What this is
This page is a visual demo of the Adaptive Intelligence Visual Atlas v1.0. It does not add a new theory. It turns the atlas diagrams into readable, labeled, interactive SVG figures, adds working calculators where the source defines gauges or math, and makes the atlas own completeness and QA rules checkable on the page.
The source states, in its own words, that it distinguishes thread-derived concepts, proposed hypotheses, engineering designs, and unvalidated claims, and that it is not empirical proof, clinical guidance, or a claim of universal law. That framing is preserved throughout this demo. Every figure that the source marks hypothetical is labeled conceptual or unvalidated here.
How it works
- Read the figures in order. Figures 1 to 11 are the diagrams the source renders. Figures 12 to 18 are the additional required visuals, drawn here from the source written specifications and labeled specification only.
- Use the calculators in the Instrument Panel section. They implement the gauge meanings the source defines in Section 13, using demo formulas stated next to each calculator.
- Load a sample, change one input, and watch the zone and the composite move. Then copy the share link to save that exact state.
- Check the completeness meter. It lists exactly which inputs are still empty.
What this is not
- Not empirical proof. The atlas is a representation layer, not evidence.
- Not clinical guidance and not a claim of universal law.
- Not a verified transcription of the raster diagrams. The source PDF carries the 11 diagrams as raster images with no machine-readable text layer inside them, so internal node labels inside those images could not be extracted. The SVG figures here are faithful reconstructions from the source captions and section specifications. See Honest gaps.
- Not a live system. There is no backend, no data feed, and no model behind these figures.
1. Visual design principles
Source Section 1. Six principles, applied to every figure on this page.
- Every visual must correspond to a defined model, decision, or operational workflow.
- Colors should encode status or mechanism consistently. Brass marks the active path or accent, ink marks structure, muted marks context. No meaning is carried by color alone; every figure also carries text labels.
- Every arrow must have a defined relationship. Each arrow in the figures below is labeled with its relationship.
- Hypothetical regions must be labeled conceptual or unvalidated.
- Every compressed visual needs an explanation and evidence expansion path. Each figure has a caption and a note naming what it compresses.
- Visuals should expose contradictions, missing transitions, and overloaded concepts rather than hide them. Gaps are listed in the Honest gaps section instead of smoothed over.
Browse the figures
Filter by family. The filter does not delete anything; it hides figures outside the chosen family.
Showing 18 of 18 figures.
2. Rendered figures 1 to 6
Source Section 2. Figures 1 to 11 are the diagrams the source renders. Status for each: conceptual model unless the caption says otherwise. No figure here claims validation the source does not claim.
Arrows: Observe leads to Orient, Orient to Decide, Decide to Act, Act to Learn, Learn to Transfer, Transfer feeds back to Observe. Emergence is drawn outside the loop because the source places it outside the guaranteed loop. Source claim.
Reading: productive tension is the upper right quadrant only. High difference without containment is destructive, not productive. Source claim. The calculator in the Instrument Panel section makes this joint dependence computable.
Arrows: each deep domain compresses to a latent schema; the schema reorganizes under a novel target. The source gives no threshold for when reorganization counts as emergence, so none is drawn.
The dashed line is the audit path: from evidence straight back up to the symbol. Compression fidelity in the Instrument Panel section measures how much utility survives that round trip.
Arrows are temporal: cutoff, then blinded test, then evidence run, then ruling. The missing parameters (sample size, blinding method, ruling threshold) are listed in Honest gaps.
The core connects to four subsystems. Decision receipts are the output that makes the core auditable. This figure is an engineering design in the source, not evidence that the core exists as built software.
Rendered figures 7 to 11
Arrows: raw corpus is refined, then branches to canonical graph, indexes, and training sets. The source gives no refining rules, dedup rule, or fidelity threshold.
Six classes with no hierarchy drawn in the source caption. Latent schema is highlighted because Figures 3 and 4 both depend on it.
Governance is drawn as a base layer, not a side box, because the source places it under all three layers. Promotion from research to canonical is a governed step, not an automatic one.
Arrows are conversion steps. Evidence is the final output and also the audit trail for the whole chain, which is why it is highlighted.
