Experiment · Code & architecture
Architecture Difference Case Study
Compare two tiny systems that return the same answer but handle responsibility and failure differently.
Repository: CinvanaAI/architecture-difference-case-study
Recorded synthetic example. An executable experiment using supplied synthetic inputs.
See the idea in action.
Both return HELLO WORLD. A closes its environment; B records one store write, one capability call, and one bridge delivery, leaving its environment open.
The same string, with surrounding spaces, passed through both toy systems.
{
"input": " hello world ",
"same_output": true,
"version_a": {
"output": "HELLO WORLD",
"environment_after_close": "closed",
"agent_state": "idle",
"work_count": 1,
"records": 1
},
"version_b": {
"output": "HELLO WORLD",
"store": {
"count": 1,
"archived": 0
},
"store_operations": [
"write"
],
"capability_calls": 1,
"bridge_deliveries": 1,
"provider_routes": [
"primary"
],
"environment_open_after_run": true
},
"provider": "inert local stub in both versions"
}Try the example.
From the repository root, follow the dependency requirements in the README. This example uses supplied synthetic material.
python -m examples.walkthrough
Both return HELLO WORLD. A closes its environment; B records one store write, one capability call, and one bridge delivery, leaving its environment open.
Complete setup and instructions ↗The interesting part.
Equal output makes ownership, lifecycle, storage, and relay differences inspectable without a large application.
Where it came from.
Controlled comparison of two source versions created in March 2026.
March 2026 — Creation of the controlled comparison working tree. Source ↗
Preserves both implementations and the difference map; excludes editor configuration.
A controlled architecture comparison, not two independently marketed products.