Bryan Allen wrote: > +------------------------------------------------------------------------------ > | On 2009-06-06 12:37:41, Shawn Walker wrote: > | > | Peter Tribble wrote: > | >It's essential that the user be able to force an override of the > | >dependency checking, > | > | No, it is not essential. > | > | There are numerous mechanisms already available to users to either > | correct or deal with broken packages. > | > | Ultimately, it is not reasonable to expect a package management system > | to be able to correctly manage broken packages. > | > | I believe that --force options are ultimate a cop-out; they fail to deal > | with the real, underlying issues. > > Dealing with the "real, underlying issues" is small consolation at 3AM (or > worse, in the middle of your business day) when you are trying to fix a > problem on a production system where a broken package has hosed a service, or > the system itself. > > Ivory tower metholody means very little in the trenches. > > --force is not a cop-out; it is a tool, and a useful one at that, when you are > under pressure to fix something.
It isn't ivory tower at all; I have been in the exact same position. That "solution" that let me get by at 3am in the morning usually came back to haunt me the next time I had to upgrade some other component on the system with a security fix, etc. Nevermind the number of times I had to go back to untangle the mess someone had created on a RHEL system because they had used --force because they didn't want to bother with the dependencies correctly. No thanks. -- Shawn Walker
