I am a bit confused. I thought we agreed that each branch/note inset should have two properties that correspond to I. and II. above. That means, in the document dialog, you can both activate/deactive an inset _and_ define its position (or, more general, its appearance) at the same time.
Actually, the more I think about it, the more I come to the conclusion thatyour approach mixes at least two different concepts which should be separated:
I. Insets that show/hide parts of the document -> i.e. Branch, Comment, Note, Greyedout -> Can contain very large contents up to a complete document, i.e. with sections, nested environments, bibliographies, child docs, floats etc. -> Can nest other Insets of type I and II without limitation
II. Insets that place text at a certain position -> i.e. Footnote, Margin, Petite -> Can _not_ contain complex contents -> Can nest insets of type I but not (or only limited) of type II
It is always possible to switch from II to I, but what do you do if a
curious user has a branch which contains a complete chapter and he changes
this to, say, Marginal Note? I think we should separate those.
I am not sure whether the semantic restrictions given above can be enforced technically. Maybe some LyX guru can enlighten us?
I see that your proposal is a bit safer but it also contradicts my idea to unify insets. IMHO there should be only one kind of "note" (aka "branch") inset that can be displayed in different ways ("footnote", "inline", and "margin note").
Tomorrow I hope to have a patch where these things are fixed.IMO we still haven't reached a point where we know _exactly_ how things should work. We should finished the discussion before you start to implement the branch inset. Otherwise, I expect a lot of frustration on all sides. If there is a consensus on what to do, it might be a good starting point to first design to user interface. If this is accepted, the remaining parts of the implementation can follow. (Just a proposal :-))
Kind regards,
Michael
