Ivan Sagalaev (isagalaev@mastodon.social)'s status on Sunday, 09-Jan-2022 21:01:57 UTC

@underlap what I'm fascinated about is that its model theoretically should not have the problem of git where commits only make sense in the timeline of a branch, so you have the classic pain of maintaining slightly divergent branches (like prod and staging) where sometimes the same change would have a conflict in one but not the other. In pijul a patch in multiple branches has the same identity, and conflict resolution is *another* patch, local to a branch.