Decision
Scope expression with nested modes (Mise en Mode): visually box a region, then supply new values to existing semantic tokens only within that scope, rather than creating dedicated tokens per expressive moment.
Alternatives Considered
Create dedicated tokens for each expressive moment
| đź‘Ť Pros | đź‘Ž Cons |
|---|---|
| Explicit and easy to manage | Token bloat Erodes the meaning and usage Every new idea is a few new tokens to maintain |
One-off component variants
| đź‘Ť Pros | đź‘Ž Cons |
|---|---|
| Localized to the component | The expression is locked inside one component, so it can’t be reused Doesn’t compose across a region of mixed elements |
Reasoning
Moving semantic responsibility from tokens to scopes resolves the tension between expression and consistency. A boxed region becomes a place for a new expression by receiving new values for the same semantic tokens, no new tokens required, so the system stays bounded and coherent.
Why it mattered
Mise en Mode reframed the question from “what token do I add?” to “what scope receives a new expression?” That shift is what lets a system stay consistent and still feel expressive where it counts.
All decisions