SciFest judge review
A short route through the question, method, present evidence, live demonstration and authorship. Each step links to the full record if you want more detail.
Timed route
The schedule is a guide. The links and checklist also provide a fallback if a live provider is slow or unavailable.
0:00–0:45
Can public road geometry, physics relationships and explicit assumptions produce an inspectable comparative signal without presenting it as a safety rating?
0:45–1:45
Trace the public default Model 2.0 first, then distinguish Model 3.2 geometry evidence and Model 3.3 diagnostic candidates.
1:45–2:45
Separate deterministic software checks, historical model-output records, empirical evaluation and independent review.
2:45–4:30
Select a road, change one scenario setting, inspect a local section, open Maths or Geometry, then return to the map without losing state.
Demonstration checklist
Use the reviewed deterministic example. It exercises the normal calculation, Maths, Geometry and export paths without claiming that a live provider or real road was tested.
Open reviewed demonstrationBefore the demonstration
Current model status
Empirical decision: not assessed
Empirical decision: insufficient evidence
Empirical decision: insufficient evidence
Software verification and deployment status are reported separately from empirical evidence. No model output is an official safety rating or collision prediction.
