You're correct. I've been using source code control systems since the days of punched cards and tape. (I even wrote a clone of SCCS in Burroughs ALGOL once, for my own edification.) However, I only have a reading knowledge of git.
It looks like I have two options when working with gerrit:
- Using an initial commit and then amending it repeatedly.
- Using multiple commits and then squashing them.
The former is pretty simple. The latter would allow me to revert to points other than the most recent one.
I guess that both indeed are features of git's distributed nature. You're free to amend history as long as you do it within the privacy of your own repository. The classic centralized/single-stream VCSs would think of that as cooking the books.
The moral of the story is Read The O'Reilly Book. (RTORB?) And then read it again . . .
Craig Jackson