Peter: Hahahaa.. because of course you are! That could be really interesting. Why not just create a tool that takes Scapple and Timeline and then exports them together directly into bookstores? :D
You are correct-- the edges can't be labeled in the Windows version of Scapple, so I use those bubbles to note the nature of the relationship, which do give me the added benefit of 1:Many and Many:Many relationships, but I'd love to have both. It seems very odd to me that an application of this nature can't efficiently handle ~100 nodes. There is a great argument for splitting up a Scapple, but I'm working with what I've got (mentally) and out-of-sight/out-of-mind. Because of the "sloppy" nature of this story, there is a complex internal reality to keep track of, as well as temporal domains.
I did look at Heimer and it is a technical fit but definitely not a functional one. For it to be functional for me (just as Timeline does), it has to convey a lot of form-differentiated information (e.g., colors, shapes). Heimer seems like it could readily represent the structure, but not hit the other measures for me-- I would be made a bit crazy by the lack of curves and shapes (it's like limiting a vocabulary). I had thought of using UML-type tools, but they all have a lot of overhead and are slow to work with, providing they can even represent arbitrary-shaped networks (you know how IT folks love our straight edges and right angles!).
Papyrus looks like an amazing tool! But, as above, not for this purpose, seemingly.
It is so. very. odd. that only one application of this very "obvious" type exists.
Right now, my only objections to Scapple are it 1) Is not performant > 100 nodes, 2) Can't inherently label edges., oh, and 3) Doesn't support persistent (updatable) styles for nodes.
I think I will be using all of the tools mentioned here at some point-- but not for this purpose!