Re: [AD] v/h flip for draw_lit/trans

[Elias Pschernig <elias@xxxxxxxxxx>]

On Wed, 2007-03-07 at 09:47 +0100, Evert Glebbeek wrote:
> On Tuesday 06 March 2007 23:56, Elias Pschernig wrote:
> > I agree in principle. I just think, 4.3 isn't ready for it yet.
> 
> Quite right. But I don't think it's going to get there if we keep tweaking and 
> adding things like this to 4.2. I think 4.3 not being ready for this should 
> be more of a motivation to work on 4.3 than to put these things in 4.2.
> 

Yes, but right now, 4.3.0 simply can't be used for the things which are
possible with 4.2.1. So essentially, until 4.3.1 or even a later version
is out, the 4.3.x branch can't really take new features - as a new
feature likely means, who implemented it needs to use it immediately.
Now, if we also put 4.2.x on a no-new-features policy, we essentially
put all of Allegro on hold. And putting a library on hold for years
won't exactly look attractive to people who want to contribute.

Now, one step around that would of course be quickly getting out a 4.3.1
which *does* change the not-useful state. For that, it would at least
need to support the common file format addons, and especially OpenGL. We
even might want to have it work together with things like OpenLayer. To
me, this doesn't really look feasible.. either we maintain compatible
versions of some addons, or try to stay compatible with addons, or merge
in addons where license permits. All messy, and keeping us from the real
ground work on the 4.3 API.

The other route would be, let 4.3.x be in this on-hold state and
secretely work on implementing new stuff (e.g. if nobody is faster,
eventually I will finish this xdummy branch stuff). The 4.3.x branch
that way gradually would get more and more useful, and around 4.3.3 or
4.3.4, hopefully surpass 4.2.x. Until then, we'd allow new features in
4.2, so users (= possible contributors) stay happy.

That's my thoughts at least..

-- 
Elias Pschernig