[This paper, by the way, applies the SICM methods to the subject of
differential geometry:
http://swiss.csail.mit.edu/users/wisdom/AIM-2005-003.pdf ]
When I grep the source, though, I do find mention of this function, which
leads me to believe (in conjunction with the date mismatch mentioned in my
previous note) that maybe the 'memory image' does not reflect the source
code, that I need to recompile the Scheme source?
At some point in the past, I installed MIT Scheme on a Windows 2000
machine and was able to recompile the Scheme code from scmutils to have a
working scmutils on Windows. I've completely forgotten where I found the
instructions on how to do this. Presumably I just execute the code in
'compile.scm'. In the meantime: I'm curious whether anyone else here has
delved into the 'Functional Differential Geometry' material.
Tobin Fricke
--
Department of Physics, University of Rochester, Rochester NY 14627-0171
http://web.pas.rochester.edu/~tobin/
> working scmutils on Windows. I've completely forgotten where I found the
> instructions on how to do this. Presumably I just execute the code in
> 'compile.scm'. In the meantime: I'm curious whether anyone else here has
> delved into the 'Functional Differential Geometry' material.
The instructions, it turns out, were in a message posted earlier on this
list, by Jose A. Ortega Ruiz on Aug 18. I downloaded the latest scmutils
from gjs' ftp site, then used "apt-get install mit-scheme" to install a
separate MIT Scheme on my machine, and followed Jose's instructions for
recompiling scmutils and making my own band file. The result is that the
latest scmutils works with an MIT scheme that works on my machine (the
mit-scheme distributed with the 20051121 tarball dies silently).
Moreover, the differential geometry functions now are available.
Tobin