[aseipp]

It's a bit of a magic trick. A "snapshot" is taken any time a command is run, and it happens implicitly before any actual algorithms or code for a given command is run (massively simplifying the internal design), so for all intents and purposes it's "automatic" from the user interface e.g. even checking repo status or otherwise small operations will cause a snapshot.

But you can integrate with https://github.com/facebook/watchman/ in order to have a truly daemon-ified option where any filesystem write will cause a snapshot to be taken.

[stouset]

What’s also amazing is this means you’re no longer completely on your own when fixing things up.

I can’t tell you how many times I got midway through a git rebase, realize I’d fucked up earlier, and had to abort and redo everything from scratch. With jj not only is this not a problem (thanks to conflict markers being first class citizens) but also any intermediate changes are being snapshotted without me having to do anything. So I can always go back to an earlier change if need be.

The first time I had to do this I’d royally fucked up a complicated commit reordering sequence while screen sharing during a meeting. I’d never needed to use the op log before and had no idea how to use it. It took maybe a minute or two to fix things completely.