Read on Twitter
Short story time. There’s a side project thing I’ve been procrastinating on for weeks because it seems daunting technically. Nothing too hard but it was a little bit out of my comfort zone and I was hoping to either not need it or use a library.
So I’ve been looking through the libraries today. I’ve already tried a couple before. None of them are quite what I wanted and fit the use case. Some may fit it but they are very generic and it sounds like I have to really stretch my code to speak in their language.
I ended up just biting the bullet and writing the code today. Turns out it’s just eight hours of work, roughly 500 lines of (a bit messy) code. Everything fully customisable, no need to contort my code to someone else’s abstractions. Feeling of relief as I can proceed deeper.
It’s been a while since I did moderately challenging UI work so my mental barrier went up a bit. But in general, this reminded me of how I got stuck in similar ways early in my career. Spending weeks trying to make two libraries talk when neither was really good for the use case.
This is something you learn eventually in your career, that third party solutions may be too imprecise and you need to make your own from first principles. And when to do (or not do) that. But there’s a lack of materials on how to think this way. That’s an educational goldmine.
