I wrote: > In terms of those metrics, I'm still in favor of a mostly anarchical > approach: [snip] Oh, and I forgot: * *No* QA of the type the original Fedora Legacy had, where all updates had to reviewed before even pushing them to testing and tested before pushing them to stable. I believe it was this excessive QA which killed Fedora Legacy. Instead, it should be up to whoever submits the updates to decide whether to go through testing and when to push to stable, and no QA whatsoever should be required to push an update to testing, just as things work in all of Fedora. If the security team insists on it, the security team approval system for security updates could be applied (though I consider that completely unnecessary bureaucracy, but that's a separate rant), if they don't want to be approving security updates for EOL releases, then they should just go straight to stable. Kevin Kofler -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list