On Fri, 24 May 2019, Segher Boessenkool wrote: > IMO the best we can do is use what we already have: what CVS or SVN used > as the committer identity. *That* info is *correct* at least.
CVS and SVN have a local identity. git has a global identity. I consider it simply *incorrect* to put a local identity in a git committer or author - I think this is just like any other data format conversion necessarily involved in a repository conversion. There's an argument that (real name from /etc/passwd on the repository system at the time of commit) plus (username @ hostname of repository system at the time of commit) corresponds most accurately to the old local identities, but we don't have the contemporaneous /etc/passwd and I don't see doing something like that as an improvement on using whatever each person's preferred identity for git commits is (or some name and email address we've found, in the absence of an expressed preference from a given committer), if there wasn't a ChangeLog entry added in that commit. -- Joseph S. Myers jos...@codesourcery.com
