Mark IV reduction pipeline -- not quite ready yet

[Stupendous Man <richmond@a188-l009.rit.edu>]

  I've been working for the past few months on a pipeline for 
turning a CD-Rom of Mark IV images into a small set of ASCII
text files with measured stellar positions and magnitudes.
It's _almost_ to the point that I can run it through without
intervention, but there are a couple of steps where I have to do
a little hand-editing of some intermediate files.  The astrometric
calibration step fails for some images, too -- this might require
just a bit of parameter-tuning, or it might require some additional
information (or more accurate information) be written into the 
FITS header of the raw images.  

  Anyway, I want to make the software available to others.
Even if no one else can make it work, it is certainly helping 
me to document the software properly.  

  As a first step, I have tried to gather together information on
the 5 different packages which go into the pipeline.  What I've
done is to build upon some "legacy" software I've written in the
past, using the TCL language to glue pieces together.  It's not
an ideal way to create a product for other people to use, but it
was less work for me ... so, I did it.  Now, perhaps, I'll find out
if it's any use to anyone else.

  What I'd like interested parties to do is to try to download one
or more of the packages, and tell me

        a) could you download the package?
        b) could you un-gzip and un-tar the package?
        c) could you compile and build the package?

  Only one of the pieces (the "match" package) has a self-test built
into it, alas.  So it won't be obvious if the software will run
properly on your system, even if it does compile and link.  But 
I'll bet you can help me to find errors and warnings in the process
of trying to build the software, so I can fix them.

  If you're interested, it will take a computer running Linux or 
some *nix-like operating system, with an ANSI C compiler.  One
of the packages uses the X Window system for a small fraction of
its programs.  Another package uses a open-source mathematical
library called "ccmath", which until this afternoon was available
via Freshmeat.org (but its homepage has disappeared since then,
it appears -- temporarily, I hope).

  You can find my first attempt at instructions for acquiring 
the 5 packages at

       http://spiff.rit.edu/tass/pipeline/pipeline.html

  This URL points to a document which will _eventually_ describe
how to run the pipeline; at the moment, that portion isn't written.
But the first part of the document, which describes the auxiliary
packages and where to get them, _is_ written.  I am looking for
feedback on that part alone, right now.

  In a few weeks, I may finish writing the part which describes 
how to run the pipeline.  I'd like to also create a small subset
of one of Tom's Mark IV disks, and 

        a) use that subset to test my instructions on how to run
                 the pipeline
        b) create some kind of "self-test" package based on this
                 data, which would allow other users to verify
                 that the entire pipeline runs properly on their
                 systems

  Part a) won't take so long, I think.  Part b) is more ambitious.
But it's part of the eventual goal.

  The HTML documentation for several of the packages occasionally
states, 
 
        " ... the current version has obvious problems here.
          These problems will be fixed in a future version ..."

  I welcome all constructive criticism :-)

                                      Michael Richmond