Re: Idea for a secondary list server - Mailing list pgsql-www

On Tue, Feb 24, 2015 at 8:59 PM, Josh Berkus <josh@agliodbs.com> wrote:
> On 02/24/2015 12:08 PM, Stephen Frost wrote:
>> * Josh Berkus (josh@agliodbs.com) wrote:
>>> On 02/24/2015 12:57 AM, Dave Page wrote:
>>>>> Having a slew of not used mailing lists would not be beneficial to the
>>>>>> community in general, in my opinion.  Again, the issue is less about
>>>>>> the workload and more about the concern of having far more lists than
>>>>>> make any sense, 90+% of which are essentially dead.
>>>> I agree entirely.
>>>
>>> So the alternative is that we'll be running folks through the gauntlet
>>> of justification whenever they ask for a list, and requiring them to
>>> prove the popularity of their project/group/etc. before allocating them
>>> one -- something which is hard to do without having a list in the first
>>> place.
>>
>> I'm missing the part where this is a downside.
>
> Because it's hostile to community members who just want to do something
> cool.  Nothing destroys your enthusiasm for PostgreSQL faster than
> having a senior project member tell you you're not "worthy" of a list.
> The more so because the approval policy is *entirely* subjective; there
> are no written rules anywhere.  The current practice makes the
> completely unjustified assumption that the admin group is a fair and
> accurate judge of whether a new group is likely to be popular or not.
>
> For example, what does telling a new PUG organizer they can't have a
> list say about postgresql.org's attitude towards starting new user
> groups, and towards whatever part of the world they're from?
>
> This has also been a succession of having the admin group move the
> goalposts.  First, when we booted pgfoundry Dave promised we'd be more
> open about adding lists to @postgresql.org.

We haven't booted pgFoundry - it still seems to be up and running, and
my migration plans were cancelled when Marc insisted on keeping it
going.

> Then he said that we
> couldn't do that for resource/time reasons.   Now you're saying that
> even without resourcing reasons, you're just opposed to new lists period.

I probably did say we were short on resources, and at that time
probably did object to large scale moves to postgresql.org. Since then
however, the infrastructure has been made much more maintainable, so
I'm not sure I hold that view any more.

Even when I did, I was only one voice - and whilst it's nice to think
I have such power, the reality is that it's a group decision, not mine
alone.

>>> Alternatively, having a secondary list server with external
>>> infrastructure support and an automated termination policy for idle
>>> lists would give us an "incubator" area where people could prove the
>>> viability of their focused sub-communities ... or not.  This would allow
>>> us to NOT spend time arguing on this list over whether a specific group
>>> deserved a list or not.
>>
>> If we had a policy for idle lists then we'd be much better off.  That's
>> a completely independent consideration from where the lists are hosted.
>>
>> I'd encourage you to propose such a policy.  Once that's been hammered
>> out, we'll clean up the existing lists and address any requests for new
>> ones.
>
> I'm not going to waste my time discussing such a policy if it's not part
> of making it easier for folks to get new lists, which I note you're not
> promising.  I will be happy to help write and enforce a policy if it
> means that new list requests get dealt with in a fair and generous manner.

I think a policy would be a fine thing, and would not object to adding
lists if we can agree a reasonable set of criteria to use.

-- 
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company