Derivable Architectures — Volume II

Paper 011 — Spaces of Derivations

Research Program: Derivable Architectures

Volume: II — Structural Theory

Document: papers/011-spaces-of-derivations.md

Status: Draft 0.1

Authors

Eduardo N. Hering
OpenAI ChatGPT


Abstract

Volume I established that derivation transforms an abstract architecture into concrete implementation artifacts while preserving architectural identity.

This paper investigates the collection of all valid derivations associated with a given architecture.

Rather than viewing derivation as an isolated process, we treat it as an object of study.

The principal observation is that a derivable architecture naturally induces a structured space of admissible derivations.

Many later concepts—including equivalence, optimization, reversibility, canonical realization, and architectural metrics—emerge from the properties of this space.


1. Observation

A single architecture rarely possesses only one implementation.

The ICA Baseline provides immediate evidence.

The same constitutional architecture may generate:

  • different jurisdictions,
  • different legal systems,
  • different regulatory environments,
  • different organizational scales,
  • different operational constraints,

without changing the architecture itself.

Thus, an architecture is associated not with one derivation, but with many.


2. Definition

Definition 1 — Space of Derivations

Given an abstract architecture A,

its Space of Derivations is the collection of every derivation that satisfies the correctness conditions established in Volume I.

This space is denoted informally as

D(A).

No mathematical assumptions are yet imposed.

It is merely the collection of all architecturally valid derivations.


3. Observation

The existence of one valid derivation does not imply uniqueness.

Likewise,

the existence of multiple derivations does not imply architectural ambiguity.

Architectural identity remains fixed.

Only realization varies.


4. Proposition

A derivable architecture defines a family of implementations rather than a single implementation.


Proof Sketch

By definition,

correct derivations preserve architectural identity.

If two derivations preserve the same identity while producing different artifacts,

both belong to D(A).

Therefore,

multiple implementations may legitimately coexist.


5. Observation

Different derivations may differ in many respects:

  • jurisdiction

  • language

  • terminology

  • implementation strategy

  • organizational decomposition

  • regulatory adaptations

  • document granularity

  • deployment sequence

without violating architectural identity.

The derivation space therefore exhibits diversity while remaining bounded by architectural invariants.


6. Definition

Definition 2 — Neighboring Derivations

Two derivations are neighboring when they differ only by a small set of implementation decisions while preserving the same architectural identity.

The precise meaning of “small” is intentionally left undefined at this stage.

Later papers introduce metrics capable of formalizing neighborhood.


7. Observation

Neighboring derivations frequently arise during practical implementation.

Examples include:

  • adapting terminology,

  • replacing statutory references,

  • changing procedural order,

  • adjusting document partitioning,

  • introducing jurisdiction-specific clauses.

These modifications alter implementation artifacts but do not alter the underlying architecture.


8. Proposition

The derivation space possesses local structure.


Proof Sketch

Neighboring derivations differ incrementally.

Incremental variation implies that derivations need not be isolated points.

Instead,

small implementation changes generate families of closely related derivations.

Therefore,

the derivation space admits local neighborhoods.


9. Observation

Large architectural changes are not always reachable through arbitrarily small derivation changes.

Certain implementation decisions require coordinated modifications across multiple artifacts.

Consequently,

the derivation space may contain regions separated by structural discontinuities.


Definition 3 — Region of Derivations

A region is a collection of neighboring derivations sharing common implementation characteristics.

Examples include:

  • a jurisdictional region,

  • a corporate-law region,

  • a constitutional region,

  • an implementation-style region.

Regions describe implementation similarity rather than architectural identity.


10. Proposition

Regions partition implementation diversity without partitioning architectural identity.


Proof Sketch

Architectural identity is determined by invariants.

Regions are determined by implementation characteristics.

Since implementation characteristics may vary while invariants remain unchanged,

regional boundaries do not constitute architectural boundaries.


11. Observation

The ICA Brazil Profile occupies one point within D(ICA).

It is evidence that at least one valid derivation exists.

Nothing in the architecture suggests uniqueness.

On the contrary,

the constitutional design explicitly anticipates additional jurisdictional profiles.


12. Corollary

A successful implementation demonstrates existence but not completeness of the derivation space.

The Brazil Profile establishes

Existence.

It does not enumerate all possible realizations.


13. Conjecture

Most practically useful architectures possess derivation spaces whose cardinality is extremely large.

The engineering challenge is therefore not generating derivations,

but selecting appropriate ones.


14. Open Questions

Among the derivations belonging to D(A):

  • How should similarity be measured?

  • Can shortest derivations be defined?

  • Can optimal derivations exist?

  • Are there disconnected regions?

  • Do derivation spaces possess symmetry?

  • Can every derivation be decomposed into elementary steps?

These questions motivate the remaining papers of Volume II.


Conclusion

Derivation is not a single process but a structured space of architecturally correct realizations.

Viewing derivations collectively rather than individually changes the nature of architectural analysis.

The implementation problem becomes one of navigation within a constrained derivation space.

This perspective prepares the way for the study of equivalence, composition, reversibility, canonical derivations, and architectural metrics developed in the subsequent papers.