supplement: question about "formal equation" for Bayes act

[elray1]

Here's the text in question:

• My question is whether we need to include Eq (30)?
• Motivations:
  • it took me a while to process what was going on here because I didn't catch all the notation, and had to think through "the point is that the left hand side gives you the same value, $\tau$, repeated $N$ times, and that matches the $\tau$ that's returned by $F_T(K)$".
  • Is this notation needed to set up (31)? It doesn't seem to be used there, and we could already just write down (31) based on (28) and (27), right?

[aaronger]

Yes, we already know (31) so (30) is not logically carrying much weight. But I find the 1_A adjoint idea to be helpful as offering another perspective and leading perhaps to a generalization to other "totalling" maps. But I can remove it if it seems too distracting.

[elray1]

I vote for removing it here -- the cognitive burden on the reader making their way through this supplement is already high, so I think we should try to avoid extra things :)

[aaronger]

OK will try to do soon, but the $\undertilde{\bm{F}}$ is starting to feel like a better solution than anything I was using before. Maybe I set myself up by saying "interesting"...