On Thu, 1 Oct 2009, Brian Mathis wrote: > On Thu, Oct 1, 2009 at 3:45 PM, R P Herrold <herrold at centos.org> wrote: >> herrold: >> because creating a problem and fixing it ex post is harder >> than not creating it in the first place > Spam issues aside, that is the very concept of Wikipedia and other > wikis, and also for all modern VCS tools, and most of them have proven > that line of thinking really doesn't hold up. That's clearly one opinion but not stats based that I can see -- it is not observationally true that I can see -- As a counter example, recall a wiki entry I saw on the Wikipedia declaring Mike Harris [the upstream's long time X maintainer] to be the 42nd Emperor of Ontario snip more theory > ... As soon as it's open, you'll have more people > monitoring and more people who can fix errors as they are > introduced. I'm one of the ones reading all commits and doing rough cut triage, and the theory you put forth is not the reality I see, even with the present 'find and ask model' Check out the created and abandoned xen articles in our wiki -- who cares enough to stop writing new pages and complete fixups? The ML pre-vetting is a talkers debating society by and large, from what I see in commits. ;) As I suggest: If it is not core variances, it goes to 'projects' wiki, and I'll ignore it and whatever else the cat drags in (with an appropriate subscription rule) so long as it has a warning sign for the unwary my $0.02 - Russ herrold