Jari Aalto <jari.aa...@cante.net> writes: > Russ Allbery <r...@debian.org> writes:
>> That doesn't make any sense to me. Why would we replace a tag which is >> accurate with one that isn't? Why is an incorrect dependency on >= 7.1 >> more descriptive to the reader than a correct dependency on >= 7.0.50~? > I was more thinking how this message may affect potential readers. > Reading the "accurate" lintian message, user may think: > "Okay I write this to shut up lintian": > Build-Depends: debhelper (>= 7.0.50~) > Where a better[1] and practical alternative would have been: > Build-Depends: debhelper (>= 7.1) > ...because Debheler is in 7.4.x in testing, so it's unlikely that the > version 7.0.50~ is ever used. debhelper (7.0.50) unstable; urgency=low * This release is designed to be easily backportable to stable, to support the new style of rules file that I expect many packages will use. backports.org I think now has an even newer version, but nonetheless 7.0.50 was the release done specifically for this support and hence is the correct dependency when one wants to use overrides but doesn't need any newer version. > [1] (In readable sense; it's short, no funny characters, no long version > string to ponder about. I guess we just disagree about this. (>= 7.0.50~) is clearly better in my opinion. If I saw a dependency on (>= 7.1), I would look at the debhelper changelog and try to figure out why this package cared about the rework of command line option handling. -- Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org