[FFmpeg-devel] [RFC] STF 2025

[Rémi Denis-Courmont]

Le 22 mai 2024 00:34:03 GMT+03:00, Thilo Borgmann via ffmpeg-devel <ffmpeg-devel at ffmpeg.org> a écrit :
>>> I hope you realize what you argue in favor of.
>> 
>> Yes. It's quoted above.
>> 
>> Are you claiming that *no* review is better than *some* review done in
>> *public* for all to see by a paid professional just because the person is
>> maybe biased?
>> 
>> First, even volunteers have their own biases. Any expert should have opinions
>> from their experience, and that by definition makes them "biased".
>> 
>> And second, you can't have it both ways. Either we want people to be paid for
>> review, and they will be answerable to their sponsor, or we want people to
>> continue to work on their free time.
>
>I think that is what you don't understand.

You're not answering the question here. The current STF funding of 153k€ for 2 years is roughly enough to pay for ONE full-time entry-level software engineer in Germany. Even if this were doubled with another similar round of funding next year, and even if that was to be reliably renewed year on year, and assuming that STF keeps an hands-off approach of not influencing the work, that will *not* be enough to pay all reviewers.

So is it better to have no reviews or reviews by skilled corporate employees?

(...)
>> Ideally so but that's the land of utopia.
>
>Of course, we talk about what should be, don't we?

Of course *not*. There is no point debating ideals that we can all agree on and that will never come to fruition. Rather this is all about how to concretely apply or not apply to STF, and more generally how to try to improve the sustainability of FFmpeg in a realistic manner.

>> And "I hope you realise that you are arguing for" Intel, Loongson, etc.
>> employees to stop reviewing patches.
>
>Syntax error. What exactly do you mean?

I fail to see a syntax error. You're saying that corporate employees should not review because "they [will] want to get [their]" or their colleagues' "stuff in" (your words).

Intel and Loongson are obvious current examples of companies whose employees are pushing and reviewing enablement patches for their commercial hardware. That is very definitely not "unbiased" nor "independent".

>According to my assumptions: No, I value reviews of company employees in general which have been proven to be useful and unbiased e.g. in getting part of the community reviewing 'stuf' but not their 'own stuff'.

I never said that I wanted biased reviews. I said some reviews were better than none, in spite of the risk of bias.

So much for your grandstanding against my alleged not realising what I am advocating for, if you end up agreeing with me...