Re: Re: These NeXTbuntu guys

[Nicolas Roard]

Hi all,

On 8/28/06, Joachim Schulz <joshu0011@yahoo.com> wrote:
> GJC,
> I'm all for cooperation. However, that doesn't mean
> beginning a dialogue with a gesture of submission.

Well nobody is talking about a gesture of submission :-)

The dialogue was initiated (well, kind of, considering it wasn't
posted on the ml but on a blog) by stating a string of facts that are
erroneous about GNUstep; of course people posted answers to rectify
that (and frankly, the answers seem just informative to me, there's
nothing in it asking for a "gesture of submission", apart stating that
GNUstep is much more advanced than rick thinks, and that creating a
"nextbuntu" is a huge undertaking no matter what; and thus it would be
much more efficient to work /with/ gnustep, not against it. Seem to me
some persons are a bit edgy (possibly on both sides, although nothing
said from the gnustep guys strike me out as rude -- but hey, I'm
possibly biased).

> Our intent was not to ignore GNUstep. I agree, our
> first contact was more like a fender bender.

uhuh... seems so indeed ;-)

> But clearly, GNUstep is not ready for prime time, not
> after 10 years of hard work and having access to all
> kinds of legacy code. This is baffling to me.
> Maybe it is time to step back and take a fresh look at
> the whole picture. Maybe a spirited collaboration can
> get things moving that neither party could do on their
> own.

Well, yes and no. The situation isn't as clear cut as you think. For
once, as stated before, gnustep _is_ more advanced and complete than
you seem to think. Don't mistake the outside (the UI look) with the
inside (the actual capacities of the code).

Now, you are of course also right -- after ten years, gnustep doesn't
seem to be "ready for prime time" (although you'll need to define
precisely what's "prime time", and what exactly you want to do with
gnustep, as in many cases it works just fine).

It is certainly lacking on the AppKit side (which of course is what
you're looking into). I mean, it's not _that_ lacking, which was the
point of these answers on the blog, and not as lacking as you all seem
to think it is. But yes, it is not as polished as it should (must) be
-- I think everybody agrees here.

The reason is dead simple: there's actually very few persons working
on gnustep-gui currently. And of course, reimplementing AppKit is a
piece of work :D

I'm not sure about the "all kinds of legacy code" the gnustep project
is supposed to have used, but one thing is sure, is that there was
indeed a lot of hard work put in it.

Which is why, again, people advocated you to have a deeper look at
gnustep and possibly to help working on it -- certainly a better path
if you want a cocoa-like environment. It's not exactly some code you
could replicate in 2 months.

GNUstep is under the LGPL, not the GPL, and as such there shouldn't be
any kind of licence problem, whatever you want to do with it.

Ok, now that was for the bad news / summary... but here is the good
news: GNUstep isn't as polished as it should be, indeed, but the code
is rather clean most of the time, not too difficult to understand, not
too difficult to modify, the different parts of GNUstep are quite
cleanly separated, etc. And the best thing is that indeed most of
what's lacking is polishing. I.e., I strongly believe that even with
just a few more dedicated persons coding on gnustep-gui on a regular
basis, the impact will be tremendous, in a very short period of time.
I think you'll be surprised at how easy it is to actually improve
gnustep a lot with just a few additions ;-) -- gnustep's problem is
more "if only there was more regular committers"...

So, well, do not take that as asking for a "gesture of submission",
but please, have a deeper look on what gnustep can and can not
actually do... (there's videos demonstrating gnustep/gorm, check those
at least).

Cheers,

--
Nicolas Roard
"I love deadlines. I like the whooshing sound they make as they fly
by." -- Douglas Adams