3) The back and forth continues and links and answers are mostly ignored because, hey, it is a bug!
4) After several messages, finally it becomes clear the unexpected behavior is being provoked by the user doing something wrong according to the docs posted in the first answer.
5) The problem is finally solved, but at an unnecessarily high inconvenience for all involved, and the user is neither happy, because no one likes to be proved wrong in public.
Perhaps this is a corollary of the Dunning-Kruger effect and it cannot be avoided, but my reflection is:
a) No matter how intelligent you considers yourself, don't discard a priori the possibility of you doing something wrong when you are experienced unexpected results
b) Read the FAQs, the relevant documentation in the wiki, search the forum for similar experiences and, especially, be open to a)
c) Once you are invested in the "bug theory" and you make it public, the process will be more painful for all participants, included you. So do your due diligence first.
Hey, are you saying GC doesn't have bugs or we should not report them?
No, of course not, I am saying not all unexpected behaviors are bugs, and treating them as such beforehand doesn't help you or the project.
Sorry for the extension, and thanks for the collaboration.