Santiago Vila <sanv...@unex.es> writes: > In this case, however, I fail to see the rationale for actually > *dropping* the extra priority, other than "it's not useful for me". > Well, it may be useless for you but it's still useful for me.
Basically, the rationale is simplification and reducing the cognitive load of people packaging software for Debian. I've taught a bunch of people how to package software for Debian, and the "extra" vs. "optional" split does not make much sense to them. Debian packaging has a reputation for being a little nit-picky and tedious, so in general it seems good to remove things that add another decision point if that decision point isn't particularly useful or isn't used consistently. An alternative would be to say that all software that doesn't warrant standard or higher should be packaged with priority optional *unless* it conflicts with something, in which case it should be priority extra. That at least would be simple and straightforward and would be easier for new maintainers to reason about. It's not what we're doing right now, though; instead, "rarely used" software has some random mix of optional and extra priorities, based on the whim of the maintainer, and there are certain classes of packages that are almost always extra (debug symbols, for example). This is all hard to explain. I think many of the issues here would be served by creating a much simpler rule instead of eliminating extra entirely. If conflicts are really the reason why we care about this priority, then let's define extra priority to be all about conflicts and simplify. -- Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to debian-policy-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/87vbmdydmx....@hope.eyrie.org
