[ddtaylor]

Many of these problems can be avoided by using a pull-request style workflow.

[aflag]

I don't see how prs would help solving any of those. It's common for me to use a few of these commands before I open the PR.

[Sammi]

They stop you from pushing anything broken to master. So you only have to worry about being in a broken state locally on your machine, not about accidentally breaking master for everybody.

Ideally your code review system should be the only one to have merge rights to master. Then nobody can singularly break master at least.

[reacweb]

Yes, there is a balance between preventing problem or fixing problem. It is useful be prevent problems, but it remains useful to be able to fix them easily.

[dahart]

How? I don’t see a single issue in the post that a PR would avoid. All the problems listed happen before you’re ready to push or present to others.