Hi Eric,
> Ralf just pointed this out to me:
> http://producingoss.com/en/producingoss.html#territoriality
This view values "egalitarian spirit" and argues against written credits
in files for contributors. My opinion is different.
> Would anyone object if I removed most (if not all) 95 instances of my
> name from those files
You should be able to feel reward by your name in a file that you wrote.
> given the argument that 'git shortlog file' will
> still give a pretty good indication of who the original author was?
Well, neither of the git tools is perfect for determining who wrote most
of a file:
- "git shortlog" falls short when there is a file rename (e.g.
lib/wchar_.in -> lib/wchar.in.h) and cannot evaluate a big commit
against multiple small commits.
- "git annotate" falls short when there is a reindentation of a large
block.
I wish there were more unit tests in gnulib. I'd welcome any incentive
to write new unit tests.
Bruno
