Where Fedora Went Wrong (nice conclusion)

[Michael Wiktowy]

On 5/15/07, Valent Turkovic <valent.turkovic at gmail.com> wrote:
> http://www.oreillynet.com/linux/blog/2007/03/where_fedora_went_wrong.html
>
> What are your comments? I feel simillar.

Everything benefits from more testing. I see a lot more focus on
making testing of Fedora 7 easier in piece-wise way this release than
any other. Some examples being F7 Boot Test CD images, periodic
rawhide LiveCDs being built and made public, etc. What I would like to
see is an improvement in the testing methodology, tools and
infrastructure available. Smolt is a nice new tool for testers to use
but doesn't seem to be marketed that way. The existing sysreport and
the upcoming sos are other tools that I just read about that are
flying under the radar and could probably see more usage in Fedora
bugzilla reports. Some guides on an acceptable period of time for
bugzilla entries to go completely uncommented on are needed. Maybe
there is a good wiki page that I am missing that outlines the steps
that a tester should go through to make sure a released is as polished
as possible.

Another note: I use both Fedora and Ubuntu. The groups are *not*
mutually exclusive. IMHO, they both have their strengths and
weaknesses and neither are appropriate for 100% of the population
(although taken both together likely have the coverage). I hate this
you-are-either-with-us-or-against-us mentality. Every distro has the
same upstream heritage. Both could learn a few things from each other.
I worry a lot less putting a Fedora box on a public IP because it has
much stronger security in place with things like SELinux and I worry a
lot less about giving Ubuntu to my sister because it is assuming that
there is not a system administrator in the loop to sort out the little
things (although I have had to sort out plenty of little things).

/Mike