Re: Adding PGInstaller to the Downloads section

On Tue, Oct 16, 2018 at 8:13 PM Umair Shahid <umair(dot)shahid(at)2ndquadrant(dot)com>
wrote:

>
>
> On Sun, Oct 14, 2018 at 1:20 AM Jonathan S. Katz <jkatz(at)postgresql(dot)org>
> wrote:
>
>> Hi,
>>
>> On 10/8/18 9:11 AM, Umair Shahid wrote:
>> >
>> >
>> > Hi. Any updates on this?
>> >
>> > If there is specific feedback on how to make this patch better, please
>> > let me know. Thanks!
>> We're sorry this is taking some time to get to - with the major release
>> and some other larger pgweb projects, in addition to some of the
>> discussion on how to set this up, we have not been able to prioritize
>> it. We will get to this in the weeks after the PG11 launch.
>>
>> Our plan is to focus on doing some restructuring on the downloads page -
>> which there appears to be consensus - before adding additional
>> installers. Getting this done will allow us to add the installers
>> shortly, if not immediately, after.
>>
>
> Thank you for the explanation Jonathan, though I am not sure I quite get
> it. The Downloads page needs restructuring, which apparently needs to be
> done regardless of whether Postgres Installer is listed there. I don't
> understand why Postgres Installer should be held up because of it.
>
> We moved quickly to address all concerns raised in the first round. It
> sends a rather discouraging message from the community if I find out about
> this decision more than a month after I sent the revised patch, and that
> too after repeated reminders.
>
> I believe this puts Postgres Installer at a disadvantage, and without good
> reason. I would urge you to kindly reconsider. Thanks!
>
>

Hi!

It's not about putting any of the installers at an advantage or at a
disadvantage. It is whether we put our *users* at an advantage or a
disadvantage, and the consensus on that is pretty clear -- we are not doing
our users any service by having a whole bunch of different installers on
the same page without a clear and working structure. And yes, adding more
options without structure and clearness to the end user on which should be
used for what, will be of negative value. What we have now does not
"scale", and piling more on it just makes it work.

Thus, we need to work on that structure first. Unfortunately, there seems
to be a distinct lack of people willing to work on *that* part, so it takes
longer.

It is definitely true that you have worked quickly to address the problems
that were there with the patch initially, and that is definitely
appreciated. But just like with PostgreSQL itself, that's no guarantee that
it will get in.

It's just like the main PostgreSQL sourcecode -- we don't apply patches
even though the author updates them quickly based on comments if it's
determined that they are counterproductive to the bigger goal of the
product. It's the same here, just for the website.

--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>