[FFmpeg-devel] [VOTE] FFmpeg leader

[Felipe Contreras]

On Sat, Oct 2, 2010 at 3:45 PM, Michael Niedermayer <michaelni at gmx.at> wrote:
> On Sat, Oct 02, 2010 at 02:35:05PM +0200, Luca Barbato wrote:
>> On 10/02/2010 02:25 PM, Michael Niedermayer wrote:
>>> This is untrue
>>> the assert changes have been discussed a long time ago with noone caring.
>>> the per codec parameters too have been discussed and ive explained that
>>> i plan to implement them through a priv_avclass in AVCodec and AVOption
>>> access to the private context, noone disagreed from what i remember
>>> and the swscale api changes have briefly been discussed in the IRC thread
>>> that was triggered due to the 10bit dithering trolling. Yes the final
>>> implementation was a hair differnt than what was discussed because its
>>> simpler as it is now but it all was discussed.
>>
>> The problem here is that we lost those discussions in the mailing list
>> flood. I think probably _most_ of this annoyance would be mitigate if
>> those changes would happen on a split branch and then merged on a pull
>> request on the mailing list. Same goes for the issue you had with Mans
>> commits colliding with your uncommitted work. We have better tools,
>> using them _should_ reduce this kind of friction.
>
> nope, i was trying to fix the very same things mans fixed, namely reformating
> the code for better readability. no tool in the world would have avoided that
> colission and wasted time.

Then write your own:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=scripts/checkpatch.pl

Sure, it's a lot of work at first, but it avoids so much more work
pointing out the same code-style mistakes over and over again.

> I do take it seriously when i commit something and i do fix issues that are
> pointed out to me (unless its bikeshed that goes far too far like the #include)
> its annoying if someone complains to me and while i start working then commits
> under my fingers that very change without asking, making me waste time

If you had sent the patches for review first, you wouldn't have this problem.

-- 
Felipe Contreras