Re: Inexpensive / free-ware similar to IDL?

[luthi(at)aura.ethz.ch (Martin LUETHI GL A8.1 2-4092)]

I was looking at several free alternatives to the IDL / PV-Wave products for
quite a time now and found these packages, which seemed to match requirements
of ease to learn, ease to use, yet providing a strong programming language:

You will find a good overview at
http://stommel.tamu.edu/~baum/linuxlist/linuxlist/node34.html#numericalenvironments.
Programs that I evaluated a bit closer are Octave and Python, which I found at
least equally convenient as the IDL/PV-Wave family of products. 

o Octave: Mostly compatible to Matlab
          GNU GPL License 
	  good plotting routines and convenient for most purposes

          http://www.che.wisc.edu/octave/

o Python: A very strong, cleanly designed interpreted language for nearly
             every purpose, very well documented, good newsgroups
          Many data types, OO-design, extendable language
          Command line and stand alone programs
          Free to copy, use and enhance for every purpose (open source,
            most modules are GPL'd)
          Versions for virtually any platform (all Unices, Win32, Mac..)
          Strong support for linking programs written in C and Fortran
          Native support for internet access (ftp, http, etc.)
          Very good numerical libraries which provide the power and 
            flexibility of IDL/PV-Wave
          Integrated plotting through external programs: Gnuplot, Dislin etc.
          A own plotting library is in currently in development
          Close integration with the Tk-widgets (real cross-platform), KDE and
          Qt-libraries and GTK widget set
          Python is widely used as scripting language and thus extremely well
          supported. My personal view is, that with the native plotting 
          routines Python is optimal for data anlysis, number crunching
          and nearly every other computing task (there is an Illustrator-like
          interactive drawing program written largely in Python)

          http://www.python.org          
          http://starship.python.org/~hinsen

As I finish my Phd I will switch to Python, which is most convenient for my
purposes and, due to its wide use in modern computing tasks, is well sought
after knowledge in the industry. The sad thing about this is, that the large
amount of my programs written in PV-Wave will be further useless for me. The
good thing is, that I can fix eventual bugs myself, not having to rely on a
tech support, which will "raise the scoring in the bug report list", even if
the bug is near fatal to the usability of the product (cf. the plotting
inaccuracies reported some months ago on this list, the discussion about the
numerical libraries aso). I hope that these high quality free products are a
real challenge to VNI and RSI, which should think seriously about the quality
of their products as primary goal (as opposed to the amount of features they
provide). My feeling is, that both products should merge in order to survive
in this very competitive, rapidly evolving marked. The benefit to us users
would be the joint ongoing development of both languages (OO-techniques,
interactive plotting and netCDF-routines from IDL) as well as the very
convenient and professionally implemented libraries (IMSL-numerics and GT-grid
from PV-Wave).

Thanks to the members of this newsgroup for the intersting discussions and
the highly appreciated help and expertise and to Carsten Dominik for his
excellent IDLWAVE.el emacs-mode.

Cheers

Martin

-- 
============================================================
Martin Luethi			Tel. +41 1 632 40 92
Glaciology Section		Fax. +41 1 632 11 92
VAW ETH Zuerich			
CH-8092 Zuerich			mail luthi@vaw.baum.ethz.ch
Switzerland
============================================================