OnlineOrNot Diaries 23

Developers will literally spend weeks designing and building a perfect system instead of just solving for the problem they have today.

Normally I would say: it is me, I am developers.

But then I wrote out the problem to a friend on Whatsapp and realized I was just procrastinating.

I built for the problem I have today

An extremely common pattern for me (that even after almost four years, is hard to recognize and stop): building overly complex systems as a means of procrastination.

For example:

I recently had a problem where I wanted to display the limits a user is subject to (for example, if they're on a free trial, don't give folks as many SMSes as a paid account, to avoid abuse).

I started planning to build a complicated entitlements system based on a key-value store, that I could update at any time, for any user. Why? Because I've worked on systems with tens of thousands of daily active users as part of my day job for years now, and that's the first thing that comes to mind.

That's wasn't the problem I had to solve today though.

I already had functioning means of limiting what types of users could do what. The problem I had to solve today was to display that limit to the user.

I was chatting through the complicated system I planned on building with a mate over WhatsApp when I realized I was just procrastinating. I didn't need a fancy entitlements system (yet).

A single variable in a shared package would do the job.

Will it scale? No.

Will updating it be a pain in the ass? You bet.

Is it enough for the problem I have right now? Yes.

So I'm pretty thankful I have folks to brain-dump this type of thing to - it's difficult to notice when you're making the problem harder than it needs to be (even when you only have two hours a day to focus on the problem).