MathGroup Archive 2005

[Date Index] [Thread Index] [Author Index]

Search the Archive

Few notes

  • To: mathgroup at smc.vnet.net
  • Subject: [mg53962] Few notes
  • From: bearophileHUGS at lycos.com
  • Date: Sat, 5 Feb 2005 03:15:45 -0500 (EST)
  • Sender: owner-wri-mathgroup at wolfram.com

This is probably my first post in this interesting newsgroup. I like
Mathematica (mostly as a programming/prototyping/visualising tool) for
its consistency, for all the builtin goodies, and for its graphics
capabilities. I also love Python (and other things/languages), and I
think there are important similarities between Mathematica (M.) and
Python...

I'm a kind of  newbie in M., but there are few things in it that I
don't like much, that I think they can be improved:

- The built-in M. editor is good to write formulas, to see graphics,
etc, but it's far from good to insert small programs... And it's even
worse for a programs a little bigger (for them I use a simple text
external editor). A simple text editor is often better for me, than the
builtin one. Adding colours (and few other simple "modern" programming
aids) looks useful to me:
http://www.mma-users.org/mediawiki/index.php/Wish_list
Such colours are quite useful during programming.

- I usually use M. on a Windows system. In all windows programs after
the selection of a piece of text the user can move the cursor, to
increase or decrease the selected part. Mathematica frontend is the
only program that I know of, that allows only to increase that size
(even using the back cursor key). I don't know why this design choice
was adopted, maybe for uniformity across different implementations of
the M. frontend on different operating systems, but I think this isn't
enough. On a Win/Mac/etc there are interaction/GUI conventions that all
programs must respect.

- For Python there is a 2D plotting library (with some nice data
processing capabilities inspired by):
http://matplotlib.sourceforge.net/screenshots.html
Its much simpler (and probably quite worse) than the normal good M.
graphical capabilities, but the graphics it produces is very smooth
(anti-aliased), because it comes from the very fast, very good (and
free) AGG library:
http://www.antigrain.com/
http://www.antigrain.com/demo/index.html
In my old version of M. graphics isn't anti-aliased, is this a design
choice? Do new M. versions produce antialeased graphics?

Thank you,
bear hugs,
Bearophile


  • Prev by Date: Re: multiprocessor vs cache/memory
  • Next by Date: JLink
  • Previous by thread: Re: NDSolve and differential equation system
  • Next by thread: JLink