[Qgis-psc] On finances

[Régis Haubourg]

Hi All

On 04/12/2018 00:24, Nyall Dawson wrote:
> On Mon, 3 Dec 2018 at 18:14, Andreas Neumann <andreas at qgis.org> wrote:
>
>> Personally I wonder if we shouldn't have a break in features for a period of one release where all of use would focus on stabilizing QGIS and improving the quality - both of QGIS itself and also in the documentation. Have one release windows (4 months) where we do not accept any major new features, but concentrate on fixing issues.
> In practice I can't really see how this would be any different from
> the LTR, which already sees a steady flow of fixes and no features.
> Having a "bug fix only" release won't directly result in more incoming
> fixes or funding for people to work solely on fixes, rather it would
> be the same rate of fixes that we see coming into the LTR release
> already [1]. Instead it's likely to just result in developers working
> on queued features which will all land the day when that release
> "thaws", probably resulting in MORE issues then if these features had
> been introduced in one at a time.
Totally agreed. I vote against a bugfix only release.
>
> (That's without even touching the difficulty in changing the published
> release schedule now, where devs and customers have already planned
> timelines and feature implementation based on the current schedule!)

+1 again


IMO, we all need to advertise to our clients, trainees, friends,
children (hum) that they should find a way to subscribe to contracts
that help in fixing issues in the long run. Support contracts are a
solution we try to push forward at Oslandia.

We need to loudly explain that bugfixing is not only coding, it is
sometimes months of discussion to reproduce the conditions where an
issue appear (QGIS server slowdown for instance).

So bug triaging is also something we need to include in those support
contracts.

If one day we have enough resources for QGIS.org, I'd really be in favor
of someone sponsored for being the master bug triager (looking to
someone with an Italian name probably). I'm happy to give a help, but I
can only find time in pre-release period or when customers ask for this
explicitly. 

We also need to raise new "edge power users" that can contribute by
testing advanced use cases (thinking of node tool / forms for instance).
I really feel guilty somehow of having let this down when switching from
the contributing user to the service provider side. I must admit that
you really can test extensively when you have the real use cases and end
users on your back. So I'd like we try to push to message again and
again to all the new QGIS users.

I chose to get involved into the French Local Chapter to follow QGIS.ch
path, to try to change the customer attitude toward QGIS. This lets me
again less time to test and report issues, pretty frustrating it is, but
we need to grow the contributor community as fast as the user base. 

Régis

>
> Nyall
>
> [1] Possibly less, because a lot of existing issues are identified and
> fixed during development of new features
> _______________________________________________
> Qgis-psc mailing list
> Qgis-psc at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/qgis-psc

-- 
Open Source GIS Expert / Water management 

mail: regis.haubourg at oslandia.com 
tél: 0033 184 257 870
---------------------------------
http://oslandia.com/

OSLANDIA IS AN INNOVATIVE COMPANY SPECIALIZED IN GIS ARCHITECTURE. WE
PROVIDE SERVICE ON OPEN SOURCE SOFTWARE FOR WHICH WE ARE EDITORS OR
RECOGNIZED EXPERTS.