3 ms·
I've always worked on side projects, even before being exposed computers at school. So for me, it's been and still is a hobby, which fortunately I can make a li
by karmakaze 8y ago
I've always worked on side projects, even before being exposed computers at school. So for me, it's been and still is a hobby, which fortunately I can make a living doing.
Given that, I choose pretty much the same way. I'm always looking for new languages, frameworks, apps and try to come up with some app ideas myself. The best ones are the small ones you wish existed and would/do use yourself.
The only general pattern is novelty. I typically use things that I don't get to use at work. I learn different things that may not immediately improve my work but eventually the cross-training as much higher value bringing in wider views that might otherwise be missed in discussions.
One time, I tried to make an app idea using a new language, back-end framework, and front-end framework. I realized that I might not even make much of a dent, so switched to using what I use at work (Spring Boot I think it was) to save time so I could get to the visible parts of the app. That was a first-hand internalization of the 'faster beats better' startup mentality.
The opposite also happens. I ran a Chef server and recipes to create a master-master MySQL DB to host a Ghost blog on a third machine. I haven't blogged much afterward but obviously I enjoyed the process and learned a lot.
Another time, I set out to make an app and get it in the app store. I chose Android since I already know Java and had watched an online course that described the lifecycles and layouts/views. It was surprisingly straight forward googling and pasting.
As you can tell, I do a lot of back-end/devops, so I did take the time to try out Vue just to have a modern SPA default. I picked it because it looked simpler to me than React. I'm currently looking for something in between Java and Go, imagine Elixir with static types, fast compiles, and small/fast. I'm hoping Crystal might be it but can't decide between Lucky or Amber frameworks to try out.
Lastly, side projects keep me sane at work. I can try anything on a whim without the (valid) constraints of choices at work. It's interesting by definition so balances out periods of powering through work I already know but needs to get done.
One last tip if you don't feel motivated. Rather than doing the thing, just get started setting up an environment to do the thing. e.g. install software, pick a name, make a repo. (I usually lose a bit of time picking, registering, later abandoning a domain name, but it's fun = worth it). More often than not, it seems so ready to use that it's irresistible.