The big tip for post #8 in the be a better programmer while still having a life series is to become a witch. A Terry Pratchett style witch, to be precise. Terry Pratchett’s witch characters are really great at two things: first sight and second thoughts. To quote him directly:

First Sight and Second Thoughts, that’s what a witch had to rely on: First Sight to see what’s really there, and Second Thoughts to watch the First Thoughts to check that they were thinking right.

And no, I’m by no means the first person to connect Pratchett-style witchery with programming or design. Hint: go read that blog post, it’s really good.

Back at my original point, first sight is seeing what’s actually there, not what you wish was there or what you thought was there or what you meant to put there. Does that remind anyone else of debugging?

Fortunately for programmers, we have tools like debuggers and IDEs to help us see what’s actually there. We also have techniques like simply getting up and taking a walk, or explaining our problem to a rubber duck (or maybe another programmer if it’s a really hard problem), or commenting out half of our code and then half of that half and so on until we find the problem line. Let’s just not think about how much programming must have sucked in the days before friendly IDEs that highlight mistakes for you :)

Unrelated image from to make this post look nicer in social media shares.

Another part of first sight for programmers is also your attitude. If you don’t want to see the problem, you’re just not going to no matter how observant you are normally. I’m by no means perfect at it myself, but I’m convinced the most useful attitude you can bring to debugging is the simple acceptance that you got at least one thing wrong. The longer you spend insisting that your code should work, the longer it takes to figure out what’s actually wrong with it.

Moving on, second thoughts are thoughts about your thoughts. When you think you know the best way to build something, why do you think that? How do you know you’re right? Is that actually the best way or is it just the first way you thought of? How would you know either way? What constitutes the “best” way to do something? Is “best” the most performant, the easiest to read, the easiest to change, the quickest to write, the easiest to test? If “best” for your project meant quickest to write yesterday, does it still mean that today? How would you know when that changes?

Checking up on yourself like that is really hard to do and that’s why this post is more for me than for you – I’m trying to remind myself to question my assumptions.

One of the traps I fall into most often is looking for an example of what I want to do in our existing code and then assuming the first thing I find is the right way to do it. Shockingly enough, codebases change over time. Just because something worked well when it was written doesn’t mean nobody has thought of a better way since then or that the rest of the app hasn’t changed enough to make the old “right way” completely different from today’s “right way.” Just like you look for a couple of sources that agree with each other when you’re Googling what an error message means, look for a couple of examples in your codebase and if they’re different, check which one is newer.

Getting into the habit of thinking about how you think is not easy (at all!), but it’s useful and, like the other installments of this series, not something that you have to devote all of your free time to. It’s also useful in pretty much every area of your life. When you have any problem to solve, how do you know you’re right about how to solve it? For that matter, how do you know you’re right about what the problem is?

When I’m stressed out, every little thing drives me absolutely crazy. I can end up convinced that what’s bothering me is that this stupid freaking feature won’t work no matter what I do when the real problem is that I’m trying to hit a tight deadline and marketing keeps changing their minds about what’s important and half the QA team is sick so they need extra time to test everything and that means I need to deliver even sooner and everything is terrible!

Okay, so what do you do about that? For starters you really should read that blog post I linked earlier, Amy Hoy goes into a lot of detail about learning to notice yourself thinking. My big tip is just to get into the habit of asking yourself “Why? Why did I decide that? Why is that the best way? Why is that bothering me so much?” Sometimes the answer is going to be stupid simple: I decided to go to cafe at the front of my building for lunch because the weather was hideous and I didn’t want to go outside. Sometimes the answer will lead to more questions, like when you ask yourself “Why did I decide to put that config file in that directory?” In my case the answer was “Because that’s where the other config file lives” which leads to another question: “How do I know both config files should go in the same directory?” From there I learned all sorts of stuff about which files were supposed to go in the original directory and why, and where the other file that was related but not the same type of config ought to live.

This is the kind of thing that takes a lifetime to master, so don’t feel bad if you don’t get it right away. Asking yourself those questions is still worth it even if you only remember to do it sometimes.