Improving our integration with KDE application teams, and supporting companies

[Volker Krause]

On Friday, 24 August 2018 00:23:29 CEST Valorie Zimmerman wrote:
> Now that I'm coming out of jetlag, catching up on email and so happy
> to see the direction this discussion is taking.
> 
> On Thu, Aug 23, 2018 at 7:51 AM Jeremy Whiting <jpwhiting at kde.org> wrote:
> > Hi All,
> > 
> > On Mon, Aug 20, 2018 at 1:29 PM Adriaan de Groot <groot at kde.org> wrote:
> >> Mostly I'm repeating this item from Cornelius because it follows so well
> >> from what Valorie *originally* asked, rather than a bunch of
> >> misintepretations and discussing-something-else.
> >> 
> >> On Monday, 20 August 2018 10:58:05 CEST Cornelius Schumacher wrote:
> >> > I don't think that anybody has a problem with having a healthy
> >> > ecosystem of
> >> > companies around KDE. That's not the debate we are having.
> 
> Slight pushback here: Boud has gotten negative feedback for having a
> foundation to pay devels for Krita. Frank definitely got negative
> feedback, enough to take ownCloud out of the KDE ecosystem.
> 
> >> The thread started with Valorie exclaiming surprise that there was
> >> pushback on the entire notion of having companies / paid development
> >> around the KDE community. Some people have chimed in saying that that's
> >> not what they hear at all, so at this point I'm inclined to say that
> >> Valorie had the bad luck to run into one or two grumpy people.
> 
> Possibly. However I've heard a lot of this over the years. It could
> have been the same few grumpy people though; I didn't keep track. :-)
> 
> >> [Disclosure: Cornelius is presumably paid to work on Free Software-ish
> >> things throughout the week; I am paid to work on Free Software things
> >> for at least half of the week, and am looking for more.]
> > 
> > I also work on Free Software-ish things myself, but haven't done as much
> > KDE stuff in the past few years.
> > 
> >> But we could run a separate email thread with this question:
> >>  - do we (as a community) want an ecosystem of companies and paid
> >>  development>> 
> >> around KDE?
> 
> This was my main question, and I and most answering here seem to say
> YES. I believe that this growth is crucial to the on-going growth and
> health of the KDE community.

Right, I didn't find anyone objecting to that during Akademy either. If 
there's opposition against that (as previously indicated in this thread), I'd 
like to see the actual concerns so we don't have to speculate what those might 
be.

> > I would argue that whether we as a community want that or not we already
> > have it. I can name on a couple of hands quickly a number of community
> > members that either are paid to work directly on KDE stuff or on free
> > software in general and do KDE stuff in "Community" paid time by the
> > company they work for.

Yes, and it's like that even since before I joined KDE more than 15 years ago. 
So we do have quite some experience with the hybrid volunteer/paid setup, and 
we obviously managed to do that without killing volunteer contributions.

> And to the extent this is true, and those companies make enough money
> to keep paying them, this is a Good Thing. Blue Systems is great, but
> do they have a plan to make money? As far as I can see, it seems to be
> more a KDE charity. I think this is great, *and* I would like to see
> profit-making companies surround us as well. Not just a few, but
> *many* of them. Many of our applications, for instance, could be
> world-class, and support small companies which in turn support them.
> In addition, I'd like to see companies doing support for companies and
> individuals using KDE software on Linux, Mac and Windows. I'd like to
> see at least on the Windows Store many more applications making a bit
> of money.

Not sure if coming up with viable business plans is in scope for KDE, but 
"onboarding" companies that want to work with us for sure is. Ie. are there 
company-specific issues to look at as part of the onboarding goal? (All paid 
work I have been involved with was with already long-time community members, 
so I have no idea what challenges an outside company might face).

Regarding making money in stores, do we want companies to do that, or rather 
do this ourselves, like e.g. Krita does? (I see the Krita Foundation rather as 
an implementation detail of "us" than a company here).

> >> > > For those people who claim that having paid people work on a Free
> >> > > Software
> >> > > project will inevitably kill all motivation for volunteers, let's
> >> > > look at
> >> > 
> >> > > some examples within or close to KDE:
> >> > We need to get clear on what we are debating. It's not that paid people
> >> > are
> >> > a problem. It's about how this is done and who is paying them.
> 
> Which is why we need a lot of companies. Having only a few means each
> has an outsized influence on the direction of the developers and what
> direction they take the software.
> 
> >> > We have a very conscious standing decision that KDE e.V. does not pay
> >> > developers. This clearly separates paid and volunteer work there so
> >> > that
> >> > there can be no issue with harming volunteer motivation. We might want
> >> > to
> >> > revisit this decision but would need to be very clear about the
> >> > governance
> >> > of this work.
> >> 
> >> You're right. That's a very separate debate. That question is:
> >>  - are there any circumstances under which KDE e.V. itself should fund
> >> 
> >> development, by paying developers directly or hiring companies to do so?
> 
> IMO the e.V. should be spending money to develop KDE infrastructure --
> the website, both developer and user documentation, and our hardware
> and the sysadmins who care for it. In addition, we need Promo (which
> we now have, and paid and volunteer people happily work together),
> sprints and other meetings, which again I think could use paid staff
> and volunteers working together.
> 
> > Yes, I think this is the real question here. We already have paid
> > developers, the question is whether e.V. should get involved in that
> > aspect.

