On Sun, 26 Aug 2012 15:15:04 +0100 Colin Guthrie wrote: > Obsoletes in packages which genuinely replace the old one seem > reasonable and uncontroversial.
It is both reasonable and uncontroversial because the User is given the choice. With obsoletes in packages Before any installation or package removal via urpmi/rpmdrake 1) user is informed that rpm A must be removed for other rpms to be updated 2) user is given the option y/n before the installation/removal proceeds. If using task-obsolete 1) user does not know Which or When rpm/s will be removed 2) user does not know Why the rpm must be removed 3) user has no choice as this is all occurring in secret Charles -- // Minor lesson: don't fuck about with something you don't fully understand -- the dosdoom source code ---------------------- Mageia release 3 (Cauldron) for x86_64$ On SuperSize....http://www.eslrahc.com Registered Linux user #182463 3.5.2-tmb-server-1.mga3 x86_64 ----------------------
signature.asc
Description: PGP signature