# Rams scorecard

1. **Good design is innovative — 2/3**
   Evidence: one custom print/geometry system refreshes conventional shell surfaces (`01-evidence.md#1-innovative`).
   Justification: the expression is distinctive, but it does not introduce a new desktop interaction pattern.

2. **Good design makes a product useful — 2/3**
   Evidence: core state and actions are direct, while most shell controls require a pointer (`01-evidence.md#2-useful`).
   Justification: the primary tasks work quickly, but keyboard and discoverability gaps add friction.

3. **Good design is aesthetic — 2/3**
   Evidence: the token system is strong, with several type/geometry exceptions and a 3.68:1 text pairing (`01-evidence.md#3-aesthetic`).
   Justification: the visual system is unmistakable, but contrast and intensity inconsistencies prevent a fully resolved score.

4. **Good design makes a product understandable — 1/3**
   Evidence: multiple rail gestures are undisclosed and several state labels read like actions (`01-evidence.md#4-understandable`).
   Justification: more than three controls require prior knowledge or experimentation.

5. **Good design is unobtrusive — 1/3**
   Evidence: compact permanent chrome expands into visually competing full-screen decoration (`01-evidence.md#5-unobtrusive`).
   Justification: the theme sometimes becomes the figure when the user’s task should remain the figure.

6. **Good design is honest — 2/3**
   Evidence: no dark patterns exist, but held-count, muted-volume, and disabled-slider presentation diverge from behavior (`01-evidence.md#6-honest`).
   Justification: the mismatches are minor rather than deceptive, but prevent a one-to-one mapping everywhere.

7. **Good design is long-lasting — 2/3**
   Evidence: reusable primitives coexist with literal franchise and gaming-HUD markers (`01-evidence.md#7-long-lasting`).
   Justification: the craft will last, while the most literal references may date the surface.

8. **Good design is thorough — 1/3**
   Evidence: success, generic disabled, broad focus, and accessibility metadata are missing or rough (`01-evidence.md#8-thorough`).
   Justification: three or more important state families remain incomplete.

9. **Good design is environmentally friendly — 1/3**
   Evidence: source weight is small, but recurring polling and idle motion have no reduced-motion gate (`01-evidence.md#9-environmentally-friendly`).
   Justification: the runtime is lean yet motion is always on when its conditions are met.

10. **Good design is as little design as possible — 1/3**
    Evidence: restrained single slabs coexist with repeated affordances and multi-layer modal atmosphere (`01-evidence.md#10-as-little-design-as-possible`).
    Justification: at least three decorative or duplicated elements could be removed without breaking a task.

**Total: 15/30**