Re: [FFmpeg-devel] STF 2025 grant request (2025-09-01 target)

[Michael Niedermayer via ffmpeg-devel <ffmpeg-devel@ffmpeg.org>]

[-- Attachment #1.1: Type: text/plain, Size: 2577 bytes --]

Hi Pierre

On Wed, Aug 20, 2025 at 09:28:26AM -0700, Pierre-Anthony Lemieux via ffmpeg-devel wrote:
> On Wed, Aug 20, 2025 at 9:24 AM Niklas Haas via ffmpeg-devel
> <ffmpeg-devel@ffmpeg.org> wrote:
> >
> > On Tue, 12 Aug 2025 08:32:39 -0700 Pierre-Anthony Lemieux <pal@sandflow.com> wrote:
> > > Quick reminder that the deadline for submitting your proposal is
> > > quickly approaching.
> > >
> > > Also note that the total amount of the proposed projects is below for
> > > the minimum threshold for STF to consider the application.
> >
> > Despite my initial reservations, I decided to also apply with another
> > continuation of the libswscale project. Just letting you know as a heads up
> > that I will add my proposal to the wiki shortly.
> >
> > How much more is currently needed to meet the minimum threshold?
> 
> @Michael Niedermayer How many of the 100 modules do you realistically
> expect to complete?

I think the main uncertainilities are
1. If teh community preferrs merges or cherry picks,
    If they want cherry picks its 100 patch(sets) passing review, adding tests
    where samples are on our server.
    If we do it as a merge, the work is still there but its different likely
    less work.
    Code would be merged in one go, then people could still review the
    individual modules afterwards, and I still would have the same effort to
    do to add tests and handle reviews. But maybe fewer people will post
    reviews with a merge than with cherry picks
2.
    when the work starts, because i intend to submit some patchset to test
    the whole process as soon as i am not busy with other work in ffmpeg
    So we might end up with fewer than 100 if we do some before STF starts
3.
    In all this I assume that the review process is fairly smooth and requests
    for major changes would be very rare because that would not work with the
    number of modules.
    And i assume that either we have testsamples or we dont and so the work
    adding tests is manageable. But the devil is in the details and tests
    can lead to detected bugs if the results are different between archs
    and so on. And then I would have to debug and fix that unknown number of
    bugs

So really if everything is smooth i expect all 100 but if issues arise
it can be less.
I will know how badly i miscalcuated this only afterwards ;)

thx

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

No great genius has ever existed without some touch of madness. -- Aristotle

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".