[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: FC4 good new tech, bad legacy support

On Wed, 2005-06-29 at 20:09, Christofer C. Bell wrote:

> While the rest of your post is not in the same tone, I feel this sort
> of "if you want stability go somewhere else" responce disappointing
> from many members of this mailing list.  It seems there are a lot of
> folks who like to  take the "Fedora as testbed" stance to an extreme. 
> The operating system is not advertised as "unstable" and it shouldn't
> be viewed as such.  The recent selinux mistake is understandable as
> all previous patches are nominally understood to be applied before a
> new one is (and under these conditions, the selinux issue doesn't
> occur in my understanding).

>From my experience running FC4 and from reading the list there are more
than a few gotchas with the latest release.  Many of which can be
mitigated by simply reading the release notes. Others that have required
some searching on the web and in this list.   But how many people
actually read the release notes or even take the time to plan out an
upgrade to their main system?  I'm guessing not very many, and very few
of those that post asking questions here.  

Actually Centos is not really going that far away from Fedora.  The
current version is essentially FC3 but with a longer shelf life.

I realize the original poster was venting.  But it does not change the
fact that if he is looking for something that just works out of the box
without some effort on his part then selecting the latest FC release a
few weeks after it is out is not the release he should be using.  At the
very least he should work with FC3.  If you are trying to build a stable
system on which to run specific packages then you need to do some home
work.  And selecting the latest FC release a few weeks after it comes
out is asking for problems if you are not prepared to spend many hours
just trying to get basic things working.  And if he is "testing" FC4
then filing bug reports on each problem has a better chance of producing
a fix/change than venting to the list.

If the Fedora team had as a goal to put out an extremely stable release
that works across most hardware then time based releases that
incorporate as many bleeding edge options as they do now is not the way
to do it.  It takes an extreme amount of time to properly test all
packages.  From what I have seen Fedora dumps much of the testing and
bug isolation on those that adopt FC early in the release cycle.  This
benefits Red Hat as they get a lot of major issues shaken out by the

Not to say this is a bad thing.  Just that users need to expect this
going in, otherwise you end up with rants in the list complaining that
things don't work the way the user expected.

Kind of like using a hammer when you have screws to drive.  It will work
but it takes a lot more effort and the results may not be as nice as if
you had used the right tool.  FC is not the right tool in all cases. 
Hence the suggestion to use an alternative that may work better for the

Scot L. Harris
webid cfl rr com

Anthony's Law of Force:
	Don't force it; get a larger hammer. 

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]