15
chenb0x
48d

What the FUCK is the point of submitting a PR, if you're going to approve my code, without looking at it, and then LEAVE ME to further refactoring.

I don't mind the refactoring. At. All.

What I DO mind, is being told "yerp, looks good" and then standing aside as I break everything.

TeAmWoRk

Comments
  • 4
    That's a bit like how it is where I am.

    We all submit PRs but really all we're doing is just using them to not merge all at the same time and break things all at once.

    I look over the PRs sent to me, but pretty for the most part ... nobody else does.

    It's shoot from the hip / cowboy town...
  • 4
    At my last job, the boss would glance over my code, often taking only a few seconds for a few thousand line diff, and say “yep, that looks good! Better than I could write.” And push it to prod.

    It always annoyed me so much. I can’t improve if I don’t know what I could do better! And yeah, he was more junior than me, but a second set of eyes are always better.

    I still miss the job, I really do, but I don’t miss that part.
  • 2
    Review workflows are important.

    Sadly mostly ignored.

    Reviewing can prevent Bus factors of 1, since - when done right - another Person at least knows what happens and when.

    Reviewing can prevent bugs... But it's also very useful for learning from each other.
  • 0
    Now you know you need better tests / CI
  • 1
    Kinda guilty of the charges.
    I have to review PRs which implement whole new features into a game and sometimes reading hundreds of lines I fail to see the big picture when reviewing and end up acting as a slightly smarter compiler/IDE: "Hey you did not unlock this mutex in the for loop", "You meant to call glUniformSetf instead of Seti" and basic shit like that limited to each code chunk separately.
    Most PRs I review end up with a simple approval since the one dev I review is a 10x one who doesn't fuck up ever.
Add Comment