Daniel Quinlan wrote: > > "Robert W. Current" <[EMAIL PROTECTED]> writes: > > > I would like to raise the issue of pitfalls in defining RPM as the > > "mandated" packaging system for the LSB. Although I will concede that > > RPM is fairly standard, it is potentially harmful to impose a packaging > > standard. > > I disagree. > > 1. We need to define a standard way for packages to be installed. If > you can't install software, you can't run it. > > 2. At this time, the best method for this is RPM. (That means: the > RPM binary format and probably a wrapper command named something > like "install-lsb-package".) > > (Everyone at our December meeting agreed with this, including the > Debian presence.)
Then, I would like to read more about how depencancy checking will be resolved on systems that primarly use other packaging methods. (eg, how does the ISV check for a library set on a system with all .deb packages when they install thier .rpm?) Is this documented somewhere yet, I haven't found it. > > Can this be avoided by an investigation a standardization effort in the > > packaging community itself? For example, can we bring representatives > > from the different packaging application developers to a single forum to > > discuss a method of sharing a common "packing accounting system" ?? > > People from Red Hat and Debian are working on something, but we're not > planning on waiting for it to be completed. If they decide to bring > that effort under the LSB, it would be okay with me, but that's their > decision. Great news to hear :-) sadly under publicised I guess. Hope more than those two are talking. If a forum is set up on that, I'd be very interested in reading about thier progress. Got a pointer/url or something? Is the table open for other packaging software developers to enter?
