Re: Question to Martin: How are your Grants and Paid DPL Proposals Differnt than Dunc-Tanc
* Sam Hartman <hartmans@debian.org> [2019-03-20 14:10]:
> Let me start by saying that I think it would be valuable to find ways
> to get more people paid to work on Debian; I was excited to see that in
> your platform.
>
> I'm nervous because of our past experience in this area.
I wasn't sure how to reply to your message or what level of detail to
go into. While I think it's important to have a discussion about this
topic, I don't think -vote is the right place, nor am I sure it's the
right time.
> I'm really hoping you have answers though because I agree with you that
I do not have the answers. However, I believe it's important to start
a conversation about this topic.
> lack of funding is something that slows down our project and I'd
> like to find a way to get more people funded, but I think it's
> important for us to do so in a manner that people are comfortable
> with.
Of course. If you look at my platform, I wrote that the project has
to figure out what kind of project it wants to be. Maybe the project
decides that the project itself should have no paid positions, but we
should make that decision after carefully considering the pros and
cons, and acknowledging that times have changed and that our approach
limits what we can do (or we have to find creative ways to get around
these limitations).
> >Yes, external grants and external paid work. None of that is
> >controversial.
>
> by which I think you're saying that the idea of helping Debian
> Developers get external grants to work on Debian would not be
> controversial.
Yes.
As usual, when you say 95% things that are completely sane and
non-controversial and 5% that is controversial, people only focus on
the 5%.
The big part of my platform is to help create a more healthy
commercial ecosystem _around_ Debian. This involves getting more
companies actively involved and showing them the importance of getting
actively involved in Debian. For a few years, I worked at HP in a
team doing Debian related work and my role was to help get the team
more actively involved in Debian (unfortunately not with a lot of
success due to various reasons mostly unrelated to Debian). From this
experience I know that companies can find it challenging to figure out
how to talk to in Debian, where to get started, etc, and the project
can make things easier.
External grants is another area I mentioned. A lot of developers are
not even aware that this is an opportunity, so just creating awareness
is a first step. For example, the NLnet Foundation has two calls for
projects funded by the EU and I'm sure some Debian-related topics
would fit in well.
The whole idea is to create more _opportunities_.
> In that same mail you talked about potentially turning the DPL into a
> paid position, acknowledging that would be controversial.
Yes, I think we ought to have a conversation about this. This is not
my main focus at the moment though or the highest priority on the
list.
However, in each DPL election the question about time commitment comes
up, which shows us something. Also, like it or not, the DPL role is
unique in the project. While the influence of the DPL is limited,
there is still a great deal of influence.
I see several problems that the current situation has:
* Conflict of interest: I'm happy to see that your and Joerg's
employers would support your DPL activities. However, I've no idea who
they are or what they want from Debian. Maybe they use Debian and
want to give back with no strings attached, but I could definitely see
a situation where a company tries to exert undue influence over Debian
by having the DPL on payroll.
* Fairness and privilege: you and Joerg can run for DPL because of
your employer support; Jonathan and I can run because we're willing to
sacrifice a hit to our income. But how many potential DPL candidates
are we ruling out because they have full-time jobs that don't involve
Debian, that have children, etc, etc.
Of course, you can argue that this applies to _any_ volunteer in
Debian (and this is true; there's a reason why the majority of FOSS
contributors are from developed nations). But in Debian, there are a
lot of opportunities to contribute if you have little time: you can
maintain a few small packages, you can do translations, etc.
Being DPL is all or nothing. The more you put in, the more you get
out of it. (Of course, we have seen DPLs putting in varying amounts
of time and effort; but I think we agree that those who were able to
commit to being DPL had the biggest impact.)
(Again, I don't think -vote is the place to have that conversation;
but I wanted to give my thoughts since you asked.)
> In the past Debian did try to pay developers for working on Debian; the
> project I'm most familiar with was the dunc-tanc experiment [1]. That
> ended up being rather controversial.
>
> [1]
> https://lists.debian.org/debian-devel-announce/2006/10/msg00026.html
>
> It seems like having the Debian Project and DPL working to get more paid
> developers might run into some of the same issues.
> In particular there might be a perception that there would be two
> classes of developers and that volunteers would be
> frustrated/disappointed they were not getting paid.
Lots of good points were raised. However, this was in 2006 and the
world around has completely changed. We have also changed (in that we
have grown older and realise we all need to survive somehow).
Anyway, a good counter example is the LTS effort. While not Debian,
it's as close as Debian as you can get (and there are good arguments
that it's something Debian ought to do directly). It has been
completely non-controversial (apart from some minor questions about
the usage of the Debian name) and it has filled an important need that
we were not able to fill with volunteers. We would not have Debian
LTS today if it weren't for the sponsors providing the funds to pay for
this work (and Raphaël Hertzog's imitative in making it happen in the
first place and keeping it going). And let's face it, the lack of LTS
was a huge reason for companies to move away from Debian.
The concerns about creating two classes are well founded, but if you
look around, I think most of these concerns are more about _control_
rather than remuneration itself. For example, plenty of volunteers
spent time on Ubuntu, which is obviously dominated by a commercial
entity. They didn't have a problem with Canonical making (or hoping
to make) money out of Ubuntu. They had a problem that as a community
member you were a second class citizen. This is of course something
we have to avoid.
Another example is GNOME which has an Executive Director, admin staff
and which recently hired an engineer. I see no controversy there.
There are many similar projects like this. The world has changed
fundamentally in the last 10 years.
As Zack pointed out, many people are _already_ paid to work on Debian,
just not by Debian. Of course, that's an important distinction but
someone could also argue that it's unfair that some people are paid by
their employers while others aren't. (Why can you spend DPL time on
your employer's time and I can't?!?)
By creating more paid opportunities around Debian, those who want to
get a job doing Debian work can do so. Of course I can see envy when
10 people _want_ to be paid to do Debian work, but only one is paid by
the project while the rest has to fund unrelated work. But if 9 of
those people can find work elsewhere to work on Debian (probably with
better conditions that Debian could provide!), then Debian would just
be one employer among many others.
Also, I think we underestimate why people contribute to Debian. Some
*prefer* contributing to Debian as a hobby. Let's face it, a job
involves a lot of things you don't want to do, but you have to do them
anyway. While being a volunteer involves certain things you'd rather
not do, you have much more liberty to decide how to spend your time.
This makes a _huge_ difference.
BTW, Debian is already paying Outreachy students to work on Debian.
The only controversy here is around diversity, not about paying
people. Debian could offer more grants from Debian money to test the
waters to find out what people are comfortable with.
Finally, I see one risk: we keep repeating that something is
controversial even though we're not sure it's *still* controversial.
By repeating this myth, we're keeping it alive. The world has
changed. Debian has to finds ways to adapt.
--
Martin Michlmayr
https://www.cyrius.com/
Reply to: