[FFmpeg-devel] [RFC] libpostproc splitout

[Michael Niedermayer]

Hi jb

On Tue, Nov 26, 2024 at 02:41:02PM +0100, Jean-Baptiste Kempf wrote:
> Hi
> 
> On Tue, 26 Nov 2024, at 14:11, Michael Niedermayer wrote:
> > But thats not what iam talking about
> >
> > with "eternal debates" i mean things like discussions about
> >
> > * the value of tests (yes they are valuable), teh discussion here yet
> >   ended with some people refusing to add tests, and in another case
> >   removing sidedata tests on framecrc
> >
> > * debates about at which point code can be in the master repository
> >   vs some external place that no user uses
> >
> > * debates on guilt about booths
> >
> > * debates on intelectual property rules
> >
> > * debates on infrastructure
> >
> > * debates on deleted twitter posts
> >
> > * debates on what is in scope for ffmpeg and what should be rejected because it
> > doesnt belong here
> >
> > * debates on slander
> >
> > * debates on lies, infighting
> >
> > * debates on VCSs
> >
> > * debates on moving control over servers
> >
> > * debates on infrastructure
> >
> > * debates on who said what 9 years ago
> >
> > * debates on hosting of servers that have been hosted at the same place 
> > for decades
> >
> > * debates on recording and remote participation on meetings
> >
> > * debates on someone leaving a booth to eat
> >
> > * debates on jokes as if they where no joke while konwing they where a joke
> >
> > * debates on removing code other community members intend to work on
> >
> > * debates on peoples contributions out of spite to take revenge and 
> > block their work
> >
> > * debates plain for the sake of debating
> >
> > * debates if trolling should be forbidden on the ML
> >
> > And i think we can add a few more pages, if we wanted but thats a sample of
> > what i meant and what i do not enjoy
> 

> This is what managing an open source project is about, or if you prefer, open source governance.

yes, some of the above is


> We understand you don't like open source project management.

no, thats just wrong. Its also a bit rude to put words in my mouth.

Iam happy to manage this. The problem is i dont have the authority to do it.
I mean technically i have. But so far i delegated this to teh CC.

Also as should be obvious. moving libpostproc to a seperate repository
is not because i dont want to manage it. Because obviously i would be
managing it. It is for the very reason because i DO want to manage it.
And do not want to/cant work on it UNLESS i manage it :)
It can be a pain to work on code thats "managed" by others who have no
interrest in the code.
(again making maintainers have final authority is a prereqesit for
 contributions from some key people)

So yeah, its exactly the opposit of what you say. I want, like and need
to manage the _PART_OF_ the project iam working on. Or i cannot and will not
work on that.

I work on the security fixes because i have the authority there. Because
noone can argue against a security fix, i mean mostly noone ;)

I avoid larger contributions recently because of the lack of final
authority and the uncertain outcome of TC decissions.
I ve tried it a few times over the years, but the process is sometimes
painfull.
Everyone who feels like it, demands and comments on patches.
You can be lucky and a patch is reviewed by someone who is genuinely
interrested in the code or interrested to use it. (and yeah thats what
happens with most small patches, if they get a review)
Or you are unlucky and some random person who has some disagreement
decides to rally all his friends or alone tries to simply make the
patch not go in.
Even with paid work it has occured 2 or 3 times to me. So i nowadays
always tell customers theres a small chance the community will randomly
block some work in a way that makes it impossible or uneconomic to implement.

again linux style development model with clear maintainers for each area
or the simply "maintainers have the final say in maintained areas"
fixes this.
I guess i would present this issue in a talk if i was going to meetings,
maybe people would understand it better then. But iam just not the guy
doing talks or meetings

thx

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Freedom in capitalist society always remains about the same as it was in
ancient Greek republics: Freedom for slave owners. -- Vladimir Lenin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20241126/4e3b2da1/attachment.sig>