[FFmpeg-devel] Develop FFmpeg through your browser?

Rémi Denis-Courmont remi at remlab.net
Sat Nov 9 12:16:11 EET 2024


Le sunnuntaina 3. marraskuuta 2024, 1.56.26 EET Michael Niedermayer a écrit :
> Should we move to a different development system ?

Simply no. The issue is the review and merge process, not the software 
development as such.

E-mail was not designed to do code reviews. It does a pretty poor job at in-
depth code review, and even worse job at tracking patchsets. Maybe some day 
Git itself will be extended to provide some form of code review support, but I 
doubt this. Or maybe some other VCS with such support will outshine Git, but I 
doubt that even more.

In the mean time, the only options for the review process are email and web 
interfaces. Web interfaces are nowhere near ideal, but having done 
considerable review work with both alternatives, it is clear to me that they 
are a hell of a lot better than email.

> I dont know, but I belive we should carefully consider if we want to move to
> a gitlab like system that is a commercal / corporate and not community
> driven system that we can get stuck in.

My gut feeling is that, from a purely pragmatic standpoint, communities are 
better off with an open-source yet commercial offering supported by a viable 
business, than with an also open-source project  precariously maintained by a 
small community of hobbyists.

With that said, I think most people (including myself) are not familiar with 
the community-based options such as Gitea or its fork Forgejo. If they address 
merge request and code reviews as Gitlab, I think most of the Gitlab 
proponents here would be content with moving to that instead.

> Companies get sold change owner/leader, and in no time gitlab can be owned
> by microsoft or another corporation

Nobody seriously suggested switching to proprietary forges such as Github or 
Gitee (or at least not this time around).

> and be merged with github or similar for example.

> Also i have a dislike for these browser based systems.

That's beside the question really. I don't exactly love Gitlab either, or web 
apps in general. I just dislike the bugged email-based workflow even worse.

And there are currently no alternative: I wouldn't mind a native desktop app, 
but I don't think that such thing exists, nor that it would be viable without 
support at the Git level (i.e. storing the review data in Git rather than in a 
relational database).

> Should we move to videolan? This is a seperate question and has nothing to
> do with changing the "development system" ?

> We can install gitlab on our infrastructure, if the community decides that
> it wants gitlab. We can also install anything else the community wants. If
> people do move to videolan, i will not come along. I see not the slightest
> reason to give up our independance.

TBH moving to VideoLAN is a purely practical question. That means Thresh and 
other VideoLAN admins would do the maintenance, and the VideoLAN foundation 
would pay for the hosting. In other words, nobody here has to be burdened with 
additional work and cost.

The community will always have the option to move to another host thanks to 
Git's decentralised design. The only thing to worry about would be losing the 
requests and issues history. But even that could probably be copied and 
archived elsewhere regularly if you are afraid that VideoLAN would turn evil.

All other things being equal, I aree with you that it would seem more sensible 
for FFmpeg to host its own web forge - but it is questionable that all 
relevant things are equal as of yet in terms of how VideoLAN and FFmpeg 
infrastructures are managed.

-- 
Rémi Denis-Courmont
http://www.remlab.net/





More information about the ffmpeg-devel mailing list