coffee_ Rebuild: Integrating Firebase

At the beginning of the year, I wanted to rebuild _coffee to include a few other features, watch this video for more context. As part of this rebuild, I wanted to completely rethink how I stored the data and introduce a way for users to create a pool of coffee data. Today, I took a good first step towards that and started working with Firebase and anonymous accounts.

Why Firebase?

I mentioned this in the video, but the main reason is so that I can just get something done and shipped. Firebase has storage, data management, and user access really easily accessible and I’m familiar with it so I decided that’s the route I want to go.

It is also really easy to iterate with whether I want to do that locally using emulators or live since I have no users. My plan once I do get users is to have a production version of the firebase console so that I can still play with live data, but that’s a bit further away.

Another bonus to Firebase is that the cost can somewhat be managed. I’m not sure how I want to monetize coffee_ yet. I don’t really want to introduce subscriptions, so my solution to data management needed to be low cost until I figured that out.

What’s the plan for accounts?

I did implement anonymous accounts. It was the quickest way to start getting the user’s data into firebase. I didn’t want to build any UI for accounts yet cause that’s a lot of work, so I went straight to anonymous accounts. My plan is to allow for users to sign in with apple and email at some point.

What’s being built next?

I stopped working while trying to get bean info to store in Firestore, so I’ll pick up there next week. The plan is to get beans writing to Firestore and get that info to come back into the app via Snapshot listeners.

If this project excites you and you’re wanting to help me, the project is open sourced and there are a couple issues I can use help on. Those are marked as “Help Wanted” in issues.