[FFmpeg-devel] Sovereign Tech Fund

[Rémi Denis-Courmont]

Hi,

Le 4 février 2024 21:28:44 GMT+02:00, Michael Niedermayer <michael at niedermayer.cc> a écrit :
>On Sun, Feb 04, 2024 at 03:38:43PM +0100, Rémi Denis-Courmont wrote:
>> Hi,
>> 
>> Le 4 février 2024 14:41:15 GMT+01:00, Michael Niedermayer <michael at niedermayer.cc> a écrit :
>> >Hi
>> >
>> >As said on IRC, i thought people knew it, but ‘the same person as before’ is Thilo.
>> >
>> >Ive updated the price design suggestion for the merge task, its 16€ / commit limited to 50k€
>> >this comes from looking at pauls fork which has around 500 commits in 2 months thus
>> >250 commits per month, 12 months, and if we allocate 50k that end with roughly 16€ / commit
>> >if activity stays equal.
>> 
>> It's very different if we're talking about librempeg or some other unspecified fork. I could make a fork that removes MMX et al, and claim that I'm merging a fork.
>
>There are so many reasons why this wouldnt work
>(first you would have to lie, i dont think you would,
> then it would not be left that way on the wiki,
> not being sent to STF that way
> and not being accepted by STF and more)
>
>But assuming one could get away with that in the short term
>Why would anyone do something like this to destroy our all opertunity
>to obtains grants in the future ?

I don't know. That was purely an example, and I prefer to be the fictional bad guy in my examples, so nobody else feels insulted. But you can't blame people for being distrustful when a proposal is brought forward on short deadlines even though it was privately known for months.


>> Indeed I don't think that a semiformal open-source community with a lot of strong and varied opinions will carry such dotting of all i's very effectively. That has been one of the arguments for delegating this to a contracting IT company rather than to FFmpeg-devel and SPI.
>
>If the FFmpeg team can make decissions about what to fund then we do not need
>any contracting IT company.

Let's face it: FFmpeg is not the healthiest of open-source communities as of now.  But that's not even relevant here: OSS communities are typically focused on development and maybe support and promotion, *not* HR and payroll, nor waterfall-style project management.

Ergo, there would be no shame in conceding that FFmpeg would suck at those tasks, and a company whose job those things essentially are would be more effective at it. And I'm not saying this out of self-interest, just  pragmatism (call it cynicism if you will).

>OTOH If the FFmpeg team is not able to make decissions, thats a far bigger
>problem and it needs to be understood and corrected

I don't think the technical development lists of an OSS community should concern themselves with funding matters. Well-funded foundations surely need to concern themselves with this, but they don't mix it with development. And FFmpeg is not sl well-funded in the first place.

> Because whoever controlls the income of developers
>effectively controlls the project.

As long as there are several parties involved, and a single trust doesn't dominate the GA and the TC, I don't see that as a major problem. Or rather, it's a lesser problem than loosing competent developers because they need to work on something else to earn their living.

>An emloyee has to do what she is being told be her employer. So if the main developers
>become employees payed to work on FFmpeg that would hand FFmpeg to some CEO on a
>silver plate,

200000€ are not remotely enough for that to happen. You'd need 2-3 orders of magnitude larger investment, without competition, to get there at minimum.

So I don't see a risk here. But it's up to Thilo really, if he insists on going through SPI or not applying for STF at all.

>This would change FFmpeg from a Free software project to a commercial company.
>I do NOT agree to this, and i belive many others also do not agree.

I think a lot of people would rather get paid to work on Ffmpeg, and would in fact contribute more effectively if they were. And conversely, quite a few contributors seem to be acting for their commercial employer already.

Also, as a consultant or maybe an associate for FFlabs, it's a rather contradictory position for you to hold.