[Mageia-discuss] Mageia logo proposals and selection

[Michael Scherer]

Le mercredi 20 octobre 2010 à 14:26 +1300, Graham Lauder a écrit :
> On Tuesday 19 Oct 2010 11:38:39 Michael Scherer wrote:
> > Le mardi 19 octobre 2010 à 09:06 +1300, Graham Lauder a écrit :
> > > On Tuesday 19 Oct 2010 04:27:29 Frank Griffin wrote:
> > > > In FOSS, it doesn't.  If enough people agree with your objective, you
> > > > may find that you have enough critical mass to produce a derived distro
> > > > with a face and personality which matches your objectives.
> > > 
> > > This is one of the interesting elements of FOSS marketing that I've
> > > talked about in the past.  That Marketing department, which in a
> > > corporate world always has the ear of management more so than the
> > > Development people simply because of human interaction capabilities, has
> > > to turn it's focus inward.  The problem is, an one I've been trying to
> > > avoid here, is that it becomes insular to the exclusion of all else and
> > > then the community stagnates and spirals into irrelevancy.  For the
> > > community to grow there has to be a dynamism, (and I'm talking grow in
> > > terms of the community of contributors)  Userland is the big billboard
> > > of that dynamism.  Ubuntu for all it's faults and annoyances has taught
> > > us one thing, high visibility in Userland attracts contributors.
> > 
> > Then what Fedora and Debian has taught us ?
> > 
> > Because AFAIK there is also lots of contributors in Fedora, as there is
> > in Debian, and I think they didn't really choose the high visibility
> > path to get them. So I do not think we can really find a direct
> > correlation between "ubuntu has lots of users" and "there is lots of
> > contribution".
> 
> 
> Debian is an interesting case in viral marketing in a highly interconnected 
> demographic.  I always remember the "OMG we have a new release!" that used to 
> race round the maillists and Usergroups.  It never really had a market share, 
> rather it had almost a monopoly in its chosen demographic.  It is deliberately 
> eclectic and famously stubborn and being part of the community is as important 
> as the software itself, I mean he named it after his wife and himself, Deb and 
> Ian, how cool is that. It was just that attitude that endeared it to it's 
> chosen community and good on them.  Slackware and Gentoo have a similar ethic.  
> And more power to them.  It wasn't until Ubuntu came along that Debian gained 
> much in the way of widespread traction.  However it was it's obsession with 
> stability that attracted the Mark. They could afford to break things because 
> they had this super stable backstop, but at the end of the day, Debian counts 
> the Ubuntu user as it's community, I would be interested to know how many more 
> developers Debian picked up in the wake of Ubuntu's popularity, I certainly 
> know quite a few.  Certainly HPs support was post Ubuntu startup 
> 
> Fedora has the benefit of age, being around a long time and focusing in the 
> corporate space is a good way to lift profile in your preferred market.   I 
> don't have any figures unfortunately but I would suspect many came from Red 
> Hat sites.
> 
> In any case, both are in fact very small in terms of the whole desktop market 
> and even in terms of all developers. 

Small in term of direct users, but they are the one with the more
contributers, and therefor, the one that are likely to survive in the
long term. And while it is not a stated goal of Mageia, I hope it is
obvious to everybody that we ( aka the founders ) forked the project
because we wanted it to survive in case of problem on Mandriva side.

If we look at the number of contributers in the overall free software
distribution community. I think that Debian and Fedora are one of the
biggest one. 

The fedora account system tell me there is 21000 members in the group of
people who signed the contributer level agreement ( CLA ), around 1100
in the packager group, around 100 in the marketing group. I suspect that
opening a account is required to edit the wiki or something like that,
hence the high number of accounts.

A quick search on debian ldap directory ( ldapsearch -x -H
ldap://db.debian.org -b ou=users,dc=debian,dc=org
'(objectClass=debianAccount)' ) tell me there is around 1650 accounts,
which roughly translate to the same number of packagers for the time
being. 

