[Qgis-psc] Voting for grants

[Nyall Dawson]

On Tue, 2 Jun 2020 at 18:00, Denis Rouzaud <denis.rouzaud at gmail.com> wrote:

> 2. Some are made by unknown people from the community, and I have mixed feelings about this (openess vs rewarding of involvment)
>

I also share this concern. There's quite a few proposals this round by
newcomers, and some explicitly state in the reply/associated QEP that
they are seeking the assistance from established community members in
the completing the work (I'm paraphrasing here, but I'm trying to
avoid singling out particular proposals). This concerns me on a few
fronts:

1. Existing maintainers are already shouldering a huge a burden. **In
my understanding** the QGIS grant project isn't a "Google Summer of
Code" style project targeted at introducing newcomers to the project,
but rather a means of funding the established community members with
conducting difficult work which would otherwise be impossible. I don't
think it's fair for proposals to actually do the opposite, and place
further stress/workload on existing community members.

2. Value for money: is qgis.org and the project sponsors getting the
best return on investment if they are funding newcomers to learn the
QGIS project ecosystem, vs already established members who know the
systems, policies, and expectations intimately.

3. Lack of a proven track in making contributions which are acceptable
to the project: while I'm sure that all submitters have the skills to
complete the work they propose, we've seen in the past that the way
some newcomers contribute to the project just isn't compatible with
the QGIS approach. I definitely don't want to mention names here, but
there's been occasions in the past where newcomers have submitted huge
amounts of work which ultimately led nowhere, because they refused to
adapt their coding style/conventions with the established approaches
used by QGIS. Raw development skills alone aren't sufficient to get
code into QGIS -- it also takes the submitter's willingness to adapt
their usual approaches to match the existing codebase.

I'm not concerned about the contributors portion of the voting members
here -- these members all will already know who is firmly established
with proven track records of contributions to the project. I'm more
concerned about how we can make the community (user group) voting
members aware of this distinction and help them make informed votes
which take this into consideration.

(Full public disclosure: while I haven't personally submitted any
proposals myself this round, I have been involved in writing up the
two geodatabase driver related proposals and am also taking a code
mentor/reviewer role in the dbManager->browser proposal. I'm not 100%
impartial here, keep that in mind while reading my opinions!)

Nyall