On Mon, Aug 18, 2014 at 12:15:10AM +0200, Clément Bœsch wrote: > On Sun, Aug 17, 2014 at 09:14:47PM +0200, Luca Barbato wrote: > [...] > > > Ive asked [1][2] back then what "policy in place" was broken > > > > - you tried to commit code that was blatantly below the already lax > > quality requirements (e.g. it contained tabs, it was (and still is) hard > > to read, it contains dubious, aka security-concerning, practices), I > > told you not to commit those as-is and you blatantly ignored me and the > > others against it. I'm referring to the mplayer filters. > > > > Weren't mplayer filters pushed post fork? I dont remember, they dont seem to be in the libav git history also there are no tabs in the mplayer filters in ffmpeg, and IIRC no security bugs in the filters have been found in them since they where pushed. Also its hard for there to be security bugs as they are only used when explicitly selected by the user > > > - your interaction with whoever wasn't in full agreement with you was > > horrible, I told you for months in public and private, you seemed to > > agree just to behave in even worse way later. The interactions with Mans > > are probably the best example of this. > > Interactions with Måns and everyone else were also pretty terrible, to the > point that he also left Libav in a rage quit (see #libav-devel on > 2011/09/17 for related, when he left. Michael was not here). Note that I'm > not saying this is the reason he completely left the project. > > > - your interaction with whoever provided infrastructure service was > > horrible[1]. As Attila already stated here[2] and here[3] > > > > AFAICT this has nothing to do with the policy rules in place. There is no question i made mistakes 4 years ago, and no question i could and should have interacted with developers in a less abrasive way back then. Iam a technical person not a diplomat nor politican. [...] > > - you putting in FFmpeg pretty much every patch from every branch you > > can come by, including my incomplete work from github (you did for pulse > > and segment with interesting results, hopefully you won't do that again > > anymore). > > Yes, with full authorship, because those were requested & needed in > FFmpeg. Also, communication does not work with you (we FFmpeg are > definitely ignored by most of the people on Libav); this means that > obviously we are not going to annoy you for information or status about > your progress. May I remind you that Michael is banned from the > mailing-list, IRC and some of the developers repositories? Why would you > expect him to even try to communicate with you about these contributions? also about pulse, we did take lucas "incomplete" code but the current code we use is quite different from that and has evolved alot. The original used the simple API from pulse "pulse/simple.h", FFmpegs current code does not. Also i think our current code is more based on work by the pulse audio developers and Lukasz Marek than lucas original. And theres pulse input and output support in FFmpeg, the original code from luca and libav only support input through the simple API [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Its not that you shouldnt use gotos but rather that you should write readable code and code with gotos often but not always is less readable
Attachment:
signature.asc
Description: Digital signature