Cyril Brulebois writes ("Bug#1124226: dgit: commits created for previous
uploads should be created in the name of the current dgit user"):
> Ian Jackson <[email protected]> (2025-12-28):
> > Firstly, I should say that I'm sorry to Cyril for this.
>
> Thanks, I know it's not a huge deal, but that's not the first time I go
> “wait a minute, I didn't do this?!” and I thought I'd ask “officially”…
Thanks for doing so. We welcome feedback (and we may not have the
effort to make every improvement) so our bug list looks quite long.
We are keen that our software's behaviour should be proper, and that
it shouldn't make people annoyed.
> A separate discussion reminded me of Signed-off-by that can be used (as
> far as I understand/use it) to take responsibility for a given commit.
> But I'm sure you're well more versed in such topics than I am.
You want to be very careful with Signed-off-by. Its presence is a
legal declaration. In a project containing this file
https://browse.dgit.debian.org/dgit.git/tree/DEVELOPER-CERTIFICATE
it is a quite specific declaration about the legal status of the
contribution, by the named person. In trees which do not have
DEVELOPER-CERTIFICATE or similar, its use is probably a mistake.
Certainly Signed-off-by ought not to be synthesised without
authorisation from the human named in it.
This is often misunderstood, so I thought I would clarify.
> Thanks for being open to altering the current behavior, I'll happily
> take whatever adjustments you'll deem appropriate.
Thanks. We'll take you off the CC list as we discuss the details.
Feel free to subscribe to the bug and/or say you'd like to be more
involved, of course.
Regards,
Ian.
--
Ian Jackson <[email protected]> These opinions are my own.
Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk,
that is a private address which bypasses my fierce spamfilter.
