D. Ben Knoble

This might be because I'm used to the old behavior, but every time `(flow (+))` signals an error I'm surprised :) That said… it is an "unused syntax." OTOH, inserting...

Sorry, close was a typo. The idea was mostly to avoid a possibly confusing syntactic sugar.

(possibly unrelated) nit: The following paragraph needlessly ties `effect` to the semantics of `tee` (the _equivalences_ in the preceding paragraph are useful, but explaining the _implementation_ seems gratuitous): > Remember...

An original comment chain on `readout`, Shrodinger's probe, and effects: https://github.com/drym-org/qi/pull/152#discussion_r1526209817

RE: effects within `esc`: https://github.com/drym-org/qi/pull/152#discussion_r1465021409, https://github.com/drym-org/qi/pull/152#discussion_r1466456812. I think some of these were addressed, but I also think the conversation re-arose during today's meeting.

RE: "syntax warnings," what produces the "unused identifer" annotations in Racket Mode/DrRacket/racket-langserver?

RE: the effect relation, _if the `ordering` function (judgement?) Michael proposed is writable_: it will likely to be easier with a smaller core syntax.

While I like allow/deny, have we considered calling `_` the "identity" or "do-nothing" flow? FWIW I think Python's `pass` is more like `ground` or `void` since it returns `None`.

As a category theory novice, much of this is a bit over my head, so being categorically consistent isn't as important to me as "making sense" for some intuitive notion...

I suggested using boxes in the implementation. I made the changes to the current compiler branch (patch at the end) and ran the benchmarks on both sides. What I saw...