mailing-list for TeXmacs Users

[Joris van der Hoeven <address@hidden>]

On Sun, Nov 13, 2005 at 01:49:57PM -0800, Karl Hegbloom wrote:
> On Fri, 2005-11-11 at 16:53 +0100, Alvaro Tejero Cantero wrote:
> > On Fri, 2005-11-11 at 09:20 -0600, Juan Pablo Romero wrote:
> > > I've been playing with sessions lately, specially R and XYpic.
> > > 
> > > I find them extremely useful, and I'd like to make a couple of 
> > > suggestions:
> > > 
> > > * What about a "copy input from above" / "copy output from above"
> > > functions? (you know, like mathematica's <Ctrl+L>, <Shift+Ctrl+L>).
> > 
> > This would be fantastic for tables -> to use in eqnarrays in which you
> > succesively process the output in the right hand side.
> 
> I wonder how hard it would be to implement Scheme code that can perform
> algebraic transformations of expressions in the process?  It could use
> pattern matching and transformation rules, perhaps.  It should not be
> too automatic...  Things like distribute, factor n, re-arrange terms,
> etc. that don't necessarily require CAS to perform.  If it could
> remember the order you've arranged terms, and after CAS processing put
> them back in that order...  What else?

I agree. I started moving in this direction for the Maple interface bis
(outcommented part in the sources), because Maple output is so badly sorted.
This also has applications to compressing big subexpressions to clickable
abbreviations. Of course, the TeXmacs part should merely be syntactic:
real computations should be delegated to the CAS. However, progress is slow...

Best wishes, Joris