[FFmpeg-devel] Democratization

[James Almer]

On 1/14/2025 4:51 PM, Nicolas George wrote:
> James Almer (12025-01-14):
>> The GA is a list of currently active contributors
> 
> Please, let us avoid that simplistic mistake.
> 
> The GA is an attempt at an approximation of a list of active
> contributors.

That's no different than what i stated. If someone sends 20 patches, 
even if trivial or "low quality", it's still an active contributor. And 
if you want to change it to qualified vote, it could be done, but then 
the list is going to be even smaller.

> 
> Why 20 commits, not 19 or 21? Because it's round: arbitrary.
> 
> Why 36 months, not 35 or 37? Same, because it's round: arbitrary.
> 
> Why count cosmetic commits as much as complex ones? Because it's easier:
> arbitrary.
> 
> Why count commits and not help to users on the ML? Because it'd be hard:
> arbitrary.
> 
> There is no clean-cut limit between an active contributor and an
> inactive one, there are just people who are more active than others and
> an arbitrary cutoff at a convenient point.
> 
> So let us not pretend we believe the rules of the GA are sacred, or even
> its principles.

They aren't, of course. They can be changed if they are found to be 
faulty or insufficient.
The current list is done with a dumb script that doesn't care about 
anything other than commit count in a given time range. But like i said 
above, it can be changed.

And please, don't forget the GA also includes people that don't send 
patches anymore but are nonetheless still active members of the 
community (reviews, infrastructure maintenance, etc).

> 
> These rules were chosen under the assumption that contributors would
> vote for the good of the project itself, and that contributors who do
> not care much would not vote much. These assumptions might have been
> when most contributors were hobbyists, but people who seek a profit from
> their contribution will vote for their own interest even if they barely
> qualified.

I have not seen any vote for profit or own interests reasons so far. And 
there has been a very small amount of votes ever since this was 
implemented, all things said. Six years or so since we adopted Condorcet 
and we have done probably less votes than that (Outside of renewing 
committees).

> 
> So let us acknowledge that the GA was a mistake, re-affirm that this is

The GA was not a mistake. The mistake, if anything, is accepting it as a 
management method but then reveal that you never truly intended to honor 
any outcome whatsoever.
And if you find it bad, what is your suggested alternative? Considering 
the idea for the past decade has been that there isn't supposed to be a 
single person who manages everything and has the absolute last word. A 
pretense we could of course drop too.

> a community project and that profit, if it happens, is only a side
> effect and not the main goal. And once it has been made clear, once the
> people who would pervert the project for their own gain have been
> thoroughly de-fanged, we can consider which governance mode we want for
> the long run.
These are some serious accusations. You need to provide names and 
specific actions you consider biased towards personal profit. Otherwise 
this is unfounded libel.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20250114/39f9745f/attachment.sig>