[Qgis-psc] Plugin Repository Policy

[Trevor Wiens]

PSC Members,

I've joined the list to put forward a concern with the policy regarding
approvals for plugins. It is important that you all understand that I
really appreciate the QGIS project and am an active promoter of the tool in
my work, but the current criteria around plugins are untenable for me and
my team so I am writing to suggest some options for your consideration.

For context I'll mention that I have been teaching GIS courses in Canada
with QGIS for the last 6 years and promote the use of the platform in my
professional interactions. I have created plugins for using QGIS with
Marxan but also for in-person map based interviews as part of my business
and I when I make updates it is in response to bugs or users requests so I
need to be able to release those updates quickly.

In the past I was granted the ability to control releases of our plugins
but this new policy has now removed this option and it is becoming
problematic for me to respond in a timely fashion. I released and update to
my recent QMarxan Toolbox plugin over a week ago and it is still waiting
approval so I've just created my own repository to deal with this delay.
This is not an ideal solution and not a path I wanted to pursue but the
current policy leaves me no choice.

As the policy currently stands because I am not involved in the day to day
operations and have not and will not ever be traveling to Europe for a
Hackfest, I can not approve my own releases. As the head of an company that
provides software as a service in addition to complementary tools, I
understand the project's desire to maintain its good name. However there
must be a more realistic approach to how plugins are released. I have a
business to run so I don't have time to zoom off to Europe for conference
that is not relevant to my business. With a small team I am not able to
dedicate staff time to be regular contributors to the project.

You have now added the "trusted" status to the repository which opens a new
possibility that appears to have not been considered before. For the folks
that are part of the core team and regular contributors to the core project
then giving them trusted status makes sense. For other contributors such as
myself that have been contributing plugins for a long time and never had
any issues raised with them, it would be reasonable for them to be given
permission to have control of their own releases but not as trusted
plugins.

I would even consider contributing money so that someone from the core team
could do an in-depth code review of our plugins to identify bugs and
suggest ways of optimizing performance to gain the ability to control our
release process. I have my doubts about this idea however since I'm not
aware of anyone on the core team to has expertise in either Marxan or
in-person direct to digital interviews with land uses so I am not going to
spend my time and money educating someone; the review would have to be
about bugs, not functional design or purpose.

If some sort of accommodation can not be reached I will have no choice but
to continue to run my own repository. This is sub-optimal for many reasons
including needing to explain to users the reasons why.

Thanks for your consideration.

TSW
-- 
Trevor Wiens
Apropos Information Systems
aproposinfosystems.com
Calgary, Alberta
Ph. 403-973-5901
Fax 780-666-4580
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-psc/attachments/20161119/2265556f/attachment.html>