> I think the main problem here is that the taskforce was attempting to > create a standard for packaging. What really needs to happen here is > for the Linux Community (and perhaps even other groups like Sun, HP, > IBM, HP, Apple, and *BSD) to come up with a BETTER packaging solution > than what we have.
Been done, and failed. IEEE 1387.2-1995 was such an effort. Someone told me the standard is now formally withdrawn. My view from very far outside that effort was that it pretty much was a best-practices effort, taking the main concepts from the HP SDU (swpackage) and SVR4/Solaris pkgadd systems and attempting to answer the problems that had arisen in practical use with both systems (two of which were patches and repairs), with somewhat of a nod to a few other approaches. I'm sure someone who was involved will have plenty of war stories and reasons why it went nowhere, if any of the participants happens to be listening here. My suspicion is that it was a solution not driven by sufficient motivation to actually implement - i.e., $$$ Much as the sysadmins of the world wish it might be otherwise, the whole planned POSIX sysadmin series seems to have fallen into disrepair, and sysadmin remains a roll-your-own frontier. One wonders if it's worth trying again?
