Sam Isaacson

We should make it a normal standalone macro too though. Maybe get that working first?

I can take a closer look later today.

@TorkelE I think the issue is you aren't calling `_parse_vars` correct for parameters. See how the `@parameters` macro is defined in MTK: https://github.com/SciML/ModelingToolkit.jl/blob/1aa480673305854e889be2211eb5b0e4487038ae/src/parameters.jl#L58-L62

The second (non-DSL) example should work -- it works with `@parameters`. As I said in the issue the first is ok to not have working as it isn't how we...

I thought spatially varying mass action jump rates are already supported? That is what a `SpatialMassActionJump` encodes.

But why not add a default via a system transformation instead of a special DSL command. The former would have the benefit of preserving equivalence between DSL and symbolic construction,...

Even if we keep the DSL command, I'd advocate that it simply correspond to applying that transformation function after creating the system within the DSL instead of merging scalings via...

Won’t that run CI on the branch for all PRs and merges forever now?

Just to address some of these points: > Adding metadata to a system. We don't use system metadata? (I know MTK has a notion of it, but I don't think...

I don’t think we should use Dicts though. That seems pretty heavy for when we have systems with thousands of reactions. Looking at that thread I’d say we are in...