[FFmpeg-devel] Preliminary announcement about the current situation

Michael Niedermayer michaelni
Thu Jan 20 16:03:52 CET 2011

On Wed, Jan 19, 2011 at 11:26:38PM -0500, Ronald S. Bultje wrote:
> Michael,
> first off, allow me to say there probably is not one single reason
> that we all subscribe to. Rather, we are collective in our dissent.
> On Wed, Jan 19, 2011 at 10:53 PM, Michael Niedermayer <michaelni at gmx.at> wrote:
> > On Wed, Jan 19, 2011 at 01:12:54PM +0100, Michael Niedermayer wrote:
> >> On Wed, Jan 19, 2011 at 12:54:48PM +0100, Herv? W. wrote:
> >> > On 19/01/2011, Michael Niedermayer <michaelni at gmx.at> wrote:
> > [...]
> >> > > Of course its their full right to fork if they see the need for that but
> >> > > thats
> >> > > not exactly what they did.
> >> > >
> >> > > And that I and everyone i spoke with dont even know the reason behind this
> >> > > move
> >> >
> >> > Have you asked any of the undersigned of "[FFmpeg-devel] [ANNOUNCE]
> >> > New FFmpeg maintainership" ?
> >>
> >> maybe not but they surely have seen me asking here in public and can awnser
> >> here
> >
> > Apparently not a single developer who signed this feels the need to say why.
> > They are all subscribed and all have read this
> A) you are stubborn. When you have an opinion, it is
> hard-to-impossible to convince you otherwise. I'll openly admit that
> that's not scientific evidence (one way or another) whether we gave
> any proper reason to convince you otherwise.

When i consider A better than B i tended to insist on A. Especially when its
about public API.
Do people really want the 2nd best option to be choosen when its not much
work to choose the best?
Yes when its alot of work i understand and i definitly tried to avoid recently
to push toward people spending alot of time for small improvments. In the
distant past ive sometimes asked people for timeconsuming rewrites for just
negligible things but i dont think ive done that in the last few years.
(exceptions in the thousands of reviews might exist of course)

> B) you fight a lot with people. These fights are persistent, get
> personal and it is hard to get you to reconciliate. That's not what I
> expect of a leader. You, of all, should be the first to offer a
> gesture, no matter how symbolic, to fix things up.

I think this is a problem of using non realtime communication like email
that could have been solved by myself being on IRC

> C) overabundance of votes,

> micro-optimizing reviews and

years ago yes, but recently ive definitly tried to avoid pushing such things
too much.

> discussions/fights lead to absence of actual development (not just by
> you, also by others). This leads to x264 being done outside

> (which is
> fine - but now if I want to contribute there, I get attacked),

true, yes, i admit i sided with baptiste in this discussion. I dont know
if its bad or not for the ffmpeg leader to be a bit offensive toward ffmpeg
developers who implement new codecs outside ffmpeg. That can be argued both
ways i guess
I would have never been offensive if you just contributed to x264 but adding
a VP8 encoder IIRC was a bit annoying as i felt ffmpeg was the more correct
place for it.

> our aac
> encoder being a little bit of a joke (sorry Alex, but you openly admit
> this yourself). Seriously, how much FFmpeg development (not review)
> did you do lately?

not as much as i wanted, that was primarely because it was not fun anymore
as too many people jumped on every commit (send a patch first you ...)
and when its a patch everyone has a comment about something that is largely
irrelevant but costs time to change and more time to fight to not change.

> D) mplayer != ffmpeg. I see that this is hard for you to accept, but
> they are not the same. This includes swscale, mpcodecs, dll loader and
> all the other endless flamewars.

having a private repo resolves this and some other issues mentioned to some

> E) you should be on IRC, it's been asked repeatedly.

maybe 1 or 2 weeks ago i upgraded xchat, looked up freenodes webpage on how
to register but then it was too late and i went in bed, next day i ended up
with a pretty bad headache and after that i was busy for a while as i had done
nothing while having a headache. Before i resumed my plan the new maintainers
where announced.

> Having said all this, I don't want you to stop developing FFmpeg, but
> that's exactly what I want you to do again: write code. Not mpcodecs
> wrappers, but actual FFmpeg code. I think there's a lot of reasons why
> you're not the ideal BDFL, but you are a great programmer and I'd wish
> you did more of that.

No doubt iam not the ideal BDFL, but noone in ffmpeg is the ideal BDFL. we
dont have linus here, not even fabrice is left.

First id like to also thank you for this awnser and nicely and politely written

But id like to know why, if resolution of the problems listed here was the goal,
people did not present a list like this to me with signatures and an ultimatum
to correct?
You guys complain to me iam not good, and true but this is actually not better
if not worse than what ive done. Thats not a good start for a new team IMHO

We have talked on the phone long ago and i now remember that myself being less
than polite and not adding whitespaces after keywords where discussed.

Ive seriously tried (but no doubt failed on occasions) to be more polite in
And after the last bikeshed from stefano about the whitespace ive written a
script that automatically fixes that in a patchset on git, ive not used that
script yet but it would have made fixing it alot easier for me to deal with
such bikeshed and would have reduced the stress from such requests for me.

Then there where accusations that i violate policy, and it turned out people
just thought things where written in there that actually where written in
policies of other projects but not ffmpeg and i had actually not violated
it beyond the unavoidable little mistakes that happen over the years.

When talking with attila long ago he wanted me to write a list of roots duties
and rights. I did and it failed miserable but i tried

Why did people overtake the server and domain and afterwards present this list
instead of presenting it first giving me an ultimatum to deal with it and then
take action if i fail?
I would understand if you have a better leader but what i remember instead was
you yourself agreeing with me that a team of leaders would be a bad idea.

Its not as if i had not tried to resolve issues in the past that have been
presented to me. And yes i surely failed to resolve some and i might have been
too slow on others. But i tried.

I admit i did not call mans, i admit i gave up convincing mans to return after
2 or 3 private emails got no reply.

> > I guess i did not spend enough years of my life working on ffmpeg to deserve
> > an awnser
> <troll>
> I wish this impatience had been more apparent when some developers
> asked you to reconciliate with Mans, Diego or Attilla...
> </troll>

The things that happened where quite unfortunate chain reactions and some
misunderstandings like a mail from me to attila and the results.

Truth is 4? years ago i wrote a will to make sure the computer stuff i have
would end up used somewhere in FOSS, both attila and diego are listed in it
to get all my computers and distribute them as they see fit.
Ive not changed that nor would i if i would write it now again.
Can i be hating them so much if i wrote that?

Theres quite alot that has happened in the last few days and ive not made up my
mind entirely what i will do, maybe ill continue to maintain ffmpeg at videolan
for a while, as long as users or developers want it or there are some who
offer little jobs on ffmpeg but in the end i will likely leave
Contributing to the new maintainers is no fun and too much bikesheding for me
besides i still feel the whole move was plain and simple wrong like it was done
and i do not want to support that.

Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The misfortune of the wise is better than the prosperity of the fool.
-- Epicurus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20110120/e623f8a0/attachment.pgp>

More information about the ffmpeg-devel mailing list