I'm -0.5 on reusing Name(ctx=Store). The reason we're using that in assignments is that in the past (before the PEG parser) the parser literally used the same grammar for the LHS and RHS of assignments, and a second pass was used to rule out invalid targets (like "1 = e" or "f() = a") and mark the Name nodes representing targets as stores.

We don't reuse Name nodes for other binding targets like parameters or imports.

PS. To reach Mark it's best to email him directly, he's not very responsive to GitHub notifications (there are too many).

PPSS. I'm on vacation until May 8 and am trying not to check my email much. But I already failed on the first day. :-)