Results of Fedora Participation Survey?

[Gregory M. Zysk]

I would like to draw attention to the survey that was administered by
the professors at Duke university over a month ago. I believe that this
survey should shed some light on the contributions and the various
priorities and methods which Fedora contributors define as their needs
and interests. This, in turn, should give a good basis to start from so
the data can be used to brand and market Fedora properly, as well as,
help build a better way for administration and governance of the
project. Eventually, if used properly, it can also provide the basis for
the community to avoid and reduce the growing conflicts that lead to
many contributors leaving the project.

So, in short, do we have any results for this survey?

Furthermore, do we have a methodology to accompany this survey to show
why these questions were asked and why the specific target groups were
chosen to survey?


-- 
Gregory M. Zysk
Fedora Ambassador EMEA

Fedora: http://fedoraproject.org/wiki/User:Gmzysk
Identica: http://identi.ca/gmzysk
Blog: http://openleadership.blogspot.com
IRC: gmzysk@fedora/gmzysk

[Paul W. Frields]

Replies set to advisory-board.

On Mon, May 17, 2010 at 01:45:57PM +0200, Gregory M. Zysk wrote:
>  I would like to draw attention to the survey that was administered
> by
>  the professors at Duke university over a month ago. I believe that this
>  survey should shed some light on the contributions and the various
>  priorities and methods which Fedora contributors define as their needs
>  and interests. This, in turn, should give a good basis to start from so
>  the data can be used to brand and market Fedora properly, as well as,
>  help build a better way for administration and governance of the
>  project. Eventually, if used properly, it can also provide the basis for
>  the community to avoid and reduce the growing conflicts that lead to
>  many contributors leaving the project.
>  
>  So, in short, do we have any results for this survey?
>  
>  Furthermore, do we have a methodology to accompany this survey to show
>  why these questions were asked and why the specific target groups were
>  chosen to survey? 
The Duke researchers have sent me a list of some additional
information that they need to continue their work.  Basically, they
need someone to run a few SQL queries via JSON:

(1) The count of unique people, determined by full names and/or email
addresses, active in at least one mailing list/group, which would
provide an estimate of the number of active people in the Fedora
Accounts System.  The count won't include actual names or addresses.

(2) The distribution of mailing lists/groups per person (e.g., X
people are in 1 mailing list/group, Y people are in 2 mailing
lists/groups, Z people in a 3 mailing lists/groups, etc), which would
provide an estimate of how many mailing lists/groups active people are
typically part of in the Fedora Accounts System.  Again, these counts
won't include actual names or addresses

In addition, there are some counts they need for specific lists/groups
based on actual responses, and I'll work with someone from
Infrastructure to get those.  I expect the researchers will be
publishing a full report along with the survey results to establish
context around it.

-- 
Paul W. Frields                                http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
  http://redhat.com/   -  -  -  -   http://pfrields.fedorapeople.org/
          Where open source multiplies: http://opensource.com

[Ambassadors] Results of Fedora Participation Survey?

[Gerard Braad]

On Mon, May 17, 2010 at 3:06 PM, Paul W. Frields <stickster(a)gmail.com&gt; wrote:
>  I expect the researchers will be
>  publishing a full report along with the survey results to establish
>  context around it. 
Is anything known about the time of publication?

Gerard — 吉拉德

[Amit Shah]

On (Mon) May 17 2010 [09:06:17], Paul W. Frields wrote:
>  
>  The Duke researchers have sent me a list of some additional
>  information that they need to continue their work.  Basically, they
>  need someone to run a few SQL queries via JSON:
>  
>  (1) The count of unique people, determined by full names and/or email
>  addresses, active in at least one mailing list/group, which would
>  provide an estimate of the number of active people in the Fedora
>  Accounts System.  The count won't include actual names or addresses. 
I'll note here that I consider myself to be a Fedora developer even
though I'm not active on the Fedora lists: I'm active on upstream lists,
though. Any way to get that data? (I know this is tough.)

		Amit
-- 
http://log.amitshah.net/

[Paul W. Frields]

On Tue, May 18, 2010 at 06:22:28PM +0530, Amit Shah wrote:
>  On (Mon) May 17 2010 [09:06:17], Paul W. Frields wrote:
>  > 
>  > The Duke researchers have sent me a list of some additional
>  > information that they need to continue their work.  Basically, they
>  > need someone to run a few SQL queries via JSON:
>  > 
>  > (1) The count of unique people, determined by full names and/or email
>  > addresses, active in at least one mailing list/group, which would
>  > provide an estimate of the number of active people in the Fedora
>  > Accounts System.  The count won't include actual names or addresses.
>  
>  I'll note here that I consider myself to be a Fedora developer even
>  though I'm not active on the Fedora lists: I'm active on upstream lists,
>  though. Any way to get that data? (I know this is tough.) 
I don't see any reason for us not to publish it on a wiki page, since
we're going to the trouble of producing it for the Duke researchers
anyway.  Anyone in the community with appropriate infrastructure
access could produce these numbers, after all.

In the future, it may be helpful for us to build these functions into
statistical gathering tools that the Community Architecture team is
developing.  This summer, Ian Weller is working on those tools and I
imagine he'd be interested in the code we use here.

