Hacker News Reader

Comment by SoftTalker | Hacker News Reader

All I know of git is clone, pull, commit, and push. That’s all I’ve ever needed. Git can be that simple. Of course it supports more complex uses, which you can learn if you need them.

View on HN

Life is simple until we play with time, once you start altering the past everything falls apart

This... seems a bit deeper than a comment about just Git.

It's not possible to alter commits with git. They are immutable. You can only make new commits that are based on existing ones in some way. The distinction is important, because the original commit (such as a commit pre-rebase) still exists after it's been "changed".

> The distinction is important, because the original commit (such as a commit pre-rebase) still exists after it's been "changed".

The original commit will only exist for as long as git deigns not to garbage collect it.

The use case for more is managing the history of it in a specific way to make it easier to understand. Most applications don't need this level of management since most applications don't have lives depending on them

I'd add Stash, unless I'm working alone

Yeah even that has caused me trouble. I usually just copy the problem file to another directory, checkout the file from the repo, pull, and then diff/merge my changes. It may sound like more work, and maybe it is, but I understand what's happening and it's rare enough for me that to do it "the git way" I'd have to look it up every time.

And that's the only reasonable subset of git there is - I agree. Branches and cherry picks aren't worth the insurmountable headaches they come with

Reply on news.ycombinator.com