Path: utzoo!attcan!uunet!bu.edu!purdue!mentor.cc.purdue.edu!noether.math.purdue.edu!jlh
From: jlh@noether.math.purdue.edu (Jeffrey Hensley)
Newsgroups: comp.sys.amiga
Subject: Re: Scientific programs on Amiga
Message-ID: <14600@mentor.cc.purdue.edu>
Date: 30 Sep 90 23:36:20 GMT
References: <1990Sep29.021658.26896@evax.arl.utexas.edu> <6669@uwm.edu> <90273.104606UH2@psuvm.psu.edu>
Sender: news@mentor.cc.purdue.edu
Reply-To: jlh@noether.math.purdue.edu.UUCP (Jeffrey Hensley)
Organization: Purdue University, West Lafayette
Lines: 44

In article <90273.104606UH2@psuvm.psu.edu> UH2@psuvm.psu.edu (Lee Sailer) writes:
>Here's a different slant on the lack of Scientific Software on <insert machine
>here>.  The Unix community, and to a lesser extent the general Comp Sci
>community have a tradition developed over the past 20 years of sharing
>locally developed software.  For example, if you create an experimental
>text editor for a machine, you write a manual, bundle the whole thing up,
>and try to export it.  I think this is because for many people in this
>community developing software is the *end product* they want to be recognized
>for.
>
>On the other hand, this tradition is very rare in the sciences and engineering.
>When some scientist develops a program that allows a user to view a data cloud
>in N-dimensions, the *last* thing that occurs to him or her is to go on and
>add a user interface and write a manual, and so on.
>
>There are exceptions to this, of course, but as a general trend I doubt you
>will see many good scientific applications floating around free because
>scientists aren't motivated to produce them.

  I would tend to disagree with this generalization.  Perhaps you have never
  needed a "Scientific" package, but there are LOTS of packages out there
  in the public domain.  Have you ever heard of Netlib?  Source code is
  available to solve a large number of problems.  Finite element packages,
  finite differences, ODE solvers, plotters, special function routines,
  etc. etc.  

  What you perceive as a rarity may be due to the fact that you just haven't
  run across this stuff.  I might as well say that there is a lack of source
  code available for X applications :-)

  One difference is real, and this probably explains why there is an
  apparent lack of AMIGA scientific packages - most scientists have access
  to machines at work which handle things for them (most notably, plotting
  routines).  I am not going to try to use my A2000 at home to plot 
  when I can go to work and do it better & faster and where I have a 
  laserjet printer hooked up (and for many people change that to an
  actual plotter).

  Since there is less of a software base for the Amiga (as compared to
  Macs) it isn't surprising to find a lack of software aimed at the sciences.
  An exception is Maple, but then the hope there is that enough schools will
  use Amigas that they can make a nice profit.

  --Jeff