Weekly Recap: Simple is Complicated

Last week was a slow week when it comes to discovering or learning about new things. Here are a few things I encountered.

Make things flat

It’s a mantra in programming that you should strive for simplicity. But often we have to solve complex problems and coming up with a simple solution isn’t always doable or seems to be utterly impossible.

One thing I discovered is that it can help to work with flat data structures. Accessing properties of objects which are nested multiple levels deep can lead to hard to understand code, especially when you have to use loops and find a lot.

Writing simple code is complicated

Often I start with a good feeling about coming up with a solution for a new feature and writing the first tests everything seems to run smoothly. But an hour or two later, I discover edge case after edge case, and my code quickly becomes complicated. I don’t have the right solution for this yet; I have to work on my simplification skills.

State machines everywhere

Both the Full Stack Radio Podcast by Adam Wathan and the Syntax Podcast by Wes Bos and Scott Tolinski had an episode about State Machines with David Khourshid.

The concept sounds pretty interesting. Still, to me, it seems, similar to functional programming, a lot cooler and easier in theory and with simple examples as it turns out to be in real-world applications with all their edge cases.

But I haven’t tried them out yet. I’m very intrigued by the concept, and I’ll play around with it over the holidays. Furthermore, I find the possibilities State Machines offer in regards to testing very exciting.

Red, green, refactor in practice

Last week I tried to put more emphasis on the refactor step. It worked well. Not putting too much pressure on myself during the green phase to write clean code makes it easier to come up with a simple solution. During refactoring, you already have a better feeling for where to put abstractions.

I had mixed results with writing tests for the internal API during development but deleting them after the fact. Although I recognize that those tests can be a burden on maintenance over the lifecycle of an app, I can also see how they can be very beneficial when refactoring small parts of the SUT.

Write a failing test before leaving work

It seems counterintuitive at first, but I think it’s a good practice to leave work with a failing test. That way, the next morning, you know where to start.

const vs let

I discovered that there is some debate about the usefulness of const in JavaScript. As I understand it, the main point of criticism is that it’s easy to believe you can make objects immutable by using const, which is not the case.

Jamie Kyle wrote a pretty good overview of the topic, unfortunately with a lot of needless profanity (I don’t get it why some people think they have to use profanity in technical writing, but that’s another topic).

The practice of declaring everything as const if you happen not to redeclare the variable elsewhere and changing const to let as soon as you do want to change the value of a variable, indeed, seems pretty pointless.

Wrapping it up

I learned a few interesting things last week and I’m very excited to start playing around with State Machines. Maybe they are a useful tool for writing simple code.

References


Do you want to learn how to build advanced Vue.js applications?

Register for the Newsletter of my upcoming book: Advanced Vue.js Application Architecture.



Do you enjoy reading my blog?

You can buy me a ☕️ on Ko-fi!

☕️ Support Me on Ko-fi