On 21/06/2009, James Carlson <carlsonj at workingcode.com> wrote: > Peter Tribble writes: > > On Sun, Jun 21, 2009 at 1:32 AM, <David.Comay at sun.com> wrote: > > > > > > Sorry for missing this the first time but what is the justification for > > > moving any utilities into SUNWcsu? SUNWcsu needs to be broken up into > > > separate packages and not have additional content added into it. > > > > Really? We already have far too many itty-bitty packages to manage a system > > properly. The need is for fewer well-defined and well-bounded packages, not > > continual fragmentation. > > > "Itty-bitty" packages shouldn't be a problem as long as the package > management software allows you to deal effectively in groups of > packages. The big unfocussed packages like SUNWcsu, on the other > hand, are an unsolvable problem for people trying to build smaller > systems. > > As for moving things out of SUNWesu, I don't get it. That package (at > least for System V) is part of SUNWCmreq, meaning that it's installed > everywhere and that you can always rely on it being present. Moving > bits out of it and into SUNWcsu makes no sense, at least from a > packaging perspective, and it's puzzling what it could have to do with > ksh93. > > If anything, moving bits out of *both* SUNWesu and SUNWcsu and into a > clearer-headed "SUNWksh93" (or some such) would make more sense.
Which sense does it make to have a separate package which makes a machine unbootable if you pkgrm it? -- Cedric Blancher <cedric.blancher at googlemail.com> Institute Pasteur
