Re: [9front] Questions about community dynamics

[Kurt H Maier <khm@sciops.net>]

On Sun, Sep 24, 2023 at 09:02:39AM +0900, ieliedonge@wilsonb.com wrote:
> For example, some communities operate in a somewhat laissez faire manner,
> letting whatever happens happens. Other communities have explicit moderation.
> Others still have strong enough rapport that people just naturally get led into
> cooperative, helpful communication over time.

9front is far from one community.  

There's the irc channel #cat-v, which has its roots in contrarian idiocy
and people being mean to each other.  In #cat-v, we ban people for
belligerent ignorance, help-vampirism, and bigotry.  There are five 
people who can moderate that channel.  Most of them write most of the 
code in 9front.  The other one is me, a historical accident nobody's 
corrected yet.  I expect to be first against the wall when the
revolution comes.

Then there's this mailing list.  It's run by sl and I won't put words in
his mouth about how he chooses to run it.  sl also runs the 9front.org
website.  

There's a 9front community on gridchat who prefer it as a chat medium to
the hellhole #cat-v.  There's another one on discord.  There are
probably others.  Some of them run their own websites and knowledge
resources, like wiki.a-b.xyz.  These communities are run by other people
according to their own priorities.  

Most people who use 9front dip a toe in many of these options and wind
up gravitating towards a couple of them.  #cat-v and this mailing list
are the oldest, but that doesn't make them the correctest.  That
decision is up to each user to make for themselves.  They're all valid
communities worth participating in.

Getting your code in the tree just requires consent from anyone with
commit access to the repository.  Ori and sl manage those resources.
Getting commit access yourself is just as easy; I'm not aware of anyone
who asked for it having been turned away.  That doesn't mean it's
impossible, it just means the people who have asked have generally had
code to contribute.

> An element of the question is me wondering what 9front's values are. I see a
> problem with the habits of some on this mailing list but maybe that's just my
> uncalibrated view as an outsider.

9front is an operating system; it cannot have values.  Each community of
9front developers and users develops their own set of values.  I can
only really speak for #cat-v, because it's the only one I have any
moderation powers in.  I focus on banning hate speech advocates and
tire-kickers who make unreasonable demands on other people's time
without attempting to develop investigative skills of their own.  Other
than that it's basically anarchy.

> It's probably obvious that I'm also thinking about the recent discussion around
> a sed patch. As an onlooker, I see a few people giving a thumbs up and one
> loud, aggressive voice simply yelling that it's a worthless patch, without
> articulating anything helpful. As someone who'd like to eventually contribute,
> that kind of thing is pretty discouraging.

I find computers in general discouraging.  Learning when to ignore hiro
and when to listen to him is advanced black magic that can only come
from years of experience.  It's the same way with unix.

> So my meaning of "facilitation" here is asking both whether such
> a sentiment is part of the 9front community and if so whether there is any kind
> of mechanism to make things more welcoming.

As you can probably predict by now, this all depends on which 9front
community you're trying to participate in.  In the short term, I
recommend not taking it personally when someone freaks out about a
patch.  In the old days, pre-9front, the process was to submit a patch
and then get ignored for years.  Arguably, the whining is an
improvement.  If you don't want the peanut gallery, you can always email
some developers privately.

As for this particular patch, I'm ambivalent, but unix compatibility has
not in the past been a high priority, because plan 9 is not unix.  Maybe
this patch belong's in ape's sed.  I don't remember whether someone
investigated whether ape's sed has the same behavior as plan 9 sed?

Thanks for doing the work, whatever the results.

khm