[Qgis-psc] Financial warning

Mon Mar 4 10:57:54 PST 2019

Hi Jonathan,

I have to disagree with you here. I am line with Régis argumentation. Bug
fixing is hard work and not exactly something you want to do in your
leisure-time just for fun. It is often difficult if not impossible to
estimate how much time is needed to fix an issue, esp in the light that it
is often / usually someone elses code the devs are fixing. This is not a
good work that can be assigned through an auction to the lowest bidder. It
would be a waste of time, because all the bidders would have to provide
estimates, which means that they have to examine the bug in depth, which
means that many devs would do the same work in parallel. In addition, I
would doubt that our core devs would even participate in such a bidding
process.

I am also 100% sure that our core devs are not introducing bugs on purpose
that they can later charge for fixing them - this is definitely not the
case in our project. If this would the case, it would be soon noticed by
other core devs, as pull request are reviewed by others.

I don't think we need a change here. Trusting devs is the thing to do here.

However, the assignment of available funds to devs is not an easy thing to
do. if there are suggestions for improvement in the process, I would be
happy to pursue them. We would have to introduce strict deadlines, until
which each potential bug fixer would declare the available times and then
someone would have to decide on the distribution. Not sure how fair and
efficient this would be.

I kind of prefer the current process where things are based on trust and
self-declaration. We just have to more clearly communicate how much funds
are available in total.

Greetings,
Andreas

On Mon, 4 Mar 2019 at 18:30, Jonathan Moules <jonathan-lists at lightpear.com>
wrote:

> Hi Andreas,
>
> Some thoughts if I may:
>
>  > "In general I think we should be grateful that the core devs
> participate in the paid bug fixing campaign for the low rate of 100 € h
> for us"
>
> This rate seems quite high to me. I don't see how paying "consultancy"
> level rates for bug-fixing is sustainable, even if they're at the
> "lower" end for consultancy rates.  I don't know what the
> overlap/relationship is behind the developers of features that have
> bugs, and the fixers of bugs, but this seems like its skirting close to
> the principle behind "write me a new minivan" Dilbert:
> https://dilbert.com/strip/1995-11-13 - thought I'm obviously not
> suggesting that's anyone's motivation!
>
>  > "- devs can only roughly tell me how many days they are really
> available. So I don't know in advance, until I get the invoice. "
> and
>  > "- information comes in at totally different times. Some devs only
> tell me quite late that they are available while others already spent
> some days working on it "
>
> I would suggest requiring standard business processes to be engaged.
> Tell folks ahead of time they can only invoice for X hours. If they want
> to do more, great, but that's their prerogative and they won't get paid
> for it. If the job takes longer than they estimate, well their loss. If
> it takes less time than the estimate, profit.
>
>  > - what would be a fair amount how to distribute the money between
> devs/companies and who would decide?
> Maybe having some sort of auction process where the lowest bids get the
> work.
>
> Just my 2p.
>
> Cheers,
> Jonathan
>
>
> On 2019-03-02 11:35, Andreas Neumann wrote:
> > Hi Paolo,
> >
> > It is very hard to coordinate the exact amount of bug fixing, due to
> > several reasons:
> >
> > - devs can only roughly tell me how many days they are really
> > available. So I don't know in advance, until I get the invoice.
> >
> > - information comes in at totally different times. Some devs only tell
> > me quite late that they are available while others already spent some
> > days working on it
> >
> > - what would be a fair amount how to distribute the money between
> > devs/companies and who would decide?
> >
> > In general I think we should be grateful that the core devs
> > participate in the paid bug fixing campaign for the low rate of 100 €
> > h for us - because if they would sell their services to clients at the
> > same time, they would probably charge 150-180 € an hour and earn more
> > and potentially have a more interesting task than fixing bugs in QGIS.
> > However through their efforts, in return they get the benefit of a
> > "better QGIS" where they could more easily sell their services around
> > QGIS, if it more stable and of better quality. And lets not forget
> > that we have also a lot of "unpaid" contributions towards QGIS.
> >
> > Personally, I would like to keep the process of accepting all
> > available bug fixing hours we get, but rather try to get more funds in.
> >
> > Thanks,
> > Andreas
> >
> > Am 02.03.19 um 12:27 schrieb Paolo Cavallini:
> >> Hi Andreas,
> >> thanks for this warning. What prevents us from sticking to the budget,
> >> and allocating to bugfixing only what is available?
> >> I think communicating our needs will help in this regard: I believe
> >> donors will be more motivated when they will know all the additional
> >> donations will go into bugfixing.
> >> All the best.
> >>
> >> On 02/03/19 12:17, Andreas Neumann wrote:
> >>> Hi,
> >>>
> >>> In that respect I hope that our change from sponsors  to sustaining
> >>> members (to be voted on during the upcoming AGM) will help to attract
> >>> more organizations to join as financial supporters. I am positive that
> >>> it will help, esp. for governmental organizations and universities.
> >>>
> >>> Andreas
> >>>
> >>>
> >>> Am 02.03.19 um 11:50 schrieb Andreas Neumann:
> >>>> Dear PSC,
> >>>>
> >>>> In the light of the current discussion around AGM and fast release
> >>>> pace, here is some additional information and concern I have as the
> >>>> financial manager of the project:
> >>>>
> >>>> Most of our financial resources go into bug fixing (> 50%) and I have
> >>>> the impression that this share is increasing every year. Currently I
> >>>> have to send out a warning that our financial resources are dwindling
> >>>> rapidly - our bank account is down below 30k €. For every release we
> >>>> spend about 10-20k more on bug fixing than allocated in the budget. As
> >>>> a consequence, we will have to either cancel the QGIS grants this year
> >>>> or skip bug fixing of release 3.8 or cut down drastically if we can't
> >>>> find additional financial supporters. I am very grateful that many
> >>>> core qgis devs can find the time to participate in the paid bug fixing
> >>>> (and they do a very good job!) - but unfortunately bugs are coming in
> >>>> quicker ...
> >>>>
> >>>> I don't want to sound overly pessimistic - but it is my duty to let
> >>>> you know about this development that we currently spend more than we
> >>>> get in. Something will have to change in this respect during the 3.8
> >>>> release or we'll have to skip the QGIS grants program this year.
> >>>>
> >>>> Andreas
> >>>>
> >>>> _______________________________________________
> >>>> Qgis-psc mailing list
> >>>> Qgis-psc at lists.osgeo.org
> >>>> https://lists.osgeo.org/mailman/listinfo/qgis-psc
> >>> _______________________________________________
> >>> Qgis-psc mailing list
> >>> Qgis-psc at lists.osgeo.org
> >>> https://lists.osgeo.org/mailman/listinfo/qgis-psc
> > _______________________________________________
> > Qgis-psc mailing list
> > Qgis-psc at lists.osgeo.org
> > https://lists.osgeo.org/mailman/listinfo/qgis-psc
>
>
>
> _______________________________________________
> Qgis-psc mailing list
> Qgis-psc at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/qgis-psc

-- 

--
Andreas Neumann
QGIS.ORG board member (treasurer)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-psc/attachments/20190304/9cc1e59f/attachment.html>