On Sat, 15 May 2021, 6:35 am Paul Moore, <p.f.moore@gmail.com> wrote:
On Fri, 14 May 2021 at 21:18, Senthil Kumaran <senthil@python.org> wrote:

> > In other words, this isn't a technology problem, it's a people
> > problem.
> Both. I didn't suggest this is technology problem. We, have to
> choose one as per majority convenience.

Fair enough. That suggests that abstaining is the right answer for
people without a preference, I guess.

> > Do enough of the core devs actually *want* to hang out in a
> > chat forum to achieve critical mass and make it worthwhile?
> Yes, that's why the choice exist that "We don't want any chat platform".

OK, as long as you don't also assume that abstention means "not
interested in chat".

I think the wording of the poll here will matter quite a bit, as what really makes a chat channel work is having a regular core of people that are on at the same time and actually use it to talk to each other.

When that core is there, it's possible for newcomers to join, either by lurking and observing the interactions for a while, or by introducing themselves or asking a question and getting a reply. Without that core, newcomers will join, not see any traffic, not get any replies their messages, assume the channel is dead, and move on.

Time zones make it harder to build that core community, but even a channel that has a definite active time zone (e.g. US or European business hours) can be viable - you just need to indicate to newcomers the times when the channel is most active.

For myself, the only period where I was a regular in Python related IRC channels (including python-dev) was when I was working for Red Hat - I considered joining them part of my regular login process for a work day.

When I left RH, and with the notable exception of Steering Council meetings, I mostly went back to treating core development as an async-only activity (sometimes very async these days, as I'll now sometimes go for several weeks at a time without checking for any Python related messages if I'm not already involved in a specific discussion).

Since then, the only times I've joined IRC or Zulip have been for core dev and post-conference sprints, where I really was spending several hours in a block on Python in general and not already focused on a specific writing or development task.

So right now, I'd want a poll option that let me say "Only for specific events (e.g. sprints), and will use whatever is nominated for the event"

But to figure out if a new channel (or resuscitating Zulip!) would be viable, the information needed would be whether or not folks are prepared to *idle* in the channel (even if they went AFK), with a view to becoming regular participants, and what times they would be prepared to do that.

If there aren't half a dozen or a dozen folks willing to at least try to be regulars in the channel (I don't actually know the minimum viable seed number, but I assume there is academic research somewhere estimating it), and their expected availability times don't have at least some overlap, then I wouldn't expect a new channel to be any more successful than the previous attempts.