Re: enriched-mode and switching major modes.

[Richard Stallman]

    That's true.  Perhaps one could have a block-begin text property and a
    block-end text property, both attached to the gap between two
    characters?

Text properties are attached to characters, not to gaps.
I don't think this idea could make sense.

Another idea is to give each box a "name", a Lisp object,
which has to be eq in order for them to be the same box.
It could be a list of data that defines the box characteristics.

    > 2. How to represent line breaks.  Saying "break one long line at
    > display time" would work ok for display, but all the commands that
    > operate on lines would see just one long line there.

    I think this is not too much of a problem.  Kim has done a great job
    of changing the behavior of some Emacs commands to do cua-like
    things.  I expect that a similar method would work to change the
    behavior of beginning-of-line, next-line and so on.

Sorry, I do not want to have to change each command to
explicitly recognize lines in two different ways.

    Imagine that you have a feature which wraps long lines at word
    boundaries at display time.  Imagine that this feature was configured
    to wrap to the window width.  Now imagine that you make the window
    narrower.  This introduces more wrapping points.  But which Lisp code
    needs to see more newlines?

Whatever code operates on lines.