[FFmpeg-devel] Project orientation

[Soft Works]

> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces at ffmpeg.org> On Behalf Of
> James Almer
> Sent: Saturday, July 4, 2020 5:37 PM
> To: ffmpeg-devel at ffmpeg.org
> Subject: Re: [FFmpeg-devel] Project orientation
> 
> Another thing worth mentioning is a lack of new blood. Despite participating
> in GSoC for a long while, i can't name a student that stuck around after the
> fact. Mind, there are new devs that started contributing for other reasons,
> but perhaps not enough?

My time as a student is long gone. Over the years, I have seen a lot and done a lot 
of things as a developer. From my perspective, ffmpeg is one of the least 
attractive software projects for contributing, that I know. There are so many
reasons, I don't even know where to start.

Let's take this one: While writing this e-mail, I have to insert manual line breaks
in order to adapt to how mails were written until the 1980ies. I cannot use 
bullet points or font formatting (e.g. to indicate code snippets).

Communication via a mailing list is anachronistic.  The world has changed 
and much better ways exist to communicate. The fact alone, that this mailing-
list approach forces me to continuously read everything - including that
sociopathic talk that is happening here as well, would be reason enough
for me to walk away. 
With something like a forum or GitHub issues, I could not only step out of 
those conversations, but also better choose topics where I'm interested in
and where not.

And using patch files with plus and minus signs for collaborating on code 
changes? Really? It's not just that it would be a matter of taste - there are 
obvious shortcomings when comparing with modern ways, like that it's 
impossible to connect code and discussion about that code.

Another problem is, that there is almost no connection to users of ffmpeg.
The user mailing list is strictly separated and developers seem to be living 
inside some kind of bubble with little contact to those who are actually 
using ffmpeg. That's my impression at least. I'm sure there is one or the 
other who does differently.

As a developer (without a well-known name) who wants to contribute a patch,
things can be quite frustrating here. When that patch accidentally hits an
area of one the very few who are caring and friendly - you're lucky.
But otherwise, a patch will either be ignored or talked into infinity.

I have a number of things to contribute, but after it didn’t work well 
with small things, I decided not to bother with the bigger ones.
As an example, I gave my QuickSync/DX11 code to Intel, in order to let
them deal with getting it merged, but even that didn't work out so far.
AMD AMF hardware decoders are still missing. They had submitted 
something 1 or 2 years ago, but they didn't get any feedback and were
discouraged to proceed. 
Windows Media Foundation HW acceleration has recently been merged,
but the actual patch existed for several(!) years. 
Things running not really smooth here, a lot more progress could be made
in some areas.

I know, the project is community driven, but sometimes I think there is
missing someone who is "in charge" of things and connects the many
loose ends that exist.

-----------------------------

I don't really have a problem with how things are. I accommodate, we got  
our own fork and platform builds and just do our changes there.

This is my very personal view. People have different opinions (luckily), 
and I didn't write this to discuss all those points in detail or to initiate
some change to the project.

But when somebody starts wondering, why there aren't any new developers
joining the project, one should maybe think about whether one or two 
of the things I listed above could be the reason...

Kind regards,
softworkz