[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Question for Stefano: New contributors



On Thu, Mar 17, 2011 at 08:49:45AM +0000, Lars Wirzenius wrote:
> The GR to be welcoming to non-packaging contributors was a good thing.
> However, it addresses the end of the membership process, the actual
> membership. Before they even enter the NM process, we need to get them
> contributing to Debian.
> 
> What is your assessment of the situation at that end, Stefano? Are we
> frequently getting non-packaging people contributing to Debian, and
> becoming a part of the development community, or are such people rare
> exceptions?

Unfortunately, I'm writing this on a plane, so I can't check actual
data, but I bet we have the means to evaluate whether non-packaging
contributions are an exceptions or not. We can look at stuff like active
Alioth users (those who have committed something in the last
$reporting_period), number of bug reporters (+ patch), etc.

No matter the current amount of those people, letting them know that we
value their contribution (as IMHO we did with the GR you mention) is the
first needed step to get more of them.

> What, in your opinion, could we do to attract more of them? What
> barriers are there that we could lower?

The main barrier I see is documentation, I've the impression we're not
particularly good at that. We are quite good at announcing stuff, both
to developers (hello d-d-a) and to users (hello wonderful
press/-publicity work), but announcements are not the same thing as
documentation. And FOSS projects without good documentations have a big
handicap to overcome when compared to similar FOSS projects with good
documentation.

We likely have a chicken and egg circle to break here. We need good
documentation; to that end we need people who is good at (and willing
to!) writing documentation; so we need to advertise it properly ...

I've had the above as a background thought for quite a while, but I
failed to propose initiatives to improve the situation. I would really
welcome suggestions on this topic.

> Are there particular kinds of skills we need?

- documentation writers
- bug triagers
- testers
- translators
- graphic artists
- journalists
- porters
- "animators" (people who periodically and reliably organize community
  activities such as teaching sessions, interviews, and any other
  IRC- or other social-based activities)
- developers (with a vocation to write the software which power the
  Debian infrastructure; we have mixed fortune on this front, but let's
  keep in mind that a packager is not necessarily a good software
  developer)
- ... tens more profiles I'm probably forgetting ...

In most of the above category you can point to Debian teams who are
pursuing those goals already, but I still have the impression we are way
under staffed with respect to what we could benefit from.

> What about packagers? How could we attract more of them, and get them to
> stay in the project?

For staying in the project I don't have a good answer, beside the
obvious one of making Debian a pleasant FOSS project. All in all, it's
probably normal to have some turn over, according to the range of
interests that hackers often go through.

Regarding attracting more packages, my experience in speaking with
potential contributors at conferences and trade show insists in saying
that they are generally at loss in understanding how they can help /
what they can do to help out with a specific package. While in many
cases that is probably a symptom of not enough determination, it's also
a symptom that we are not good at telling them what we need. I routinely
point people to http://wiki.debian.org/Teams/ and at the packaging team
pages which are there, but I'm not that sure those pages are routinely
updated. IOW, we are back at documentation! I feel we need to convince
ourselves that taking the time to *tell* people what we need actually
helps in getting help.

At the same time, we can work to reduce the overall need of packaging
manpower, by reducing *internal* barriers (AKA "strict code ownership").
We have ways to do that already: using NMUs more, having more and more
packages whose VCSs can be committed to by all DDs, etc. I know I sound
repetitive when I write about those, but is part of the strategy *g*.

Cheers.

-- 
Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Quando anche i santi ti voltano le spalle, |  .  |. I've fans everywhere
ti resta John Fante -- V. Capossela .......| ..: |.......... -- C. Adams

Attachment: signature.asc
Description: Digital signature


Reply to: