Thought about the flow for how I would like to start going about displaying my financials publically. The GitHub Repo is here, but it doesn't have much yet. No real programming has happened yet, just schema setup.
The public facing version of my financials will be extremely bare-bones and stripped down. This might even be a statically rendered HTML page. If you are interested in more details on my financials I will be exposing those as well, but there might be some kind of paywall. I am not sure exactly how this is going to look yet. Probably will give anyone access if they email me firstname.lastname@example.org. If there is a paywall access will probably be in some funny tiers or something.
Set up the basic schema for Investments and Transactions. It's all GraphQL via Hasura. Will be trying to use Apollo on the backend to play with. I'm also very curious if I can link this into my main DB via external connection in Hasura. That would be pretty awesome, though there are some things that would definitely be missing which I will need to consider or hack up. (Specifically how to add entities). Going through these DB's is interesting and helps to give me an understanding of what is existing out there. I would also like to play with a graph database.
Have 3 parts to the overall 'app'
So far the experience with the API is okay, but there's a lot of layers to go through. They really want to force you through certain paths, which forced me to rethink my own approach. This mainly had to do with login of financial institutions. It seems the best to go directly through their prebuilt front end flow (Plaid Link). Otherwise the API seems pretty nice and well documented. Definitely helped set up some of the tables I am thinking of as well.
We have thought out some things and begun to flesh them out. So next steps:
Oh man so many hacks but this is fun 🤪. So far my favorite is learning that in Go, map access is not threadsafe. So what's the solution?! Just throw a mutex around it 😝. Didn't even consider where to throw the mutex, just put it around the whole func 🤦🏽♂️. For now this is fine, but really I will have to think about where this mutex needs to go in the future, it should be pretty clear, but the code is evolving so fast (over half the file is deprecated) that I am not too worried.
Anyway overall this was a massive success and went really fast. There were a few hiccups. Mainly with me not paying enough attention to the code. But anyway, all the podcast data from my podcast app is now going into this centralized DB.
Basically the central DB has an
entity sitting over any row in any table. This allows any two arbitrary rows to be connected to eachother. I plan to use this in some sort of writing tool (with awesome autocomplete), to help write these faster.
May-li's thread: https://twitter.com/mayli/status/1308287703268487171
Also this: https://twitter.com/mayli/status/1308605022276587527
My brain is starting to fall apart now. Definitely taking note of this and thinking of writing more as I go and making sure to take the time to do this.
inboxfor things to process
inboxwill have inputs from: