Comparative analysis

This page reads the six reconstructions next to one another. It projects the book’s Chapter 6 material straight to the web, drawn from the same model, so the two surfaces cannot drift. The two convergence tables are the exact tables the print edition lays out; the correspondence matrix and the modeling-ceiling ladder carry the same data the book renders, shaped here as tables where print presents it as cards and prose. MAGE ran at none of these companies. Every mark records how the book reads an independent report against the theory, never a claim a company makes about MAGE.

Reading the tables

The convergence and ceiling tables use a small set of marks. A check means the source establishes the pattern in full. A half-circle means it establishes the pattern in part. An em-dash means the source is silent. The modeling ceiling adds a hollow circle for a representation a team exercises without naming it. The correspondence matrix keeps its words — strong, partial, and the rest — because there the word carries the reading.

Correspondence: each system against each construct

The matrix places every reconstruction against the MAGE constructs, with DocAble — the system the theory was built on — as the first row. Read down a column to see how widely one construct recurs across independent systems. Read across a row to see which arm of the theory a single system leans on.

The modeling ceiling: how far each system’s models reach

The ladder runs from the shallowest representation a system keeps to the deepest. A system’s ceiling is the highest rung its described practice reaches. Some stop at topology and policy; others climb into behavior, invariants, and model-derived verification. The book reads the rung a source reaches, not a grade.

What every system establishes

These patterns recur across all six. Each was engineered independently, for one company’s own reasons, yet every source lands on it. Convergence this broad is the evidence that a pattern is structural rather than a house style.

The key names the construct each pattern instantiates and states what the source establishes.

What industry is independently discovering

This second set recurs too, but the support varies from site to site. Some sources establish the pattern in full, others only in part. MAGE generalizes what these systems reach for; the gaps in a row mark where one source stops short.

The key for this bucket reads the same way.

What none of the six generalizes

MAGE is a theory of the engineering ENVIRONMENT ITSELF as the object of engineering: everyone else engineers an agent, a runtime, a policy engine, or a model; MAGE engineers the governed environment in which commodity intelligence operates.

None of the external cases we examined describes the following machinery in the generalized form MAGE does. Not 'nobody in industry has ever done this.'

None of the reconstructions states the following machinery in the general form the theory does:

• The strongest notion of model — a Model Zoo spanning logical / behavior / process / physical / scenario / invariant / lineage / traceability views.

• Explicitly binds the model to the territory — general model<->implementation parity plus a mechanical drift check.

• A common grammar for governance mechanisms — representation / constraint / sensor / validator / gate / feedback as cross-technology types.

• Governance conversion as a general engineering operation — classify the failure, then choose the durable mechanism from model|constraint|sensor|validator|gate|interface|architecture|no-mechanism.

• Governs the accumulated governance itself — strengthen / reconcile / retire.

• Explicitly theorizes engineering capital — the compounding feedback loop as a dynamic, not just an accumulation.

• Integrates Modeling and Alignment as the two irreducible moves — better models make more obligations deterministically enforceable.

• Treats the environment itself as the object of engineering — THE SPINE (RH2), not a peer of the other seven; they are its consequences.