Generated image
|| #blog #2408-dailies

Resisting the urge to automate too soon

👋

I love automating stuff.

It's a fun challenge that has the potential to save me (or someone else) a lot of time.

For this website alone I have loads of ideas on how to automate things (big and small):

These all sound like fun projects, I'm excited to do them (or most of them!)

But I need to think about the cost to value ratio of each one. How much will it cost me (usually time) to automate this? And how much value (usually time saved) will it bring me?

For a lot of these ideas they math doesn't make sense. It will cost me a lot more time to write the automation than I will ever save with it.

Why spend 30 minutes doing something when you can spend 6 hours automating it?

  • My colleagues and I at some point

But there is another thing to keep in mind: change.

The more automated a system is, the less flexible it becomes.

For example, let's say I have a script to generate a new image for each post. Then if I ever want to change something about how those images are linked to a post (e.g. the frontmatter field name) I'll need to update the code. And the more automations I create, the more often I'll need to go back and change code if I change something in the system.

So before I can automate something I also need to think about how stable the current setup is. Because by adding automation I'm increasing the cost of change.


So I'm holding off on automating too much for now as I figure out how the system will work going forward and what makes the most sense to automate.

Even though I really really want to.

See you tomorrow 👋


Maybe try another one?

Nextcloud: Mount as a Folder How to mount your Nextcloud account as a regular folder on Linux
2021.05.02
How I fight tab hoarding Addressing my tendency of accumulating dozens of open tabs ⚔️
2024.09.18
Intro to Server Monitoring How to monitor your server with Prometheus and Grafana
2019.11.01