some editing tools for maxima /Grumpy old man..

I have thought about responding earlier to a bunch of issues, but held my
tongue, thinking I was just revealing myself as old fashioned. But I decided
to say it anyway, and let people react .  I was prompted by this last note.

To a large extent the messages on this newsgroup can be classified according
to 
(a) how can I use Maxima to do some computation
(b) is this a bug in maxima
(c) here's a new package for maxima
(d) other mathy / computation questions


(e) questions about implementations CMUCL, SBCL, GCL, CLISP, Mac, Windows
Linux

For a certain percentage of the questions in (e), the problem is not a
maxima problem at all, but a bug or a mismatch in expectations between the
programmers write Lisp and the programmers implementing Lisp. 

For a certain percentage of these problems, the difficulty is in the version
of Makefile, Perl, Tcl, gplot, etc etc.   

Here's my gripe. It seems to me that any of these features that are already
in lisp should be used in their lisp form.  Defsystem (or similar) in lisp
instead of Make.  Lisp instead of Perl.  

Now if you are trying to choose between lisps, I can understand a simple
Makefile that determines if you are talking about GCL or CMUCL, and sets a
home directory but after that, I think lisp AND SHOULD can take over, at
least to the extent possible.   Viewing lisp as a "special purpose
programming language for symbolic math kernel construction" and that
everything should normally be written in C or Perl or ... just strikes me as
a wrong world view.  If there is a piece of code in the Maxima code base
that is not in lisp, it should be because (a) it is a huge piece of code we
borrowed from somewhere else (e.g. fortran, gplot?),  or (b) it absolutely
cannot be written in Lisp.  I think it is really sad if NEW code is written
in C instead of Lisp for other reasons (like the author didn't know Lisp).
This tends to make code that is fragile, e.g. breaks if numeric constants
exceed 2^32-1, can't be compiled except on linux without major effort, may
break in "other" C compilers, and is a hazard generally for portability. 

Maxima compiles on allegro CL  without perl, tcl, or makefile, on both
Solaris and Windows XP.  True, a version made this way doesn't link with
gplot and describe and fancy displays. 
...........

The "history" feature below can be done by looking at data in the LISP
program. (As indicated, $playback shows how., as someone else mentioned..)
So that is really a better solution, at least at first glance.
RJF




> -----Original Message-----
> From: maxima-bounces at math.utexas.edu [mailto:maxima-
> bounces at math.utexas.edu] On Behalf Of Daniel Lakeland
> Sent: Thursday, December 07, 2006 7:23 AM
> To: maxima at math.utexas.edu
> Subject: Re: [Maxima] some editing tools for maxima
> 
> On Thu, Dec 07, 2006 at 01:45:11AM -0500, sen1 at math.msu.edu wrote:
> > This works great. Thanks, Robert.
> >
> > Now, how about a 'history' like function.
> >
> > I know one can use 'save' and 'stringout' to write commands,
> > functions, etc to a file.
> >
> > But, if one used a certain 'plot' command, or 'solve' command, it
> > would be nice to call it up again without
> >        saving to a file, searching in the file, and then cutting and
> >        pasting to get it back into the current session.
> 
> I think the way that most people deal with this issue is by using one
> or another of the UI packages on top of maxima. In particular I use
> both wxMaxima and the maxima interactive mode in emacs. On Debian
> systems these packages are called "wxmaxima" and "maxima-emacs"
> 
> 
> 
> 
> --
> Daniel Lakeland
> dlakelan at street-artists.org
> http://www.street-artists.org/~dlakelan
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima