This one still needs to stay a bundle.
The important change here is not for MUIOGO to make local decisions about upstream result semantics. The important change is for MUIOGO to correctly honor upstream v5.5 result metadata and variable definitions.
That means the default rule here is:
- upstream
Variables.json is the source of truth
- MUIOGO should adapt its result rendering, registry, unit generation, and Pivot behavior so upstream metadata works correctly
What should land here:
- adopt upstream
Variables.json as-is
- expose the upstream-owned result groups and metadata correctly in the MUIOGO frontend
- make MUIOGO's result handling honor upstream unit rules and grouping cleanly
- keep existing result screens working while aligning them to upstream
v5.5 metadata
What should not land here:
- no local reinterpretation of upstream units or semantics by default
- no formatting-only churn in
Parameters.json
- no splitting this into half-ported frontend changes that leave result screens inconsistent
Parameters.json stays out of scope unless we find a semantic upstream change that actually matters. Current evidence still points to formatting-only churn there.
Done when:
- upstream
v5.5 result metadata is loaded as-is
- MUIOGO renders those result groups and units correctly
- Pivot works on the demo data with the upstream metadata
- existing non-broken result screens do not regress
Related work:
This one still needs to stay a bundle.
The important change here is not for MUIOGO to make local decisions about upstream result semantics. The important change is for MUIOGO to correctly honor upstream
v5.5result metadata and variable definitions.That means the default rule here is:
Variables.jsonis the source of truthWhat should land here:
Variables.jsonas-isv5.5metadataWhat should not land here:
Parameters.jsonParameters.jsonstays out of scope unless we find a semantic upstream change that actually matters. Current evidence still points to formatting-only churn there.Done when:
v5.5result metadata is loaded as-isRelated work: