mailing-list for TeXmacs Users

[Henri Lesourd <address@hidden>]

>Thank you Henri for posting this package.  It comes close to my needs.
>
>However, there is one key defect.  Consider a tree with root A which
>has children B and C, and C has children D, E, F, and G (i.e
><proof-edit|<tree|A|B|<tree|C|D|E|F|G>>>).  The proof-tree package
>will make the bar between A and B and C wide enough to cover D, E, F
>and G.  For large Natural Deduction proofs this becomes annoying.  The
>line should only be wide enough to cover B and C but not their
>children.  Otherwise I could do the same thing with clever use of
>fraction bars.
>
Yes, but as soon as the proof becomes big, editing it becomes
very cumbersome. This is why I wrote this package in the first
place.

> Look carefully at the widths of the bars in
>http://www.phil.cam.ac.uk/teaching_staff/Smith/LaTeX/nd.html to see
>what I mean.
>
Yes, I see.

>Unfortunately, I don't know enough about the internals of TeXmacs to
>determine to read the source and determine if or how the proof-trees
>package could be tweaked to handle this.
>
It will be difficult to do it, because the proof-trees package
uses the \frac primitive of TeXmacs, which doesn't handle this
at all.

So a solution is possible, but this is more complex than just
cleaning my code and posting it on my web page :-(. I will have
to think about how to do it, but I cannot tell you when a solution
will be ready.

Anyway, it is interesting, because none of the other people
that use the proof-tree package ever raised this question,
which is real, indeed (but there are so many questions that
can be raised... :-). So in the future, the problem with the
fraction bars will be included in the requirements for typesetting
these kinds of proof trees.


Best, Henri