re changing from vista

[Charlie Kester]

* Jeremy Chadwick <koitsu at FreeBSD.org> [2008-11-14 14:56:26 -0800]:

><opinion> But why are we interested in "converting" people?  That
>borders on religious, which an operating system should not be.
>
>
>I'm not saying "we don't need new users" -- I'm saying: if we took half
>the energy used "converting" people and applied it to fixing bugs and
>improving FreeBSD, there wouldn't be a need to "convert".  "Build it
>(and secure/stabilise it) and they will come".

Indeed, what IS the value of more users to a volunteer project like
FreeBSD?

Microsoft, Apple, etc. want more users on their OS because it increases
their profits.  But who gets more money if ten thousand users switch to
FreeBSD?  

FreeBSD already has a large enough user base to attract the attention of
developers deciding which platforms to target with their apps.  But even
if it didn't, it has a large developer community of its own, and they've
done a great job porting apps, as well as creating new apps themselves.  
New users who are also developers can contribute to this effort,
so it makes sense to actively recruit them.

But why should we want to increase the number of ordinary, non-developer
users?  If these new users also contribute to the project, by working on
documentation or other non-programming tasks, then it makes sense to
actively recruit them too.

Perhaps there's an implicit calculation that only x percent of new users
will actually contribute to the project, so if you want/need C new
contributors, you should aim to recruit N = C / x new users.

Some of the comments in this thread have expressed one of the problems
new users can bring: an expectation and demand that things work the way
they used to on their old OS.  People who voice these concerns want to
preserve the Unix philosophy and culture, so they don't welcome
immigrants who refuse to assimilate.  They don't see those immigrants as
potential contributors to the project; they see them as people who want
to replace it with a different project altogether.

...which perhaps explains why some people want to impose something like
a Unix citizenship test.

Users can also contribute by helping to refine the requirements for
software.  For example, my son is an animator and he and I have often
discussed various graphics tools.  In his opinion, the Gimp is a
powerful tool which provides almost every tool or technique an artist
might want, but it's unusable because its user interface doesn't reflect
the way artists actually do their work.  He says this isn't just that
they're used to Photoshop or whatever; there's something about the
nature of the task that the Gimp fails to accommodate in a natural,
effortless way.  He says the Gimp feels like a tool designed by software
engineers rather than artists. 

We need users like that, who aren't developers but who are experts in
their own domain.  How much of FreeBSD's strength as a server derives
from the fact that so many of its users have been sysadmins with a keen
awareness of the day-to-day problems in that domain? (It's also been an
important fact that many of them are developers too.)

So when new users appear and start requesting changes to make things
more like the system they came from, we shouldn't automatically classify
them as "unassimilable immigrants".  We should try to understand what
they're really looking for, and whether or how our current software
supports it.  

It's especially important to understand why they left their old home.
What was the need that inspired them to consider a change?  How did
their old OS fail to meet that need?

Sometimes our answer to them is going to be, "No, sorry, our project
isn't designed to do that" or "That isn't one of our project's goals."
Maybe you should consider Project Y instead."  There's nothing wrong
with that kind of answer.  It's coheres with the Unix philosophy of
clarity of purpose (e.g., tools that do one thing and do it well.)

So, in conclusion, we DON'T need new users because growing the userbase
is good in itself. Sometimes growth is cancerous, and kills the body.
We DO need new users insofar as they help us meet the goals of our
project. 

(And sometimes new users suggest new goals for us to pursue.)

-- Charlie