diego at biurrun.de
Wed Dec 31 16:06:17 CET 2003
On Tue, Dec 30, 2003 at 06:30:47PM +0200, Ivan Kalvachev wrote:
> I told you that --enable is idiotic.
> I was the one that wanted to change --enable to meen autodetected
> (I also proposed --force to meen force), but I was threaten that
> my cvs write access will be taken away if i do such thing.
Well, hasn't everybody been threatened wth that some day? ;)
It's pretty clear that this is a very disruptive change that should be
discussed on dev-eng first. IIRC you were going to do the "speak up or
I'll commit in a few days" thing.. But let us not linger on this subject.
I agree that --force is more intuitive than --enable and I would actually
support such a change, but it is disruptive and backwards incompatible
unless you add the same options twice. Why not do it for G2 then?
But --enable for autodetection is not intuitive either. Besides having a
switch for autodetection is missing the point anyway. If you are supplying
a build option you do not want autodetection.
Anyway, I am going to add a few lines 'configure --help' that explain how
configure works. Maybe that helps clear up the constant confusion.
> Arpi think's that the mplayer way, and Gabucino is happy if he can
> confuse/abuse as many users as possible. The rest of developers
> don't want to take side or doesn't care.
> Or think that this is "a bug, but very small".
See above. I think your approach was backwards incompatible, that is the
I think that there is a consensus among the developers that having a
possibility to force options avoiding autodetection is a good idea.
More information about the MPlayer-dev-eng