Typed input
Numbers, Hebrew text, dates, times, identities, roles and sources.
RTM LAB · RESEARCH PLATFORM
From documented records to a network that can be reconstructed and tested.
RTM Lab takes records, text, identities, dates and times; reconstructs Evidence, finds structures in Discovery, builds a dependency-aware network, and runs null-model tests to compare the observed network with alternative worlds under the same search freedom.
This page is public and indexable. The compute environment and private API remain access-controlled.
RTM Lab
R46● API ONLINE
01 Input02 Evidence03 Discovery04 Network05 QA06 Statistics07 Export
INPUT MATRIX
One finding, modeled as one information system
Text / contentDateTime
Identity / roleCore anchorSource
FROM RAW INPUT TO AUDIT
Seven stages — from structured input to a full statistical audit
The engine separates finding relations, building topology and testing extremeness. You can see not only what was found, but where it came from, what it depends on, and what happens when null worlds receive the same search contract.
Reconstruction
Legal paths are checked against frozen targets with an audit trail per closure.
Search
New structural relations are searched under the same locked search rules.
Topology
Closures become a network of nodes, families, bridges, convergence and dependency.
External QA
Manifest and validation tools separate source, representation, computation and interpretation.
Null & audit
Alternative worlds run through the same search contract and are compared structurally.
Export
Inputs, network, provenance and test settings can be retained for reproducibility.
What the engine actually measures
The unit of analysis is the architecture — not a single numerical match
A network-level result asks how many independent evidence units remain after deduplication, how many distinct families connect, how many anchors are covered, and how the largest connected component is organized.
≡
Dependency-aware
Algebraically or semantically dependent paths are not automatically counted as independent evidence.
⌘
Role-aware
Sender, recipient, date, time, text and core anchors can receive different event-model rules.
◇
Network-first
The structural statistic considers component size, family diversity, anchor coverage, arity and convergence.
↻
Same search freedom
Null worlds are not handed the observed equation. They receive the same rules and search from scratch.
✓
Audit trail
Fixed variables, resampled variables, transformations, seed, world count and observed signature are preserved.
T
Tanakh lookup
A separate tool searches verses, words and contiguous phrases without automatically turning the result into Evidence.
Search profiles
Three search profiles, three levels of freedom
RTM Lab keeps Standard, Strict and Super-Core profiles separate, including at the statistical stage.
STANDARD
Standard
Full search under the active search contract.
STRICT
Strict
A fixed profile with reduced search freedom.
SUPER-CORE
Super-Core
The hardest core of the locked search rules.
Important: profile-score p-values and the structural-network p-value answer different questions and are reported separately rather than multiplied.
R44 · Structural Network Event Null
The event-level structural test.
R44 freezes the rules of the game, not the observed equation. The engine declares what is fixed and what may vary, generates an alternative event world, runs the same discovery process, and compares that world's best network against the observed signature.
Time-onlyExhaustive scan of all 1,440 legal minutes when the rest of the event is fixed.
Time + DateDate and time are resampled according to their typed legal spaces.
PrimaryText + date + time vary while identity/core slots can remain fixed by schema.
Population identitiesA separate mode requiring an explicit population model rather than an implicit guess.
Why the Null matters
The question is not “did a closure exist?” but “how much freedom existed to find one?”
When many values, representations and operations are available, chance can produce attractive relations. RTM Lab attempts to price that search freedom inside the null model.
1
Same Search ContractObserved and null worlds use the same search rules.
2
Max-over-searchEvery null world may return the strongest network it finds.
3
Dependency preservedRepresentations derived from the same source do not become artificial independent evidence.
4
Exhaustive when feasibleSmall finite spaces are enumerated rather than approximated by Monte Carlo.
What a small p-value can say
- Under the specified null, networks at least this strong were rare.
- With frozen rules and anchors, it quantifies extremeness relative to that model.
- Sensitivity modes show which parts of the event carry the anomaly.
What it cannot say by itself
- It does not independently prove a physical mechanism or causal theory.
- It cannot erase undocumented retrospective search freedom.
- Zero hits in N simulations is not p=0; resolution and uncertainty remain.
Designed for auditability
Built so a finding can be inspected, not merely admired
RTM Lab keeps raw inputs, computation, topology and statistical inference conceptually separate, making it easier to reconstruct what actually contributed to a result.
A
Typed variables
Each anchor can be fixed, variable, dependent or derived, with an explicit role.
B
Provenance
Seed, ranges, null model, world count, search contract and observed signature are recorded.
C
Export
Exports support reconstruction outside the session in which the result was discovered.
Post-audit AI report
After the audit: an RTM AI report on the complete finding
Once Find and statistics have completed, RTM Lab can pass RTM AI a frozen Audit package containing the inputs, anchors, closures, network structure, provenance and Null results. The AI layer does not replace the deterministic engine; it analyzes the already-computed finding and returns a structured research report.
Evidence→Network→Statistics→RTM AI Report
FAQ
Frequently asked questions
Is RTM Lab a gematria calculator?
No. Gematria is one representation type. The central object is an information network made of anchors, roles, paths, dependencies, topology and null-model tests.
Why is the engine private?
This overview is public. The full compute environment and its API are access-controlled, separating public methodology from private research compute.
Does RTM Lab automatically “prove” RTM?
No. It provides infrastructure for reconstruction, audit and extremeness testing under declared models. Moving from statistical anomaly to a theoretical claim requires additional evidence, confirmatory prediction and external replication.
What is the difference between Profile Score and Full Network?
Profile Score summarizes search characteristics under a profile. Full Network evaluates the structural signature after dependency deduplication. They answer different questions and are reported separately.
Private compute · Public methodology
Ready to enter RTM Lab?
The public page explains the methodology; the full engine remains a private, authenticated research environment.