[crossfire] Leaderships(s?) (was Re: Platform statement)

[Kevin R. Bulgrien]

> > I have not enough SVN access to do a release.

That would be sourceforge access.  No upload, no release.  Last time
I commented I had taken the pain out of building the client, I was told
the hassle is the SF stuff, so the neverending tale goes on.  If noone
will upload, all release effort is wasted.  I've gone out of my way to
facilitate a release and the run-around needs to stop. Every time I ask
for some support, some more load of hooey pops out as yet another
reason not do it.  I'm not wasting my time until it looks like someone
is serious and says what needs to be done for them to upload, or I get
the access it takes. 

> You have commit access.  That's enough access :-)

No, its not.  SVN write doesn't get the poop on the download tab.

> if you want to do this,
> what I'd expect you to do is decide what will go in the release, and
> merge anything that needs to be merged.  (If you just decree that your
> client release == trunk, then your work is done.)

There's already consensus on this, and as far as I know, no one has
argued against.

> If you *do* have time you want to spend on it, then any amount of
> building, testing, and fixing bugs is welcome.  If something happens to
> draw you away, I'll handle it, or someone else will, or we'll be late.
>
> If I feel adventurous and work allows me, I may try to set up the windows
> build environment, and see if I can convince gtkv2+glade to compile.  But
> from what I heard of people who did try, probably not ;-)
>
> Of course, once the alphas/betas are out and bug reports start popping
> in, someone has to fix them before we dare make a "final" release.  Since
> not many people are actively doing client development, I'd say there's
> about a 35% chance that someone would be you anyway; but if real life
> keeps you away from doing it, or even makes us miss the target date?  It
> doesn't matter at all.  What matters is that a release eventually
> happens, and that it's the best quality we can do.

2.x client isn't going to start popping bugs...  people are using it.  If it
has issues, they need not stop a release.  The current releases have
more/worse bugs than the trunk gtk-v2 client anyway.  I am somewhat
confused about the question of working on it.  Most of the recent client
commits in SVN are mine...  I'm not promising anything, but the proof
is in the pudding.  I do not plan to be gone from the project anytime
soon.  All this worry about debugging seems kind of odd... Can anyone
say double-character bug... showstopper...  Can't get any worse than
that.  Do you know how hard I worked on that one?

> I do believe regular releases on a fixed timeline are a good thing.
> However, I'm not expecting *this* release to be the first of those.  We
> have a lot of accumulated work to catch up to, and we have to figure out
> our own workflow, so that will take some time.  Better do it well than
> fast.  Then the next one can be on time.
>
> The other good thing about volunteering to lead the client are that,
> well, you get to make decisions.  Want to drop the x11 client?  The gtkv1
> client?  Go ahead.  Change the whole thing to C++, or, I don't know,
> Erlang?  It's your baby.  The last few decisions you did make (glade,
> layouts, themes) have been successes so far, so I don't think there's any
> reasonable justification for objecting to you making future ones.

I don't need a title.  I have been doing it.  What's the deal with
"volunteering".  That got done years ago.  I never unvolunteered,
but I also know that mwedel wrote it, and am not that interested
in claiming something not offered.  But the point is also I don't know
enough to even write a client, so how can I commit to be "the" guy
when others are changing stuff I have no clue about the server/protocol/
or the tall stuff being worked on in jxclient?  I can try, but that's all I
can commit to, and I guess I don't see a lot of people out there making
noise about keeping the GTK clients alive, so it seems I'm kind of out
there on my own.

> I suppose, in the interest of fairness, I should say that I've been half-
> secretly writing a "libcfclient" in C++, using boost and asio.  I have no
> intention of rewriting the desktop client using this lib, or of competing
> with it, though; the lib was meant for other things, including bots and
> maybe portable (phone/pda/pmp/ds/psp) clients.
>
> best,
>                                                Lalo Martins