Package Review Stats for the week ending January 18th, 2009

[Robert Scheck]

On Thu, 29 Jan 2009, Conrad Meyer wrote:
> In either case, the package owner gets an email summarizing changes to CVS, 
> and can revert the CVS change. If the newbie was malevolant and/or the problem 
> persists, they can be kicked out of Fedora. If not, they learned their lesson 
> (similar to Wikipedia's "please go play in the sandbox, kid").

We usually make a Package Review BEFORE a package goes into Fedora. Why
should somebody be allowed to change something at a package being in Fedora
without having it reviewed BEFORE the CVE commit by somebody else (in this
case by the package maintainer)? If we drive this way, we seriously do not
really need a package review, we can fix/correct stuff always afterwards.
A very interesting thought, you've brought me to. This even would solve my
claims regarding the Merge Reviews. And Wikipedia also works this way - so
everybody can do everything (also create new pages, which equals to Fedora
CVS imports) and if it's considered harmful/low quality, it's handled then
afterwards. What is preventing us from letting initial CVS branchings every
packager do on itself and performing no longer any package review in front
of the CVS branching/import?

This is a real thought, no kidding or irony. That would make us just more
open as we currently are. Yes, I know you don't believe me that, after all
of my e-mails now, but I really have spent some time to think about that.


Greetings,
  Robert