Walter Manger | To The Triforce
Quests for Knowledge

Side Project Scheduling

I’ve recently starting getting up earlier to work on side projects. When I say earlier, I mean, at about 4:15AM. This should give me a good 1.5 hours to work on something and get to a point that I can commit something meaningful before I have to start getting ready for my day. The goal is to get that number down to something like 40-45 minutes so I can spend the other half doing something else like reading or researching new technologies. What I am finding is for that 1.5 hours to be effective, planning the night before is pretty important. I’ve spent a few mornings getting up to speed for a half hour. It’s really important to eliminate that discovery time.

What am I working on?

When I first wanted to learn Go, I started making a game that could be played in Slack in the form of an integration. It is called Slack Battle. What I thought would be easy to slap together in a night or so is turning out to be a little more difficult than I’d planned. Creating games is difficult – even really simple ones. Add the requirement for playing games over HTTP and the need for some sort of persistence, and the difficulty increases.

The New Plan

The project was first created with Google App Engine, but coming back to the project after it had aged a bit, I thought that it would be so much simpler not being boxed into what GAE offers. Really, the only thing GAE was providing me was a datastore. I decided to use Mongolab instead. Why? Because it doesn’t really box me in. It feels as if I had an instance of Mongo running locally.

  1. Rip Out GAE
  2. Integrate Mongo & Mongolab
  3. Refactor for better readability and prepare for testing
  4. Add tracking & logging
  5. Actually write tests
  6. Create a front-end for testing (React / Angular)
  7. Deploy to (Digital Ocean, Heroku, …)