[RFC] proposed user doc for nested trees

[Stephen J. Turnbull]

Vincent Ladeuil writes:

 >     >> I don't feel safe allowing any kind of branch (repository really)
 >     >> for the local branches.
 > 
 > Argh, *every* not *any*, sorry :-(

Actually I read that as "I don't feel safe allowing *just any* kind of
branch ...".  Aaron's just being (justifiably) careful.

 > A side note: You said: 'bzr should create local branches in
 > .bzr/branches' and 'they shall be called "subbranches"', why not
 > use .bzr/subbranches ?

But they're not going to be "subbranches" as far as you can tell from
the content of .bzr/[sub]branches, right?  That's what Aaron wrote,
and as far as I can see there's nothing that could get in the way of
implementing it that way.  They're just going to be "branches"; the
difference between a branch and a subbranch is how they're referred to
in the containing branch.  So it seems to me that that is reasonable
to name it "branches".

Question: could this mechanism be adapted without too much confusion
to colocated branches?