Graphs & Hierarchical Semantics

At the core of the EdgeNode modeling engine lies a singular, unifying design philosophy: Node = Graph = Entity.

Unlike traditional tools that enforce rigid distinctions between a "diagram," a "folder," and an "element," our environment treats these concepts as mathematically equivalent. This enables infinite, fractal architectural depth without structural bloat.

The Recursive Graph Model

To understand how EdgeNode processes complex hierarchies, we can express the architecture through recursive graph theory. In a standard computational graph, we define the topology as a tuple of vertices and edges:

$G = (V, E)$

However, in a strictly recursive environment, every vertex v belongs to V is not merely a terminal endpoint, but a potential container for an entirely new topological space. We extend the standard model by introducing a structural mapping function rho:

$Gsys=(V,E,ρ)G_{sys} = (V, E, \rho)$

Here, rho maps any vertex to the set of all possible graphs. If a node contains no internal structure, the mapping results in an empty set. If a node acts as a parent, it maps to a nested graph configuration. Because this sub-graph is subject to the exact same rules, the architecture becomes infinitely nestable.

Computational Elegance: Because every entity is fundamentally just an immutable identifier mapped to a data snapshot, the engine never passes heavy, nested graph structures into memory. When you navigate down into a node, the system queries the graph layout by targeting the parent identifier directly against the active node configuration.

The Three Equivalencies

This mathematical foundation manifests in the workspace through three practical equivalencies:

1. The Node is an Entity

A visual circle on your canvas is never just a shape; it is a direct projection of a domain entity. It holds absolute state, cycle capacities, and contour definitions. Its spatial coordinates are simply properties attached to its underlying identifier snapshot.

2. The Node is a Graph

Because the mapping function loops back into a new graph space, every entity inherently possesses an internal canvas. An entire enterprise architecture model can exist inside a single node named Enterprise, and the algorithms governing flow, limits, and edge detection remain mathematically identical at any depth layer.

3. The Graph is an Entity

Conversely, any complex graph of interacting nodes can be collapsed, viewed, and evaluated from the outside as a single functional node. The throughput of the parent entity is simply the evaluated aggregate of its internal topology.