plans for long term support releases?

Mike A. Harris mharris at mharris.ca
Wed Jan 17 03:03:54 UTC 2007


Thomas M Steenholdt wrote:

> This is why I suggested an more stable (in terms of changes) LTS spin, 
> perhaps, for every 2-4 normal Fedora releases, to provide a Fedora that 
> could actually be used in these situations.

There is already.  It's called "Red Hat Enterprise Linux".  There are
alternatives as well based upon rebuilding the SRPMS from RHEL.  The
only difference is that they do not have the "Fedora" brand name tacked
onto them and they're physically built on computers elsewhere.  To the
best of my understanding, all the rebuilding work is also done by people
whom contribute directly or indirectly to the Fedora Project in an
ongoing basis in some manner.

So, what people are asking for _already_ exists, just under a different
name.  To duplicate _their_ existing effort would be a waste of
resources to whoever was doing the work.

But then that's the problem isn't it...  Nobody _wants_ to do the work,
rather they want _someone_else_ to do the work.  But those someone
elses that could do the work - are already doing the work, and naming it
CentOS, or one of the other similar projects out there.

Of course anyone out there with a strong desire to have an actual
Fedora Core release "maintained" (not supported - Fedora is unsupported)
for n years, can always go right ahead and start doing just that.
It would appear such a person does not exist however.

So, it appears that there are only people who want long term Fedora
maintenance out there, but zero, or close to zero people with enough
motivation and/or desire to actually do it.

It doesn't come as much of a surprise however, at least not to me.  What
motivation does someone _have_ to start such a project?  We could
certainly theorize about that, and I have my own theories of potential
reasons why someone could in theory want to do that.  In particular the
"I'll do it to scratch a personal itch, because I want to use it myself"
theory is likely the most common one why people get involved in such
things.  For small projects that are manageable by a single person, this
theory works very well, and is probably the main factor responsible for
the plethora of OSS in existance in the first place.

As the amount of work a potential project requires to be successful
increases however, the likelyhood of a single person working alone on
it being successful doing so decreases at an inversely proportional
rate.  A large enough project more or less dictates that multiple
people working together in an organized fashion as a team toward a
common goal is required, and that the motivations of each individual
are the same or at least similar enough for them to come together and
do something.

I say the motivation for a project like this _has_ to come directly
from the population who wants to use it.  There could be some minor
exceptions to this rule, but any exceptions don't break the rule.
(ie: someone willing to kindly maintain a few packages they don't
even use is a possibility, motivated simply to help out.)

So while we often can witness many people stating how much they
desire to _have_ a specific Fedora release receive long term
maintenance updates, we almost _never_ see anyone actually want
to do all of the work on their own, and we almost never see anyone
actually put effort into trying to start off such a project and
gather and lead others whom are like minded.

The Fedora Legacy project is the closest anyone has gotten, and while
it's had a shaky life, it's done some good stuff too.

Another factor to consider, after considering all of the above, is that
should anyone out there be motivated enough to actually grab the bull
by the horns, and to try to gather like minded individuals, they also
need to be prepared to handle the plethora of community oriented
discussion and feedback they'll almost certainly receive, both from
like-minded individuals, and people with very different opinions.
A successful project leader could expect a lot of public
discussion/debate/flamewars/etc. both from the community at large,
as well as contributors and potential contributors.  (I've yet
to see a project with more than a small number of people ever agree
on decisions about even the most minor things the first time around.)
<grin>

Then, once things get going, you have to realize that the vast
majority of users out there to which you are likely to receive any
feedback from, are ones who don't like something.  Expect lots of
negative feedback, thankless emails from people with attitudes, etc.
and be prepared to take it all with a coke and a smile, and to
continually focus on why you did all of this in the first place.
Every decision you make, someone out there will be displeased with,
or would have done it another way for one reason or another.  Expect
them to tell you this.

Oh, you're likely to get good feedback from friendly users who love
what you're doing and are eternally grateful too.  Those people will
make your day on the rare occurances they shine through the massive
wall of negative energy surrounding every move you make.

So why aren't people jumping up and down to go ahead and do this
now?  Volunteer to maintain Fedora for 4 years on your own for the
community, then run out and get a good therapist and some prescription
medication.  Everything will be alright, I promise. ;o)




More information about the fedora-devel-list mailing list