On Mon, Jan 15, 2007 at 08:58:58PM +0100, Patrick Useldinger wrote: > So I guess I am missing something decisive here. Can anybody shed some > light on _why_ there are 2 different ways to update?
It might help to think about it as the process for keeping up with -stable being identical to the process for keeping up with -current. In both cases, you start with the latest relevant release and update it by building from source. The only difference is that in one case the starting points change only twice a year, while in the other they may change twice a day. Ideally, there would be stable as well as current snapshots. But that would require a lot more hardware and time for the developers to keep up with the last two stable releases plus the version they are actually working on now. For each platform. That doesn't seem very practical. If you have more than one machine (with the same architecture) it makes sense to build a release in one and use it for a binary upgrade on the others. Emilio (NOT a developer.)
