gtk+ stuff moving out of Core

[Jeff Spaleta]

On 4/5/06, Christian.Iseli at licr.org <Christian.Iseli at licr.org> wrote:
> I'm not sure I like this idea.  What would be the point ?
> What worries me a bit is that once a binary package is in FE, there will be
> little incentive to review and push the update.

I have a suggested fix for that. Have people raise their hand now and
commit to doing the review by a certain point in time. Start the clock
and if the review isn't completed within the time frame of that clock
and new binaries pushed... the existing binaries that were imported
from Core are removed breaking the dep chains. We can do this sort of
thing now before the next testing cycle even begins and allow for as
smooth a transition as possible.  Things are much more difficult to
deal with smoothly if things are dropped from Core late in the testing
phase.. cough gstreamer08 cough.

The idea here is to allow a "good faith" effort for someone to take
"proactive" responsibility to clean this package up which minimizes
the impact on the package repository. Good faith meaning that a
package maintainer and a reviewer(most likely someone who needs this
package as a dep) raise their hands now and say "yes we will work on
getting this compliant in X number of days".  If people don't raise
their hand to take the responsbility then no.. its not appropriate to
bend the rules and let the binaries in from Core.  If people sign up
now and say they are going to be responsible for pushing this through
review within the stated timeframe then we can publicly shame them
when they fail. I think its an appropriate tradeoff to let the
binaries in now if people take responsibility upfront to work on the
review... because I like giving people enough rope to hang themselves
for public amusement.

-jef