Hello Felix, Am Tue, Mar 14, 2023 at 09:37:23AM -0700 schrieb Felix Lechner via Development of GNU Guix and the GNU System distribution.: > With the core-updates process now abandoned, I retitled the issue to > "[PATCH wip-inetutils 0/4] gnu: inetutils: Update to 2.4".
apologies for not replying earlier - this was because I did not have the answer... I think this is a place where the feature branch idea is not yet fully thought out. It is quite clear how branches closer to the leaves could work (language branches, a qt/kde branch, such things); and probably also "core things". There I see two kinds of cores: First, things that are very close to Guile/Guix (switching to gexps, rewriting the daemon in Guile, these kind of things). And maybe something more like "bootstrap" and "tool- chain" (changing the Guile version, updating gcc/mpfr/mpc, probably rust, and so on). I do not yet have a clear picture for things on a "middle level", but which cause a lot of rebuilds. Which kind of team would be responsible for inetutils? Suggestions would be very welcome! Something completely different: I am not sure whether inetutils should have so many dependencies. It only contains binaries and no libraries. For instance, it is sometimes used as native input for tests, but tests do not have access to the inet anyway. So maybe it would make sense to identify the "root" where this package is pulled in, drop the input and modify the test suite? Andreas