[tor-project] Localized mailing lists

[Roger Dingledine]

On Thu, Apr 05, 2018 at 07:30:53PM -0300, ilv at torproject.org wrote:
> Someone might "hey, but if we have a mailing list for language X, we should
> have one for language Y". To avoid this (supposing that we want) we could set
> some requirements like for example there should be at least one core member
> willing to run the list. 

I think yes, we should try the experiment of making this list and seeing
how it goes.

I like your notion of "there should be at least one core member willing
to run the list." I'm tempted to suggest "at least two" to prove that
we have some sort of critical mass before making a list, but I think we
can figure out the more exact policies when we have tried this list and
we're considering a second one.

But to be clear, it needs to be more than just "one core member
interesting in *starting* the list" -- they need to commit to *running*
the list, meaning keeping its discussions on track, moderated against
trolls, etc.

A second principle that I would propose is that we have a periodic
check-in point, like every six months or something, to decide whether
the experiment is working as intended. If the list hasn't been used much
lately at the check-in time, or things are otherwise not going as you
originally imagined, that's a great time to decide to change things.

George is totally right that there are risks with creating new lists,
first because maybe you make a list that never takes off, frustrating
the people who signed up hoping it would be something, but second because
splitting communities can kill existing lists *too*.

So a third principle that I would propose is that the list runners should
keep an eye out for things that happen on the list that people from the
more "mainstream" lists probably want to know about, and send a periodic
summary or something. Same thing goes for watching other lists for things
that folks on the new list probably want to know about. And if you find
yourself spending a lot of time telling people from different lists about
what happened on the other lists, then it's time to stop and wonder if
things are set up wrong.

--Roger