[Ffmpeg-devel] compiler warnings

[Diego Biurrun]

On Thu, Mar 30, 2006 at 03:16:48AM -0500, Rich Felker wrote:
> On Thu, Mar 30, 2006 at 08:57:07AM +0100, M?ns Rullg?rd wrote:
> > Steve Lhomme <steve.lhomme at free.fr> writes:
> > 
> > > That's why the famous quote from here "fix the compiler" is also not
> > > realistic in the real world. Most of the time you are given a compiler
> > > (and an IDE) that goes with the hardware you're working with (I'm not
> > > talking about PCs) and you just do with it. You'll just get fired if
> > > instead of fixing bugs you were saying your boss "it's not my fault,
> > > it's the compiler". It's an engineering practice that will get you
> > > nowhere.
> 
> Things are the opposite when you work with free software. You can't
> expect people to be happy when you try to impost your bad practices
> from dysfunctional corporate environments on us. If a compiler/system
> is hopelessly broken, the best thing to do is not to support it and
> encourage people to switch to one that works.

Besides, with free software it is actually possible to fix the
environment instead of piling one workaround on top of the other.

I have sent a patch to Cygwin to fix their /proc emulation instead of
working around the bug in MPlayer configure.  It got accepted.

There is no guarantee for this strategy to work in all cases of course
and I'm also not saying that Steve does not have a point, especially in
situations where you are stuck with certain tools.

But in our situation you can modify not just your code but also your
environment.  This also explains why proprietary toolchains that don't
allow this are loathed so much around here.

You see, this coin really does have two sides.  Fixing bugs in the right
place and refusing workarounds causes short-term pain but promises
long-term improvement...

Diego