Security updates for Firefox 4 in F-15

[Mark Bidewell]

>
> For me the most important benefit is OS independent software, especially
> web browser.
> --
> devel mailing list
> devel at lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
>

Debates like this expose a weakness in packaging philosophy.  The
current philosophy seems to be all packages are equal.  An update to
Open/Libre Office is handled the same as an update to the kernel or
glibc.  It seems like in mainstream Linux distros your options often
are:

1) Wait 6 months for new software.
2) Download / build from source and deal with system integration issues.
3) Download unofficial package.

The reality is that not all packages are equal, and some could/should
be pushed faster than others.  Firefox and Libreoffice are good
examples of this.  If parallel installs of Xulrunner are needed, in my
opinion, so be it.

Ultimately, packaging philosophy could hold back/prevent mainstream
(i.e. non power user) Linux adoption.  These users will ask themselves
"I can get the latest version well integrated on release day using
Windows or Mac, why not Linux?".  Ubuntu is moving (once again in my
opinion) in the correct direction with its combination of LTS (a
stable base) and PPAs for more rapid updates of things like Firefox.
The fact that they have LTS tagging also prevents the question of "How
many versions to support?".  The answer is clear:  two, the current
LTS and the latest 6 month release.

Finally, with packaging cycles changing might it be time to revisit
the decision to merge Fedora Core and Extras?  Creating separate repos
with different goals might be wise.

I apologize for the long rant.

-- 
Mark Bidewell
http://www.linkedin.com/in/markbidewell