Greetings. As the subject says, I've encountered a problem using
Mathematica on a system running Scientific Linux 4.4. I can't tell if
the problem relates to X-windows (xorg vs. XFRee86), or to changes in
Mathematica, or ... Hence, this may be wildly off-topic, and, if so, I
apologize.
The issue has to do with executing Mathematica on a remote system and
getting the display on a local system via X-windows. Mathematica uses
its own fonts for its GUI. If those fonts are missing, the front-end
(GUI) is, for practical purposes, not usable.
We have Mathematica available on a bunch of SL 4.x boxes in our
classroom. I'd like to be able to connect via ssh to one of the
classroom systems from my office, run Mathematica on the classroom
system, and display the results on a SL 4.4 system in my office.
We're currently using SL 4.4 on our classroom systems and the system in
my office. The version of Mathematica is 5.2.
In the not-too-distant past I've been able to do this combination of
remote execution/local display by using the cheap, but, if I understand
correctly, entirely legal trick of installing the Mathematica fonts, and
no other part of Mathematica, on my local computer in exactly the same
location in which the fonts are located on the remote system.
I.e., if I do:
local# mkdir -p /usr/local/Wolfram/Mathematica/5.2/SystemFiles/Fonts
then put all the Mathematica fonts in that local directory, then I have
been able to do:
local$ ssh -X remote
remote$ Mathematica &
and have the full Mathematica front-end appear on local.
Note that this worked without my having to modify any X-windows
configuration at all! No "mkfontdir", no "xset fp+", etc., etc.
Now what happens is that I get the input window, but I don't get the
associated "palette", which contains all the integral signs, square-root
signs, etc. At some level Mathematica DOES seem to be finding the
fonts, because if I take a quasi-command-line approach and start typing
things into the input window, the function names are in bold, the
brackets and parentheses are automatically highlighted and matched, and
if I type something like:
Integrate[Sin[x^2], x]
I get legibly-formatted output with Pi's and square-root signs looking
more or less the way they look in math books.
If you know where the palette went, or if you have any other words of
wisdom about this, please let me know. Thanks.
- Mike
--
Michael Hannon mailto:[log in to unmask]
Dept. of Physics 530.752.4966
University of California 530.752.4717 FAX
Davis, CA 95616-8677
|