Hey Sergio,

I'm a big fan of Zulip as a communication tool. I've used it for various
teams, and I think it's shockingly good. I did want to comment on one
thing, though.

On Tue, Feb 10 2026, Sergio Pastor Pérez wrote:
> I'm not familiar with it, I think it does not provide the sync/async
> hybrid workflow that Zulip does.

I think conflating sync and async communication is not purely positive.

As someone who lives in a different timezone  of the Guix
community (UTC+11 at the moment), I benefit from the separation between
sync/async communication in two ways:

1. I know what's expected for response times. With something like Zulip
it's not clear how long you should (or will) wait for responses. I think
the Zulip UI nudges you towards the sync paradigm, despite supporting
async well. If you get some quick responses then you may end up feeling
like you've resolved a conversation before everyone has had a chance to
be involved. This can make things go faster, but can also exclude people
who aren't able to engage at that pace. I've seen this happen in my own
distributed teams in the past, where people disengage entirely because
they're usually asleep during significant conversations.

2. I don't feel pressure to keep up to date with what's happening on
IRC. I generally assume "if it's important, there will be a thread on
guix-devel about it", so I'm only on IRC opportunistically (and usually
just to help people who need help). The assumption that guix-devel is
the place to watch has already been partially eroded by the switch to
Codeberg (it's still async, but I find it much harder to keep track of
everything). Switching to Zulip would put more pressure on me to keep up
on the sync conversations to stay in the loop.

I'm not completely opposed to the idea, because I think Zulip is great,
but I would like us to be mindful of the costs of collapsing sync and
async communication into the same form. I think it is possible to avoid
these costs by establishing expectations carefully, but that won't
happen automatically.

Carlo

Reply via email to