(no subject)

[Chad Walstrom]

I ran into this interesting piece of information in the automake
manual:

    6.3 An Alternative Approach to Subdirectories
    =============================================

    If you've ever read Peter Miller's excellent paper, Recursive Make
    Considered Harmful
    (http://www.pcug.org.au/~millerp/rmch/recu-make-cons-harm.html),
    the preceding sections on the use of subdirectories will probably
    come as unwelcome advice.  For those who haven't read the paper,
    Miller's main thesis is that recursive `make' invocations are both
    slow and error-prone.

The article contains a lot of good information, and I'm mostly
convinced to take this route.  I'm testing a variation of the patch I
sent out to see if it corrects some of the build issues I had and will
let you know how it turns out.

If anyone has opinions one way or another, let me know.  The only
down-side I have seen so far is that some of the path names get to be
a bit long, so I'm using variables to store that info.  The layout I'm
using is as follows:

        root
         |- Makefile.am
         |- gnats
         |  |-gnats.am (included in Makefile.am)
         |  `ds-file
         |    `-ds-file.am (included in root/Makefile.am)
         |- doc
         :  `-doc.am (included in root/Makefile.am)
         .

Again, I have not committed any of this to the CVS tree yet.  When I
get a working build, I'll post another patch for review.  With no
recursive calls to make, I'm hoping to see a bit of a build
performance enhancement. ;-) (I'll post results of my tests.)

I've also updated configure.in to configure.ac and corrected the calls
to obsolete M4 macros. (This *will* get rolled in to CVS, regardless
of how we decide to do the Makefile.am's.)