[FFmpeg-devel] [PATCH 1/2] opus_pvq: add resynth support and band encoding cost function

wm4 nfxjfg at googlemail.com
Sat Apr 15 19:04:33 EEST 2017


On Sat, 15 Apr 2017 16:33:41 +0200
Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:

> 2017-04-15 15:06 GMT+02:00 wm4 <nfxjfg at googlemail.com>:
> > On Sat, 15 Apr 2017 07:55:50 +0200
> > Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
> >  
> >> 2017-04-15 2:35 GMT+02:00 wm4 <nfxjfg at googlemail.com>:
> >>  
> >> > Legally, all these copyright headers are meaningless, as long as the
> >> > license is correct.  
> >>
> >> You are joking, right?  
> >
> > Well, are you joking? While it may seem so, you apparently aren't.  
> 
> I am not joking:

Oh that's unfortunate.

> The first paragraph of all open source licenses I know of say something
> similar to "you may distribute this source code as long as you
> don't change above copyright notice". If you say now that this
> copyright notice is meaningless, you are either joking or - well,
> you can imagine.

So changing the project name (like FFmpeg does it all the time) does
not constitute changing the copyright notice? Or does FFmpeg violate
the law? Or are you joking?

Some files don't even have a name in the copyright notice. And I bet no
file lists all copyright headers in the notice. Thus it's meaningless.

What really counts is who actually has the copyright on a piece of
code.

> >> > It's possible that Libav is not always careful with attribution, but
> >> > it's the same with FFmpeg. For example, af_pan.c is LGPL, even though
> >> > it was ported from MPlayer GPL code, and the author could not be
> >> > contacted. Recently, I contacted the same author about relicensing the
> >> > same MPlayer code (of which af_pan.c was a subset of) to LGPL, and he
> >> > explicitly denied relicensing. So if I had been cehoyos, I probably
> >> > wouldn't shut up about how FFmpeg violates copyrights (fortunately I'm
> >> > not cehoyos). I think nowadays af_pan.c certainly does not violate the
> >> > original author's copyright though, because absolutely all code was
> >> > removed/replaced.  
> >>
> >> Do I understand correctly that you are neither defaming Clement nor
> >> doing something that is illegal?  
> >
> > Is that some sort of trick question? I'm doing neither.  
> 
> Clement says that he wrote the code he committed by himself and
> that it is not based on MPlayer code. You say he violated somebody's
> copyright: Who is defaming anybody here?

http://ffmpeg.org/pipermail/ffmpeg-devel/2011-November/116431.html

Panning was later reimplemented in libswresample (independently?
unknown), and some time after that, this code was removed and replaced
by using libswresample API.

> > But what _you_ do is defaming the entire Libav project  
> 
> No.
> I am (to quote you) stating "facts" namely that 1) FFmpeg's
> gitlog contains many commits that indicate avconv developers
> have not correctly given attribution to FFmpeg developers
> when moving or copying code and 2) that other commits in
> addition to those linked by Ivan (probably myself in the past,
> I didn't remember) exist.
> 
> > (and possibly me),  
> 
> I don't remember doing this.

There are a lot of good doctors who specialize in memory problems.

> > which is illegal  
> 
> Please stop calling my actions illegal

Can you tell me why I should do this? Or are you joking?


More information about the ffmpeg-devel mailing list