Code Review Confusion

I struggled to come up with a banger opening for this story, so I asked ChatGPT for ideas. Let me tell you: based on the response, ChatGPT is also clearly confused about what “code review” actually means.

Wikipedia calls code review a “software quality assurance activity,” which only adds to the confusion. Because “software quality” means different things to different people. Here are some common misconceptions I’ve heard about code reviews in ABAP:

• “It’s to make sure there are no bugs”. Nope, that’s what testing and unit tests are for.

• “It’s to check that code matches specification.” Again, that’s something testing should uncover.

• “It’s about compliance with coding standards.” Uh… Do you even have standards? When were they last updated, 2012? That’s what static check tools like ATC are for anyway.

• “It’s a CYA activity for the development team and/or just an ego trip”. Please seek professional help.

As Matt Billingham put it in a comment to this blog post:

Essentially the goal is for all programmers to code as though if a bug ever needs fixing at 3am on a Sunday morning, the person who has to deal with it is a psychopath who knows where you live.

Code review is your chance to talk to that psychopath in advance.

Ideally, peer reviews should be a routine part of the development process, not a step that follows. Feedback from another developer can be surprising, regardless of their experience level. Don’t wait until the transport release to hear it. JP