>>>>> "Norbert" == Norbert Preining <norb...@preining.info> writes:
Norbert> Hi Sam, surprising statements from you ... Norbert> On Wed, 22 May 2019, Sam Hartman wrote: >> The same is true of package maintainership though. We sometimes >> do change the maintainership because we're unhappy with how >> someone maintains their packages. That rarely uses the formal >> policy that goes Norbert> ??? This seems to be new - at least when I became DD some Norbert> 10+ years ago this was not the case, and it was completely Norbert> out of discussion to do this. I'm reasonably sure there are situations over the years where we as a community have concluded that a package highjack was acceptable. I might be wrong. I'm quite confident there are cases where someone has started NMUing a package because a maintainer is inactive and has eventually declared themselves the maintainer without following the letter of documented practice. Norbert> Why would we need "package salvaging" (thanks Paul for Norbert> that!) Norbert> https://www.debian.org/doc/manuals/developers-reference/ch05.en.html#package-salvaging Norbert> if we can change package maintainership just like that? Because "just like that" involves a lot of careful thought, sometimes a flamewar, and sometimes long discussions of whether something is the right answer. When we've done something enough that it's worth writing down a right answer ahead of time to shortcircuit discussions, we sometimes do. Package salvaging is in my mind one of those cases. Norbert> I will remember your statement the next time I consider Norbert> another maintainers packaging efforts insufficient. OK. but let's make sure you understand what I'm saying fully. I'm saying that as a DD you have the technical capability to change the maintainership of any package. If you do that outside of the written procedures you should be prepared to defend your actions and suffer consequences if the community disagrees with you. Imagine that you write to d-devel, propose some action and get a fair bit of support. But you didn't quite wait long enough or the maintainer pops up just as you do the upload or something. It's likely the only consequence will be that we might conclude as a community the best option is to revert your action. If you don't ask for input and go off wildly on your own it's likely the consequences will be significant. My point is that as a community we don't typically jump at "you broke this rule, bad!" We typically also think about whether enough good was served to justify breaking the rule and whether you might have found a case where the rule was poorly crafted. >> As a matter of technical capability we can all do a bunch of >> arbitrary things. As a matter of practice we sometimes do things >> that according to written policies and procedures seem kind of >> arbitrary. And if Norbert> I am not sure what you mean with *we*, but I am sure that Norbert> most "normal" DD are not allowed to overstep the rules that Norbert> easily. I don't think it is easy at all. Going and doing something and then waiting to see whether the community agrees with you that unusual action is justified doesn't seem easy to me. My point is that the planet admins are taking a position that seems fairly consistent to me with the same position we take for package maintainership. Normally, you follow the written rules. Sometimes there are exceptions. If you act on what you believe is one of the exceptions, then the community's trust in your actions will be re-evaluated based on whether the community agrees with what you did. --Sam