Hi James - My understanding, from my experience with Cygwin, is that the
Windows Maxima binary (the one currently available from math.utexas.edu) was
constructed using the "mingw" version of Cygwin. I haven't used mingw, but
from what I understand it is a minimal implementation of the system
including the complete (I hope) C compiler chain. Some of this is in fact
installed the Maxima binary itself, as you will see if you poke around in
the subdirectories. I don't see why one couldn't continue to use this
approach for generating Maxima binaries, but it would of course have to work
with all of the recent modifications and improvements. Also, it is worth
mentioning that the Cygwin/mingw gcc compiler is a bit idiosyncratic in that
it doesn't exactly follow the corresponding thing in a Linux installation
(eg., header files). This of course leads to the current problems producing
GCL and Maxima. I expect as well that if you want to generate Maxima
complete with a graphical interface, you will also need a compatible and
consistent Xfree86. Please forgive me if this all sounds a bit elementary,
but I do speak from experience...
Dave Holmgren
> -----Original Message-----
> From: James Amundson [SMTP:amundson at fnal]
> Sent: Wednesday, December 12, 2001 8:05 AM
> To: dan.stanger@ieee.org; Maxima List
> Subject: Re: [Maxima] Windows binaries for maxima-pre59
>
> On Tue, 2001-12-11 at 12:37, dan.stanger@ieee.org wrote:
> > When pre59 is released, is a windows binary planned? I think that
> > we would get much wider use if we do.
>
> There are two issues here. I will address them separately:
>
> 1) Window binaries. I have to confess that I haven't even looked at how
> to generate maxima binaries on windows. (Is the procedure even
> documented? Does it require cygwin?) I definitely would like to have
> windows binaries available, but we need a volunteer to produce them.
>
> Any volunteers to produce the window binaries?
>
> 2) The release plans. The goal is to have a stable, well-documented
> Maxima 6.0 that works with multiple lisp implementations. To that end,
> we will have a series of development releases starting with 5.9.0. The
> development series releases may not be stable enough for everyday use. I
> hope that 6.0 will be distributed far and wide. I do not want to promote
> the 5.9.x releases to new users, lest they be frightened away.
>
> Best,
> Jim
> _______________________________________________
> Maxima mailing list
> Maxima@www.math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima