Re: [java-gnome-hackers] Proposing Java-Gnome-4.x for GNOME

[Andrew C. <an...@op...>]

On Tue, 2008-11-04 at 05:51 +0200, Serkan Kaba wrote:
> Can we make a proposal (in this short time) or should be postpone it
> to next GNOME release cycle?

java-gnome, as a project, has been an "official" GNOME bindings since
sometime in 2004.

My personal feeling has always been that just because the java-gnome
community moved on by consolidating the project as it was at the state
of version 2.x to a new codebase at version 4.0, as a project nothing
essential has changed.

I've long been a bit annoyed that release-team hasn't got around to
updating the set of what they put out, but on the other hand it matters
little because what is important is that distros carry the software. And
most do. Thus if release-team changes its set, then we're done.

So I am *supremely* ambivalent about "proposing" java-gnome for GNOME
because java-gnome is already a part of GNOME.

I suppose I can get over that ambivalence. Frankly, what I really want
is someone else to lead the charge; I can't do all the work. People like
Serkan speaking up is a big step in the right direction, but others need
to as well.

Part of the problem in this community has long been that people that are
happily using the library tend not to speak up or talk to me one to one.
While all that positive feedback (in the case of people talking
privately) is wonderful, it doesn't help fight the usual Open Source
problem of people chiming in who have nothing to do with a project
thinking their opinion is more important than that of the people who
have been involved in it for years. I've learned that being an Open
Source maintainer is being tolerant of such, but seeing as how no one
pays me to work on this, I must admit that at the moment I don't have a
whole lot of patience for such nonsense. I'd rather spend my time on
people who contribute code and helping newcomers who turn up in
#java-gnome and on java-gnome-developer, so that's what I do.

Anyway, not my intent to sound arrogant or anything; this is still the
same project that I reluctantly took up leadership of 3 years ago, and I
still feel a huge debt to people like Jeffery Morgan and Andrew Overholt
- serious Open Source hackers who were also Java people and wise enough
to understand the trick was to find a way through the turbulent waters
that happen when massive communities like Java, Linux (remember, most of
Java developer-land knows nothing of the Linux, Open Source & Free
Software Way) and GNOME come together. Trying to serve all three, so as
to bring new Linux people to GNOME while doing the best job we can to
provide a faithful home for existing GTK developers has been an enormous
challenge.

We've done ok, and I appreciate the hard work of all the people who have
contributed over the last few years to make this project a reality. I
for one have just about everything I need now from java-gnome. The
applet mystery is really the only thing left on my wish list, and that
can wait. There's lots of work in progress software I want to get back
to, but personally I'm taking a bit of a pause to recharge my batteries
a bit so I can come back to hacking with it being fun; certainly
fighting email battles is the last thing I want to do.

++

So, on to concrete matters, the release guys want the library to be
frozen forever.

We're actually getting reasonably close to that. More to the point, we
still have 4.2 (and 4.4, and...) available as future API breaks as
necessary, and 4.0 has been [more or less] API stable since 4.0.3 or so.
I'm still sticking to the escape clause in the warning at the bottom of
http://java-gnome.sourceforge.net/4.0/ but in essence, things have been
pretty good. It's like all of Free Software: if your code is open then
you have no problem rebuilding if we change. And if your code isn't
open, too bad. Having had java-gnome packaged since 4.0.3 and having had
a [so what if it's little] application using the library packaged there
too have taken us through most of the issues. I think we're in good
shape in release management terms. I have some doubts about the Debian
package, but ultimately that's Debian's problem.

Anyway, taking 4 as epoch and {4.0, 4.2, 4.4, ...} as major release, and
{4.0.6, 4.0.7, 4.0.8, 4.0.9, ...} as minor releases (which has been my
intent) I think we're good. Demonstrated practise and demonstrable
history and all that. And since the GNOME rules provide that major
releases are allowed to API break, I think we're actually OK and have
been all along.

I've worked insanely hard to ensure our releases have been of good
quality. I'm proud of our track record. I think it's important, and I
think it's a good legacy to the person who takes over as maintainer
after me.

++

If people who have been serious hackers on and/or developers with
java-gnome over the past 2+ (or 6+) years want to make a fuss with
desktop-devel-list, then I understand that I as maintainer have to write
the original proposal. I will certainly do that, but I'm only going to
unearth that can of worms if there is a strong group of people here -
and on that list - that will fight off the inevitable, harmless, but
nevertheless very persistent people who always like to comment in such
debates.

On the other hand, if you just want to leave it be, that's fine too.

I agree that it's probably time we had it out with release-team; I'm not
really sure that we need to go to desktop-devel-list to do so, but I'll
leave that to others here to decide.

Either way, but I'm not going to do it alone. This project has for 10
years been one person carrying the torch with others dipping in and out
from time to time. That's fine - and not uncommon for small Open Source
projects like ours that are a labour of love, but if there's on time I'm
not going to walk by myself it's fighting battles on GNOME mailing
lists.

After all, *I* don't need a release to use java-gnome; *neither do you*.
That's something Rusty Russel taught me: doing releases at all is itself
a massive gesture and contribution to stability for other people - work
that doesn't do *you* nor *I* any good. Going one step further and
having debates in the broader GNOME community is a **huge** operation. I
will absolutely do my part as maintainer, but only if enough people are
going to be there to fight the battle with me.

So my friends, what's it to be?

AfC
Sydney