The three dimensions and three domains drawn here are illustrative examples, not source-specified values. The source caption only states that capability is multidimensional and domain-dependent. That is the honest limit of this figure.
3. Additional required visuals
Source Section 12. The source does not render these seven visuals. It specifies them in writing. Each figure below is drawn from that written specification and labeled specification only. Count note: the source title page describes a fifteen-visual architecture, but Sections 2 and 12 together name 11 rendered plus 7 specified visuals, which is 18. That count mismatch is flagged in Honest gaps rather than resolved silently.
Specification only. The source gives the inputs and the five branches but no pressure formula and no branch thresholds.
Specification only. Node positions here are illustrative. The source names no concepts, so no real concept is placed on this landscape.
Specification only. The source lists the states and the order but gives no transition conditions. None are invented here.
Specification only. Per the atlas own QA rule (no unlabeled arrows), edges are omitted because the source defines no relationships between these eleven systems. Drawing invented edges would violate that rule.
Specification only. The four row labels reuse the case study domains from Figure 18 because the source names those four domains elsewhere; no cell score is filled in because the source gives none.
Specification only. The timeline shape is drawn, but no event is dated because the source supplies no dates for any concept event.
Specification only. Panel content is not invented; each panel is an empty frame waiting for the source case material.
4. Emergence Instrument Panel
Source Section 13. Ten gauges, with the meaning the source assigns to each. The source gives meanings, not formulas. The three calculators below use demo-defined formulas, stated in full next to each one, so nothing here looks like a source formula.
| Gauge | Meaning (source claim) |
|---|---|
| Polarity intensity | Magnitude of meaningful representational difference. |
| Containment capacity | Trust, governance, resources, and recovery. |
| Learning velocity | Durable update per unit time or experiment. |
| Adaptation depth | Parameter through ontology or identity rewrite. |
| Damage accumulation | Loss of viability or optionality. |
| Transfer score | Breadth and distance of successful application. |
| Retrieval precision | Relevant evidence recovered at required resolution. |
| Compression fidelity | Utility preserved relative to original evidence. |
| Governance state | Authority, reversibility, and audit compliance. |
| Emergence confidence | Novelty, persistence, transfer, and organization change composite. |
Calculator A. Productive tension (Figure 2)
Demo-defined formula: productive tension = difference times containment divided by 10. Inputs are 0 to 10, conceptual units. Zone: difference at least 6 and containment at least 6 is Productive; difference at least 6 and containment below 6 is Destructive; difference below 6 and containment at least 6 is Stable; otherwise Inert.
Calculator B. Emergence confidence (Section 13 composite)
Demo-defined formula: emergence confidence = mean of novelty, persistence, transfer, and organization change, times 10, as a percent. Each input is 0 to 10, conceptual units. The source names the four components but gives no weighting, so equal weighting is used and stated here.
Calculator C. Retrieval precision and compression fidelity
Demo-defined formulas, standard ratios: retrieval precision = relevant retrieved divided by total retrieved, times 100, as a percent. Compression fidelity = utility preserved divided by original evidence utility, times 100, as a percent. Both need a nonzero denominator.
Governance state
Sample loaders
Each sample fills all panel inputs, then runs all three calculators.
Try-it walkthrough
- Start empty. The completeness meter below reads 0 of 11 fields. That is the honest starting state.
- Load Sample 1. Click "Sample 1: Productive high tension". Difference 8, containment 7 gives tension 5.6 and zone Productive. Emergence inputs 8, 7, 6, 9 give 75 percent. Precision 18 of 24 gives 75 percent. Fidelity 85 of 100 gives 85 percent.
- Break it. Change containment to 3 and recalculate. Tension falls to 2.4 and the zone flips to Destructive. That is Figure 2 working: the same difference, less containment, a different outcome.
- Load Sample 2. Difference 9, containment 2 gives tension 1.8, zone Destructive, emergence 40 percent, precision 40 percent, fidelity 45 percent.
- Load Sample 3. Difference 3, containment 8 gives tension 2.4, zone Stable, emergence 35 percent, precision 90 percent, fidelity 95 percent. Low tension is not failure; it is stability.
