[Haskell-community] Is it reasonable to poll the community on this ML?

[Paolo Giarrusso]

On 29 August 2016 at 17:21, Gershom B <gershomb at gmail.com> wrote:
> On August 29, 2016 at 11:15:19 AM, Paolo Giarrusso
> (paolo.giarrusso at uni-tuebingen.de) wrote:
>
>> If the poll was announced there, there would still be extra friction.
>> But IIUC only the mailing list was announced there.
>
> There is no poll. There is a modest discussion kicked off by Jason
> Dagit (who used to serve on the committee, but has not been on it for
> some time now) about a modest change (at this point swapping the
> bitrotted minimal installers for the HP minimal installers which are
> current).

One might initially think that "what's the best entry point into using
Haskell" is a simple question.
However, even without taking a side, we all know the topic is in fact
highly contentious and it has been in the past.

And since you're a community leader and I'm just a modest Haskeller,
I'm sure you realize the discussion topic is not perceived as modest.
At best one could argue the discussion *should* be modest, but there's
technical content to it.

It's also known that the committee's decision have been questioned for
allegedly not listening to the community on this topic.
I dislike the allegations (and find the tone unproductive), yet I
think community input would be important and it's in the committee's
best interest to both listen and be perceived as listening.

> The committee does not operate by poll, it operates on the
> basis of broad discussion (with this list being the preferred venue)
> and then making choices amongst committee members as informed by that
> discussion. This is laid out at
> https://wiki.haskell.org/Haskell.org_committee

Are these procedures the best to achieve the Committee's stated goals?

>> I might understand the concern about archiving, but haskell-cafe
>> solves that. And "the committee can't be expected to follow
>> discussions" and "is empowered to act" does sound like "the committee
>> can't be expected to listen to the community”.
>
> It means that committee members should be expected to chase all over
> social media and sort through lots of poor signal/noise ratio to find
> potentially relevant discussions at all times. Rather, it is better to
> centralize these things to the extent possible.That’s all.

I think that's a strawman. I didn't propose to spend the day on
Twitter, but to solicit input on questions of general relevance in
venues where the community is. To send a post to haskell-cafe and
follow the discussion—that doesn't imply following the rest of the ML
(at least with the Google Groups interface, I'm sure there are many
other solutions).

I realize that might require time, but I frankly don't expect that
"seeking to service the open source Haskell community" is easy.

Cheers,
-- 
Paolo G. Giarrusso - Ph.D. Student, Tübingen University
http://ps.informatik.uni-tuebingen.de/team/giarrusso/