Ubuntu "only" have 650 people in the ubuntu-member group
( https://launchpad.net/~ubuntumembers ), which is a superset of the
various groups. There is 160 people who can upload to Universe ( ie,
their version of what mandriva called "contribs" )
( https://launchpad.net/~universe-contributors ), and there was 970 who
subscribed to have their packages reviewed
( https://launchpad.net/~revu-uploaders ).

When you compare to the 3000 people who committed to gnome since the
beggining ( source, gnome census of Dave Neary ), the 500 currently
active contributers of kde
( http://www.kdenews.org/2009/07/14/growth-metrics-kde-contributors ) or
the 700 who contributed to 2.6.20 ( http://lwn.net/Articles/222773/ ),
you see the number are not much different. 

So while they may be small in term of market of users, they are the one
who successfully attracted some of the biggest community of
contributers.

And attracting contributors is the key of the sustainability we should
aim.
 
> > 
> > My own opinion is that Canonical pay 5 people full time to take care of
> > the community growth
> > ( http://www.jonobacon.org/2010/07/26/the-five-horsemen/ ), and that's
> > the main reason for contribution from outsiders. 
> 
> Tsk a badly dressed marketing team  ;) 

Nope.
The team of Jono have been quite concerned with organisation ( see the
various track at UDS ), they produced some code to help on various level
( acire, python-snippet ), wrote some documentations ( for the various
community process ) and they also try to act as mediator ( see Jono
book, chapter 9 ) when there is a conflict.

This IMHO exceed the scope of a marketing team.


>  I'm not denying that marketing  to 
> bring in Code Contributors is a necessary thing and in fact we've already 
> identified this group as our initial, primary target market, however the fact 
> that Ubuntu is high profile out in the market place gives Jono and crew a hell 
> of a lot more leverage to bring in new talent.   

They do not seem to attract so much new talent, if we look at the
metrics I gave before. Or at least, they are not more successful that
Fedora or Gentoo ( back in the day when Gentoo was all the rage, some
years ago in 2005 ). Of course, they are more successful than we were in
Mandriva, so that's not bad either.


> > The same goes for
> > Fedora and Redhat
> > ( http://fedoraproject.org/wiki/CommunityArchitecture )
> 
> It's interesting that you point to that URL, I'm a big believer in the Biology 
> of Community that the Fedora guys talk about.
> 
> The principle idea behind it is that once a community reaches a critical mass 
> it becomes self sustaining, in the case of the Mageia community that would be 
> the point where you could remove all of the founders from the mix and it would 
> keep going.  
>
> To me that requires a whole community, it is a holistic beast.  Yes you can 
> continue a community that rides on the coat tails of a single person or core 
> group but is it self sustaining.  
> 
> Fedora has reached this point I think and would continue if RedHat was removed 
> from the equation. 

Time may tell us sooner than we think.

>  Would Ubuntu continue without Shuttleworth and Canonical, 
> I'm not sure, but I reckon they are a long way toward it.  OOo wasn't, but 
> LibreOffice has the opportunity to be.  Debian, I don't know the community 
> well enough to comment.

There was a point were the Debian infrastructure was almost forked some
years ago, according to a story I heard in Zurich ( but I do not have
public source ). And there was also the old rumors of a Debian fork in
2003 ( http://lists.debian.org/debian-devel/2003/08/msg00389.html ),
which may have lead or be fueled by the creation of Ubuntu at that time.


> The point is that community goes right across the spectrum of users
> Not enough of the community at the User end of the spectrum is as untenable as 
> not enough at the Makers end.  The trick is balance, that's what the Fedora 
> project has taught us

Then the balance decided by Fedora is not really in favor of people in
the User end, if we look at this interview : 
http://howsoftwareisbuilt.com/2008/12/21/interview-with-jeroen-van-meeuwen-fedora-project-vice-president-fedora-emea/

"Jeroen: One of the big, essential differences between Fedora and other
distributions is that we’d rather gain one contributor than a dozen
users. In fact, if I could lose 1000 users right now and gain a
contributor, I’d do it. It’s not up to me, but if it were, I’d do it."

-- 
Michael Scherer