This is the mail archive of the java@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: GCJ and Sun going GPL (again)


>>>>> "Marco" == Marco Trudel <mtrudel@gmx.ch> writes:

Marco> Tom, what are your plans, what do you think?

I am extremely happy that Sun has done what they've done.  It is, for
me, the culmination of a 10 year dream -- to have a free software Java
implementation.  The fine folks at Sun deserve all kinds of praise and
adulation.

I think it is a little early to measure its impact on gcj.  For one
thing not all of the code has been released yet.  Currently only javac
and hotspot are available, not the class libraries.

I do think that our long history has shown that it is preferable to
work on a single good implementation than multiple competing ones.
Every time we've merged code bases -- first with Classpath, then later
with kaffe, etc -- quality has gone up, coverage has gone up, and the
developer pool has grown.  So, I tend to think that we'll want to work
closely with OpenJDK.

Hopefully this won't cause too much dismay but I do tend to think that
Fedora, for instance, would be best served by shipping OpenJDK as its
primary java implementation.  There are still some issues with this --
not just waiting for the remaining code to be released, but also
platform coverage (at least, I thought Fedora supported PPC.  I have
trouble remembering).  So, it probably can't fully replace gcj even
once all the source is available.

Meanwhile we know that for Fedora 7 we still need to move forward with
gcj.  We plan to ship a generics-based release there.

At this point I think the best thing to do is be patient.  We will
know more in 6 months.

Marco> The other thing is the gcj/ecj/javac thing. There, I can't tell much
Marco> about, because I was never directly involved in these. But I think it
Marco> would make sense to use everything from Sun instead from different
Marco> places. But again, I don't know how big the advantage would be.

Looking at this would definitely be useful -- not just in a purely
technical way but also as a way to start cooperation.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]