On Sun, Jul 14, 2019 at 09:21:37AM -0400, Theodore Ts'o wrote: > but as saying goes, "patches gratefully accepted". Whining for developers > to do extra work via Debian Policy is, well, not.
If patches are actually accepted, that's fine. But this Policy item currently servers as a tool to _sometimes_ get a patch through. In some cases such patches get stonewalled, and if you _then_ NMU a package in question, you can get calls for getting demoted from DD. We released Buster basically useless in GUI mode in !systemd, despite a solution been discussed[1] and _implemented_ in January 2018, with multiple later submissions. And it's not that they haven't been noticed -- the existence of that patchset was used to justify removal of (unlamented) shim. So even that on-paper severity is not enough to get patches accepted. Without it, regressions would be even worse. Meow! [1]. Only one member of the systemd team was helpful, so this can't be called consensus on their part -- but not for a lack of trying. -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢰⠒⠀⣿⡁ ⢿⡄⠘⠷⠚⠋⠀ NADIE anticipa la inquisición de españa! ⠈⠳⣄⠀⠀⠀⠀