The "rule" that the e.V. doesn't pay for development is extremely old I think, 
from before the time we started using the more inclusive term "contribution" 
instead of "development" (and in that sense we are actually already no longer 
following it). From what was mentioned in past and current discussions, the 
reasoning behind that seem to be mainly three things:

(1) KDE e.V. should not influence the direction.
(2) The decision on who gets paid will result in jealousy and internal 
conflicts.
(3) It hurts volunteer contributions.

Things have changed though. With the goals we do have direction, defined and 
voted on by the wider community. KDE e.V. paying people to work on these goals 
would not be influencing direction, but merely doing its job to support the 
community.

For (2), yes, there is the potential of things turning awkward, but far less 
than it used to be 15 years ago when most of us were students dreaming of 
getting paid for what we were doing anyway. Today a lot more people here have 
decent well-paying jobs (and families to support), and are therefore 
presumably a lot more reluctant to leave their jobs for a possibly not long-
term sustainable KDE job. And on top of that, Blue Systems already provides 
such jobs for a number of obvious candidates.

For (3), yes, there are examples of this happening elsewhere that were cited 
in that discussion, but in our long history it hasn't happen here. And whether 
it's KDE e.V., the Krita Foundation or Blue System paying doesn't make such a 
big difference for this IMHO. Yes, there are ways of getting this wrong for 
sure, but see below for an example on how to do this in a way that would IMHO 
help rather than hurt volunteer contributions.

Therefore I don't think KDE e.V. paying for contributions should be off-
limits. (That in turn opens new fundraising possibilities, as you can then 
actually promise specific things being done, see Krita).

Contracting specific tasks/bounty programs have been mentioned as one approach 
to this, to get work done. While this gives us decent cost control it would 
limit the pool of candidates to people working as freelancers or development 
service companies, in both cases you pay extra for the limited commitment.

Cornelius suggested to focus more on "investments" than getting work done. 
Paying someone who doesn't just get work done but also looks at growing the 
product and the community would be going into that direction IMHO.

Taking Krita as example: Contract Boud to fix bug A, B and C and implement 
features X, Y and Z, or pay him for a year to do what he does to make Krita 
and its community even more awesome? IMHO the latter is far more interesting, 
for everyone involved. And I think focusing on paying people with a Boud-level 
of dedication is also be a good way to avoid hurting volunteer contributions.

Regards,
Volker

> > One possible way to remove the emotional aspect of this would be to have
> > the board or some work group come up with a bounty list of long-standing
> > issues we would like to see fixed and whoever (individual or
> > group/company) is able to properly fix the issue gets the money for that
> > bounty. This could be tricky for board members to stay out of the
> > conflict of interest gray area. (e.g. Hey, Lydia created that bounty
> > because she knew the company/ friends she works for/with could tackle it
> > or somesuch) but with enough of a selection process of which bounties we
> > want that could be decreased quite a bit (e.g. The whole kde ev
> > membership voted on these bounties, no way for an individual to have
> > influenced the bounties enough to matter).
> > 
> > BR,
> > Jeremy
> 
> I think the coming effort to survey the state of our documentation is
> such an effort, and there was already some pushback over spending even
> a modest amount on that. I hope we've gotten past that, because we
> need someone outside of KDE to do the survey, in my opinion.
> 
> >> .. and there's a third question, raised by yourself at Akademy and
> >> touched by>> 
> >> Sven just now:
> >>  - if KDE e.V. has money, and doesn't spend it directly on development,
> >>  how
> >> 
> >> can it best support KDE development indirectly?
> 
> It all depends on how we define the terms, right? Is working on our
> fundamental infra "development"?
> 
> >> These are three distinct questions, and really we should be very very
> >> clear on which one we're tackling (which, thank you, you have pointed
> >> out -- as has Sune, and others, and now I'm just repeating stuff :) )
> >> 
> >> [ade]
> 
> Valorie

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-community/attachments/20180826/1211a1c8/attachment.sig>