- Share it. Click "Copy share link". The link encodes the current inputs in the URL hash, so opening it restores the same state.
5. Interactive graph requirements
Source Section 14. Seven requirements. This demo implements the ones a static page can implement, and labels the rest as not implemented here.
| Requirement (source claim) | Status in this demo |
|---|---|
| Zoom from canon to evidence | Partial. Figure 4 shows the expansion path; there is no zoomable graph and no evidence store behind it. |
| Filter by date, speaker, domain, project, status, and confidence | Partial. The figure filter above filters by family only. No date, speaker, or project data exists in the source to filter on. |
| Display aliases and supersession paths | Not implemented. Figure 17 shows the timeline shape with no events. |
| Show supporting and contradicting evidence separately | Partial. Figure 13 marks a contradiction link; no evidence records exist to show. |
| Reveal domain bridges and latent schemas | Shown as figures. Figure 3 draws the bridge; no live graph traversal exists. |
| Allow any visual object to open its decision receipts and source records | Not implemented. No receipt store or source record store is supplied with the atlas. |
| Support before and after snapshots for representational change | Not implemented. No snapshot store is supplied. |
6. Visual QA standards
Source Section 15. Checked against this page.
| Standard | Check on this page |
|---|---|
| No unlabeled arrows | Pass for drawn arrows. Figure 15 draws no edges at all rather than unlabeled ones, because the source defines no relationships there. |
| No color-only meaning; include labels or patterns | Pass. Every colored region also carries a text label. |
| Readable in grayscale | Pass by design. Bone, paper, ink, and brass differ in lightness as well as hue, and labels carry the meaning. Not verified by a printed grayscale test. |
| Every metric has units or an explicit conceptual label | Pass. Calculator inputs are labeled 0 to 10 conceptual units, counts, or units. |
| Every hypothetical diagram has a validation status | Pass. Conceptual and specification-only labels appear on the relevant figures. |
| Every graph can trace to machine-readable data | Fail in this demo. The atlas supplies no machine-readable data file, and this page has no data store behind the figures. Listed as an honest gap. |
Honest gaps
- Count mismatch in the source. The title page describes a fifteen-visual architecture, but the source names 11 rendered visuals plus 7 specified visuals, which totals 18. This demo reproduces all 18 and flags the mismatch instead of silently dropping three.
- Internal diagram labels are not extractable. The 11 rendered diagrams are raster images in the PDF with no text layer inside them. The SVG figures here are reconstructions from the one-line captions and the section specifications. Internal node wording, arrow wording, and any numbers inside the original images could not be verified from the text layer.
- No formulas in the source. The source defines gauge meanings but no formula for productive tension, emergence confidence, pressure, or any composite. Every formula on this page is demo-defined and labeled as such.
- No thresholds in the source. Zone boundaries (6 and above), the Case Zero ruling threshold, the Emergence Furnace branch thresholds, and the Claim Lifecycle transition conditions are all absent from the source. The zone boundary used here is demo-defined.
- No data behind the figures. No concepts, dates, scores, evidence records, receipts, or snapshot data are supplied. Figures 13, 16, and 17 therefore show structure with illustrative or empty content, and the interactive graph requirements that need a data store are not implemented.
- Figure 11 dimensions and domains are illustrative. The source caption only says capability is multidimensional and domain-dependent. The three dimensions and three domains drawn are examples, not source values.
- Figure 16 row labels are borrowed. The four domains reuse the case study domains because the source names no matrix rows of its own.
- Layout not eyeballed in a real browser. Controls are tested in jsdom (see receipt below). Visual layout in a real browser was not eyeballed.
Related sites
- Neuruh Hub: the master map of Neuruh builds.
- Conceptual Lexicon: sibling vocabulary build.
- Intent Preservation Protocol: sibling protocol build.
- Modular AI Research Agent Architecture: sibling research agent build.
Sibling links are related Neuruh builds, not sources for this atlas. The only source for this page is the Adaptive Intelligence Visual Atlas v1.0 PDF named in the document control header.