Experiments

A core set of canonical experiments that connect HYWE's theoretical research questions directly to empirical evidence. These experiments trace the growth of complexity and the deterministic guarantees of the spatial compiler.

Planned Experiments

Experiment 01 — Same intent, alternate configurations

• Question: How does the same topological intent map to multiple valid geometric realizations?
• Status: Demonstrated (Completed in Batch capabilities)

Experiment 02 — Increasing relational complexity

• Question: How does the system handle an increasing density of constraints?
• Status: Future (Planned)

Experiment 03 — Boundary-constrained resolution

• Question: How do external site boundaries constrain and guide configuration generation?
• Status: Future (Planned)

Experiment 04 — Deterministic repeatability

• Question: Can the compiler produce identical output for identical input across distinct environments?
• Status: Active (Partially addressed by existing benchmark)

Experiment 05 — Configuration-space growth

• Question: How does the number of valid configurations scale against intent complexity and computational cost?
• Status: Active (Partially addressed by existing benchmark)

Experiment 06 — Architectural interpretation

• Question: How successfully can the generated relational configurations be interpreted into recognizable architectural elements?
• Status: Future (Planned)

Experiment 07 — Syntactic perturbation and structural sensitivity

• Question: How much of the configuration space is preserved under a minimal change in spatial intent (e.g., altering a single relational token)? Does the system exhibit local sensitivity (small geometric adjustments) or structural sensitivity (topological bifurcations leading to largely disjoint configuration spaces)?
• Proposed Metric: $D(C_0, C_1) = 1 - \frac{|C_0 \cap C_1|}{|C_0 \cup C_1|}$ (where $D=0$ means identical spaces, and $D=1$ means completely disjoint spaces).
• Status: Future (Planned)