The Forgotten "F": A Tale of Fedora's Foundations

[Stephen John Smoogen]

On 22 April 2014 05:53, Stephen Gallagher <sgallagh at redhat.com> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 04/21/2014 05:31 PM, Stephen John Smoogen wrote:
> >
> > On 21 April 2014 11:19, Stephen Gallagher <sgallagh at redhat.com
> > <mailto:sgallagh at redhat.com>> wrote:
> >
> > -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
> >
> > On 04/21/2014 01:08 PM, Bruno Wolff III wrote:
> >> On Mon, Apr 21, 2014 at 12:37:57 -0400, Stephen Gallagher
> >> <sgallagh at redhat.com <mailto:sgallagh at redhat.com>> wrote:
> >>>> Does Fedora need to be that gateway OS? Maybe Ubuntu would be
> >>>> a
> >> better intermediate step?
> >
> > If Fedora isn't that gateway OS, why are we bothering? What makes
> > it likely that any user would switch to us if they've entered the
> > FOSS community via Ubuntu? (Don't get me wrong, this is a question
> > we also need to answer, but I don't think it's wise of us to be
> > recommending that Ubuntu handles gathering our new users for us.)
> >
> >
> > It is an interesting question... "why are we bothering?"
> >
> > When people bother because they need to be THE gateway.. they are
> > setting themselves for a lifetime of disappointment. That ship
> > sails completely with little to no control.
> >
>
> Maybe I should have phrased that differently. "If we aren't trying to
> be that gateway, why are we bothering?". Without users, we can't grow
> our contributor pool. Without growing our contributor pool, we won't
> innovate as fast as other distributions, which in turn will further
> reduce our user and contributor base.
>
>
>
Actually you will find out that while having a healthy contributor pool is
needed, having a large contributor base will inhibit development at times
because so many people rely on old stuff that you tend towards only
conservative changes if any at all. Debian is a pretty good example of this
in action. Making medium to deep changes make our flamewars seem tame. If
you want to be the keystone for innovation, you need to focus on the people
who are looking for that which is a small segment of the population of
users.




> > I have found that if you are going to bother.. do it because it is
> > making something better for you, for something you care about. That
> > is
>
> I'm certainly not trying to rule that out (it's why I'm here after
> all). But it's not *enough* (in my opinion).
>
> > stuff you can control and not items left to the fact that people
> > choose to use what everyone else uses or by the fact its name
> > sounds exotic or they like Orange over Blue.
>
> Of course there will always be people who make frivolous choices, and
> I'm not expecting to cater to them. You're right, that way lies
> disappointment. I do think we *can* improve our appeal, though. We
> just need to agree that this is a real target and go after it.
>
>
>
The majority of people make choices due to frivolous choices. They usually
come up with some sort of rational reason afterwords but the initial choice
is 'frivolous'. [Humans are not rational creatures, we are creatures who
use rationality to justify our stupidity later.. ]



> Maybe some real ideas now instead of me just spewing platitudes? :)
>
> I've argued for quite some time that the path to code contributions
> would be best paved by making Fedora the first Linux distribution with
> a fully-integrated development environment. Take something like
> Eclipse and Red Hat Developer Toolset and build our "Microsoft Visual
> Studio" with a public API. With a basic recompile of RHDTS for Fedora,
> we can carry backwards-compatible support for three years, making it
> actually possible to do development for Fedora (and as a bonus, stuff
> that will also run on RHEL with RHDTS). I'd also love to see such an
> environment designed from the beginning to integrate well with
> OpenShift/Docker for Continuous Deployment.
>
>
Good idea, First we will need to interest the developers who want to
scratch that itch to agree (either through payment or magic) to agree to
work on one set of existing tools versus everyone building another set from
scratch. Developers seem to be a very egotistical bunch who tend to think
that they are the only ones who can do something right... and then
reimplement LISP and emacs (or Algol and vi) poorly.




-- 
Stephen J Smoogen.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fedoraproject.org/pipermail/devel/attachments/20140422/0e3bacd5/attachment.html>