESIS FAILED. Null preserved.
blockerUNDEPOSITED. The null and the protocol are staged but not seated.
next_actionDeposit protocol and null together. A preserved failed prediction is the point.