[Qgis-psc] Developer roles

[Vincent Picavet (ml)]

Hi,

On 26/06/2020 11:27, Nyall Dawson wrote:
> Vincent - snipping out the 99% of your email I agree with -- don't
> take my comments below as meaning I disagree with the bulk of your
> reply! :)

Ok, great to hear :-)

Let's discuss the remaining disagreement then.

> On Fri, 26 Jun 2020 at 18:56, Vincent Picavet (ml)
>> I tend to disagree here, because I consider that growing the community and
>> onboarding new developers has much more value in the long run, than having a
>> code developped 20% faster.
> 
> Honestly, in the case of the current crop of "core developers", I
> think 20% is a huge disservice to their skills. I'd honestly,
> conservatively, estimate that a developer who already has intimate
> knowledge of the QGIS codebase, Qt, GDAL etc will be at a minimum 10x
> faster than a new contributor to the project. I don't think anyone
> could ever argue that we only get 20% more value from funding Even's
> time vs someone completely new to the project! (Try 20k% more value,
> and you're in the right ballpark...)

I did not want to underestimate the efficiency of long-term core developers. Of
course they are faster, and probably produce much better quality for the code.
"20%" was not to take as an absolute number, forget this number. My point is not
about numbers anyway.
The reality is much more complex than this : QGIS codebase is huge and even a
long term contributor may have trouble in some areas of the code. And other
contributors having never contributed to QGIS could be really efficient because
they focus on an area they are expert in ( imagine Even before any contribution
to QGIS core, but knowing extensively GDAL ).
My point is : diversity and growing developer's community is difficult, while
highly required for a project to be sane, avoid bus-factor and continue to strive.


>> I would personnaly like to see the grant program as an onboarding program. But
>> this is not a strong request, and maybe we could have another program dedicated
>> especially to onboarding new developers ?
> 
> I disagree - I think given the current round of high quality
> submissions it's clear that even when we restrict the program to
> established contributors there's more high value work vs funds
> available. If we divert some of these grant funds to onboard new
> contributors then it's definitely going to be at the cost of the
> high-priority submissions we already receive.
> 
> But regarding a dedicated program -- isn't GSOC ideal for this? That's
> exactly what it's oriented to, and doesn't cost us anything...!

I think we are not really talking of the same thing. GSoC is for students,
junior developers. While it is a great program which should probably be used
more to onboard new junior developers to the project, it is not really what I
had in mind.

I am talking about onboarding new developers, not junior developers but people
with real talent and skills, experienced developers with knowledge C++, Qt,
Python, etc.
GSoC is not adapted for them, and we do not really have any incentive for this
kind of people to join the project. This is where room for improvement lies.

>>> Fully agree, we should have the same thing for a "non-coding" contributor
>>> (think Richard, Harrisou, Giovanni ...) which do an amazing job for QGIS as a
>>> project but don't express it in C++
>>
>> +1
> 
> I'm also +1, but I think we should separate the terms explicitly here.
> I would like to see the "developer endorsed by QGIS" title as a
> reflection of their **coding** talents, and something which can be
> used on a CV/resume as a reflection of their development skills. The
> "endorsed, high value" community contributor title should be used for
> non-developer contributors only, and have it's own set of guidelines
> for eligibility.

I always find differentiating "the developer" from other contributors tends to
minimize "other contributors" importance and value. But I understand that
eligibility could have different rules, and therefore a specific role would be
required. We would have to be careful to give them as much importance as
developers, if not more.

Best regards,
Vincent