I use rebase, for two reasons:

1. (major): it lets me commit at my pace, and then publish my work at the pace the team prefers (which is a combination of what the team wants and what won't choke the CI, but that's a story for another time). The ratio of my local commits to commits sent to code review is often 5-10:1.

2. (minor): because I hate to see a Git history that's over 50% made of "Merged X, Y into develop". It's pure noise. I guess there's probably a command line flag to filter them out, though.

That's because they know how to use git so that they don't have to pick a child when they bisect.

My point on my now flagged and dead?? post was that people think git history is linear, so they believe they have to do special stuff so it's a 'good' linear history when this is pointless.