"Readability" is subjective

I’ve talked about the importance of code readability before, and so has one of my podcast guests. But “readability” is pretty subjective.

Quick. What does this mean?

/(cat|dog)s?/

If you understand regular expressions, this should be pretty readable.

If you don’t understand regular expressions, you’ll be at a loss.

So is the code “readable” enough? Obviously it depends. Do you expect the readers of your code to understand simple regular expressoins?

So here’s my rule of thumb to determine if code I write is “readable enough”:

Will the least experienced developer likely to read this code be able to understand it?

When developing a simple web service, I usually assume that the least experienced developer to see the code may be an intern, or a recent coding bootcamp graduate. So I strive for “very readable” code.

However, if I’m writing some deployment scripts for Kubernetes, it’s far less likely that a 3-month bootcamp grad will be tweaking those files. I’m willing to be “a little less readable” if it makes the job easier for me.

Or maybe you’re hacking a kernel module to improve the performance of some network filtering. Most likely, the types of developers reading that level of kernel code have many years of experience, and you don’t need to focus on child-like readability very much at all there.

The bottom line: “readable” is subjective, so keep context in mind when writing your code to be readable.

If you enjoyed this message, subscribe to The Daily Commit to get future messages to your inbox.