Ok, I know I've brought this up before, but didn't seem to get a
satisfactory answer. 

What exactly is the purpose/goal(s) of maintaining a seperate unnoficial
linux distribution of csound?

Ah, well, I may as well just go ahead and explain my perceptions as a
casual user of Csound on the Linux platform.

I haven't used Csound for quite a while on my Linux box, and hence
decided I was in need of an upgrade. Naturally, my first thought would
be to download John Fitch's distribution. I did, and although it did
build, I noticed that it did not have the code for grabbing POSIX
real-time priority. Fine, I thought. That's what the unoficial Linux
version is for. I grabbed that, and found the build process quite
different. So, rather than having the comfort of having the same
makefiles as the official version, everything was quite different, and
in fact I ran into some glitches with the unnoficial version that I
wasted some time on to resolve. It is not my point that one of these
versions is better than the other; rather, the problem is that they are
different, and hence, to install both versions entails debugging the
quirks of two entirely different build mechanisms. 

Well, once I got the unofficial Linux distro up and running, I naturally
wondered (once again) why there needs to be a seperate distribution in
the first place. And I was quite perplexed as to why, for example, the
Linux/POSIX real-time scheduling option was not submitted into the
official release. If it were, I would not have had to venture out and
learn to install a seperate version of Csound.

The most basic problem I find, as a casual user, in running a
non-official version, is how to deal with apparent bugs that pop up!
It is not at all clear to me how the sources betweeen the two versions
differ, and hence when I come across a core dump or other anomaly (which
with csound being as buggy as it is, is guaranteed to happen in *every*
session), I have no immediate way to determine whether this bug is
specific to the unoficial version, or is common to the real version. I
am instead forced to grep and browse through the sources for both
versions in an attempt to learn whether this is a generic bug or not.

Ok, enough ranting, by now I hope my point is obvious: It is unworkable
for Linux users of Csound to have to use a different source tree and
build method, just to get their platform-specific advantages. Instead,
you Linux developers really need to talk to John and come to some sort
of agreement on including the Linux specifics into the standard
distribution.

Larry