[FFmpeg-devel] [ANNOUNCE] upcoming vote: CC election

Marth64 marth64 at proxyid.net
Thu Dec 19 06:27:40 EET 2024


Hi Michael,

> 2. and the most recent case
Continuing where I left off.

Let us analyze the s337m situation.
* Contributor shares significant work, as an RFC, in good faith
* A back and forth exchange happens with technical feedback and history
* Then, two senior contributors exchange inflammatory remarks,
presumably fueled by damaged relations
* Conversation was reported to CC but outcome could not be reconciled
with consistent enforcement

The s337 thread actually is irrelevant to the broader issue, but let
us pick it apart for a minute.
This particular thread should have been handled in two angles,

(1) The technical debate between the senior engineers should have been
weighed in on by TC
Yes, TC wasn't directly "summoned", but I imagine we can employ some
proactivity if we see a tense
technical discussion. The point of the TC, in my eyes, is to settle
deadlocked technical debates.
That is, as a collective response from the TC. Even if that is, "We
already talked about this once, case closed."

(2) The inflammatory remarks weren't moderated. This is the broader issue.
These types of interactions are different from some random person
coming on say, IRC, and offending people.
These are senior contributors in a deeply fueled discussion. No ban
will fix this.

In fact, saying the word ban, voting for a ban, or applying a ban
doesn't seem to do anything.
Based on the example you provided, even after 3 attempts some people don't care.
I'm not saying we jump to ban people. This is not a PHP video games forum.
But, waving the word "ban" should be used judiciously and may need to
be redefined, so that it is taken seriously.
Likewise, exercising and applying the ban should be taken seriously
too, especially when considering senior contributors.
Rubber voting/stamping bans and ignoring similar offenses is not
effective. I would hope a CC of 5 can work through this.

So we need to be more clear with a nudge of improvement and push: get
this toxicity off the ML.
Take it offline or refine your communication approach.

Folks should be encouraged to think,
(1) Is what you're saying adding value and even relevant to the conversation?
(2) If we're stuck in a bitter technical debate, can the TC help? (yes probably)
(3) Did I buffer my thoughts or react on impulse?
(4) Could I have ignored the "igniting" remark and focused on the
technical part of the discussion?

We have a Code of Conduct, does it need to be improved?
Do we need a pre-canned message to respond with when CoC is starting
to break in a conversation?

If people can't articulate with some basic guidelines of cordiality,
they shouldn't be emailing.
If a CC group of 5 can't see this without bias, then something is
wrong with the elected CC
or there should be an explanation. *Everyone* is biased subliminally,
whether we like it or not.
But we should put our best effort here to do better.

That said, I believe in forgiveness and second chances.
If the folks involved have had some legacy or current significant
contribution to the project,
IMO the CC should work with them collectively and, if it makes sense
in the situation, privately.
I am not saying CC should be a therapy group, but it hurts
productivity if talent leaves.


More information about the ffmpeg-devel mailing list