Re: Non-linear / full-2d writing systems?

... and a few more, since I lied about the "one" anyway... (:-P)

* How do you deal with the discreteness of your writing surfaces, if
they are so? (E.g. pieces of paper, or paper book format... for that
matter, what would replace that?) If not, then you'll still inevitably
have "windowing" (if simply by the amount your visual field can
encompass at once at any given level of detail, though more likely a
monitor), so how do you deal with that?

* What about pro-forms? How would you reduce a subtree (subgraph?)
into a proform? Would you?

* Mulitply used / referenced atoms? Wouldn't want it to become
overcrowded; might have to have a trunk-and-branch system for hooks
that are used a whole lot, or have multiple instantiations of that
item, or special reduced proform links (e.g. that mean "this link is
linking to Pro1"), or ...

* How do you optimize for different writing implements? E.g., it's
easier / faster to write if you don't have to reposition / pick up a
pen; would this entail multiple pens at once, or other innovations?

* Level-of-Detail (LOD): Would be nice to have it chunk well (to 7+-2
of course ;-)), so that any given LOD has only an encompassable number
of chunks. How to do this visually, so that the same-LOD, connected
items are easily (visually) grouped together and separated from their
context? (Probably a similar question to general
channel-discrimination in a noisy environment.)

* Structure: How would you structure basic things? An argument? A
conversation (would that mean anything?)? Etc., for a given
communicative intent...

 - Sai