Am Dienstag 26 Juni 2007 schrieb Douglas Allan Tutty: > On Sat, Jun 23, 2007 at 11:01:05AM +0200, Hans-J. Ullrich wrote: > > Oh sorry, I did not prezise my question corretly. I know that both > > conflict. This is clear for me. What I want to know is, WHY such a > > conflict happens. Why can (in my case) nexuis not access to libcurl4 and > > the other ones stay access to libcurl3 ? This was my question, as IMO > > both libs seem independent for me. > > > > On the other hand I wondered, why apt does not inhibit the deinstallation > > of the other programs or the installation of libcurl 4. Is it, because > > the philosophy says, in linux everything is allowed to be done and > > controlled by root ? > > > > My question aimed less to the technical side, but to the philosophical > > side. > > You want to install libcurl4 which conflicts with your installed > libcurl3. Lets assume that they both contain identially named files > that would overwrite each other on installation. They may not be > destined for eventual coexistance so that is not planned for in their > namespace. > > So apt will remove libcurl3. > > However, your packages A, B, and C depend on libcurl3 (which is now > removed). > > So apt will remove A, B, and C. > > Sounds like you're running unstable. Things like this should never > happen in stable. > > The maintainers for A, B, and C can't update them to work with libcurl4 > until its available. So the timeline looks like this: > > libcurl4 becomes available. > > New package D needs libcurl4. > > A, B, and C already exist and need libcurl3. > > Maintainers for A, B, and C, start to transition their packages to use > the new libcurl4. > > Here's where you're at now. > > Eventually, A, B, C, and D will all depend on libcurl4 and libcurl3 will > be obsolete. > > So philosophically, one must be philosophical about problems when > running unstable. > > Doug.
Hi Doug ! Thank you very much for this explanation. This is exactly, what I wanted to know. Yes, I am running unstable. Now I know, that this could be a normal behaviour, but mostly in unstable. Again: Thanks for the help. It explains a lot of things for me. Regards Hans -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]