-- 
Paul W. Frields                                http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
  http://redhat.com/   -  -  -  -   http://pfrields.fedorapeople.org/
          Where open source multiplies: http://opensource.com

[Mel Chua]

> > I'll note here that I consider myself to be a Fedora
> developer even
> > though I'm not active on the Fedora lists: I'm active on upstream lists,
> > though. Any way to get that data? (I know this is tough.)
>
>  I don't see any reason for us not to publish it on a wiki page, since
>  we're going to the trouble of producing it for the Duke researchers
>  anyway.  Anyone in the community with appropriate infrastructure
>  access could produce these numbers, after all.
>
>  In the future, it may be helpful for us to build these functions into
>  statistical gathering tools that the Community Architecture team is
>  developing.  This summer, Ian Weller is working on those tools and I
>  imagine he'd be interested in the code we use here. 
I'll contact the Duke researchers and ask them if they can join the 
conversation on this list briefly to answer the questions that have been 
raised here, although I'll also note that from experience, academic 
researchers can't always be completely open about their research (even 
if they'd like to) due to institutional requirements. So... gimme a 
moment, and I'll cc the list on that message.

--Mel

[Mel Chua]

Hi, Jonathon!

Hope the research study is doing well. A number of us in the Fedora 
community were wondering what was going on with the study - if you have 
a moment, could you take a look at this thread and shoot an update to 
the mailing list (just reply-all to this email) with a quick note on 
what's going on and how we can help, and when we'll be able to see the 
data from your study?

The thread in question is here:
http://lists.fedoraproject.org/pipermail/marketing/2010-May/012849.html

Thanks!

--Mel (on behalf of the Fedora crew)

[Amit Shah]

On (Tue) May 18 2010 [10:38:56], Paul W. Frields wrote:
>  On Tue, May 18, 2010 at 06:22:28PM +0530, Amit Shah wrote:
>  > On (Mon) May 17 2010 [09:06:17], Paul W. Frields wrote:
>  > > 
>  > > The Duke researchers have sent me a list of some additional
>  > > information that they need to continue their work.  Basically, they
>  > > need someone to run a few SQL queries via JSON:
>  > > 
>  > > (1) The count of unique people, determined by full names and/or email
>  > > addresses, active in at least one mailing list/group, which would
>  > > provide an estimate of the number of active people in the Fedora
>  > > Accounts System.  The count won't include actual names or addresses.
>  > 
>  > I'll note here that I consider myself to be a Fedora developer even
>  > though I'm not active on the Fedora lists: I'm active on upstream lists,
>  > though. Any way to get that data? (I know this is tough.)
>  
>  I don't see any reason for us not to publish it on a wiki page, since
>  we're going to the trouble of producing it for the Duke researchers
>  anyway.  Anyone in the community with appropriate infrastructure
>  access could produce these numbers, after all. 
Oh; I just meant that if we're skimming Fedora mail lists to gauge the
number of active developers against the number of responses we got, we
may not get accurate results (due to some developers being more vocal on
upstream lists rather than Fedora lists).

		Amit
-- 
http://log.amitshah.net/

[Paul W. Frields]

On Tue, May 18, 2010 at 11:05:28PM +0530, Amit Shah wrote:
>  On (Tue) May 18 2010 [10:38:56], Paul W. Frields wrote:
>  > On Tue, May 18, 2010 at 06:22:28PM +0530, Amit Shah wrote:
>  > > On (Mon) May 17 2010 [09:06:17], Paul W. Frields wrote:
>  > > > 
>  > > > The Duke researchers have sent me a list of some additional
>  > > > information that they need to continue their work.  Basically, they
>  > > > need someone to run a few SQL queries via JSON:
>  > > > 
>  > > > (1) The count of unique people, determined by full names and/or email
>  > > > addresses, active in at least one mailing list/group, which would
>  > > > provide an estimate of the number of active people in the Fedora
>  > > > Accounts System.  The count won't include actual names or addresses.
>  > > 
>  > > I'll note here that I consider myself to be a Fedora developer even
>  > > though I'm not active on the Fedora lists: I'm active on upstream
> lists,
>  > > though. Any way to get that data? (I know this is tough.)
>  > 
>  > I don't see any reason for us not to publish it on a wiki page, since
>  > we're going to the trouble of producing it for the Duke researchers
>  > anyway.  Anyone in the community with appropriate infrastructure
>  > access could produce these numbers, after all.
>  
>  Oh; I just meant that if we're skimming Fedora mail lists to gauge the
>  number of active developers against the number of responses we got, we
>  may not get accurate results (due to some developers being more vocal on
>  upstream lists rather than Fedora lists). 
Oh, I see Amit.  Thanks for the clarification.  Yes, we will likely
miss a lot of the upstream development.  Since the researchers are
concentrating specifically on the Fedora Project as an integration
point, upstream lists are outside the scope for this particular
survey.  An additional complicating factor, besides the obvious ones
of just a multiplication of effort, is privacy concerns for the lists
that are not operated under the Fedora Project's privacy policy.  That
policy allows us to use the contents of Fedora archives for research
purposes but of course doesn't apply to independent projects outside
Fedora.


-- 
Paul W. Frields                                http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
  http://redhat.com/   -  -  -  -   http://pfrields.fedorapeople.org/
          Where open source multiplies: http://opensource.com