gnu.* namespace discussion

[Warren Levy <warrenl at cygnus dot com>]

Folks,
As part of the effort to combine works of the Classpath and libgcj
projects, I've been folding in things like java.beans and object
serialization into libgcj.  In doing so, I noticed the potential for
namespace confusion and/or conflicts in the gnu.* hierarchy.

Currently, libgcj keeps implementation details of the JDK classes we
implement in a gnu.gcj.* hierarchy but comparable code for Classpath seems
to be in a gnu.java.* hierarchy.  As these 2 conventions seemed 
inconsistent with each other I was originally tempted (and started) to
move things into the gnu.gcj hierarchy when bringing them into libgcj.

Very quickly I realized this is bad for (at least) 2 reasons:  1) it means
extra overhead when bringing the classes over (along with extra overhead
in trying to keep them in sync), and 2) it mixes things together that
might be easier to track if they were kept separate.

My feeling at this point is that a gnu.classpath hierarchy would be more
appropriate than a gnu.java one.  I realize that would mean a change to
the existing structure and that is not fun.  Is there a better approach?
I'm sure that there are other aspects to this that I'm not bringing out
here.  Please bring out any issues you see associated with this for
discussion.

Warren Levy