[Visualpython-users] Red Hat Linux

[Bruce S. <ba...@an...>]

Roger Fearick sent this interesting and informative note and asked me to
forward it to the VPythyon user list, which had forgotten to do. Thanks
much for the information, Roger, both on Red Hat and on your use of VPython
in physics education. We will put your Red Hat information on the VPython
web site.

---------- Forwarded Message ----------
Date: Tuesday, November 21, 2000, 3:16 PM +0000
From: Roger Fearick <FE...@ph...>
To: Bruce Sherwood <ba...@an...>
Subject: Re: [Visualpython-users] Red Hat Linux

> Hi Bruce,
> 
>> Thanks much for the file. Could you possibly write a brief paragraph
>> aimed at Red Hat users, to accompany making the file available on our
>> web site? Not being a Linux user, I couldn't follow your instructions
>> well enough to know exactly what to say. Thanks.
> 
> How I got VPython working on Redhat 6.2
> =======================================
> 
> There seem to be two problems:
> 
> 1) The cvisualmodule.so supplied in the VPython linux distribution
> makes calls to a version of a C++ dynamic library that is not
> compatible with the version on RH 6.2.
> 
> Solution: I have recompiled cvisualmodule.so: just copy the recompiled
> version over the original.
> 
> 2) There seems to be some thread-related bug in RH6.2. I couldn't even
> get the thread examples in /usr/doc/glibc-2.1.3 to run.
> 
> Solution: I applied two updates (probably only the second is needed,
> but the first is worthwhile anyway):
>     a) I updated the c library.
>        See http://www.redhat.com/support/errata/RHSA-2000-057-04.html
>        for details and instructions.
>        Probably only the file
>         ftp://updates.redhat.com/6.2/i386/glibc-2.1.3-21.i386.rpm
>        is needed: it can be installed by running rpm -Fvh [filename]
>        as superuser.
>     b) I updated the kernel to 2.2.16.
>        This is probably the essental step.
>        See http://www.redhat.com/support/errata/RHSA-2000-037-05.html
>        for details and instructions.
>        Read these carefully if you haven't done this before !!!
> 
> After this, VPython ran.
> 
> 
> 
>> P.S. I see you are in physics. What is your (physics?) interest in
>> VPython?
> 
> I've been running an introductory course on computational physics for
> our senior students. There is usually a wide range of programming
> skills to deal with (down to 'never programmed before')., and I
> decided this year to try Python as the 'language of record' in which
> I gave my examples. (Students are free to use other languages if they
> are more familiar with them.)
> I wanted something that would be freely available to students so that
> they could install software on their own computers. I have previously
> used Lahey ELF90, a Fortran 90 subset, but the freely downloadable
> version of this was discontinued in the middle of last year, so this
> year I tried Python (using Numeric of course).
> 
> I have been pleasantly suprised how easily my F90 code has translated
> into Python, slices and all. The students have generally found it
> easy to program in Python (after the initial "Python? What...?").
> 
> Graphics has been a bit of a problem, but graphs can be drawn with
> dislin, and Tkinter can be used although it requires constructing
> a GUI application to do anything, so it's rather messy.
> But it's nice to have some animation, so I showed them how to collide
> a wavepacket with a barrier in Tkinter.
> 
> VPython on the other hand makes this sort of thing essentially
> trivial. It is really an incredible package which shows how graphics
> should work in a threaded environment. So congratulations and thank
> you for making it available.
> 
> It arrived a bit late to use in this year's course, but I'm sure that
> I will make use of it next year (when the course is moved down to a
> more junior level). Nevertheless I've had fun animating the numeric
> integration of the solar system, pulses on a string, etc.
> 
> I'm sure it will also make an appearance in my first-year mechanics
> lectures next year: I can see many little programs along the line of
> your cross-product demo being useful to illustrate things.
> 
> Roger.
> 
> --
> Roger Fearick
> Department of Physics
> University of Cape Town
> 

---------- End Forwarded Message ----------