On Wed, Aug 17, 2011 at 12:26 PM, Markos Chandras <firstname.lastname@example.org> wrote:
> Personally, I want to shrink portage. There is no way for 250 listed
> developers ( I would be glad if 100 of us were really active ) to
> maintain thousands of ebuilds. I have already said that many many times
> and I will say it again. We need to stop pretending that everything is
> fine. We need to support only the packages that we can *really* support
> and lets hope that more people will join in when they see their packages
> going away.
I think this depends on your definition of "support." If a package is
stale but works fine, I'd prefer to just leave it alone. Obviously a
package that is broken or has security bugs should be removed if
nobody steps up.
Think about the typical use case. The typical user has probably 10-30
applications they REALLY care about, and if they were two years old
they'd be upset.
However, the typical user probably has 100-200 packages installed on
their system, usually as dependencies, or because they do trivial
things like display the time in the corner of the screen. Users need
these packages and the functionality they bring, but nobody really
cares if they're running a 5-year-old version of agetty, slim, or
libX11 unless it has a security bug or simply doesn't work with the
software they care about.
So, I think it is in our interests to leverage the work that has been
done in the past. If that version of slim stops working with the
latest xfce release, chances are somebody will fix it. If agetty has
a security hole, somebody will fix it, and so on. However, merely
ditching software simply because it is old reduces the usability of
Gentoo without really getting us anything. Those packages cost
NOTHING to maintain, since nobody is maintaining them anyway.
If we simply drop lots of packages, I don't think users will volunteer
in droves to become devs and fix them. More likely, they'll just go
elsewhere, and we'll actually end up with fewer devs.
I think the average level of quality in Gentoo is adequate. There are
a few problem spots that crop up and should be dealt with. However,
packages that are simply older than upstream aren't automatically a