I had a hard time understanding rules 3 and 4 because they don't really seem to be related to global actor isolation at all even though that's the framing of the document. I also don't really see those rules discussed in the motivation section nor mentioned in the proposed solution section, unless I'm totally missing something?

@Jumhyn good point! I can restructure the proposal a bit to make it clear that while these rules are not strictly related to global actor cut-off, they still enable nonisolated to be written in more places than today.

@simanerush Thanks, I think that restructuring reads a lot better!! I might even argue for dropping any mention of global actors from the document title as well, since I think that sets the stage wrong. But also I see how perhaps the global actor bits are the 'primary' focus and the other parts are more ancillary.