The world is rich with raw data waiting to be curated. This week we explore this data provided by informational API's and build the connected data sheet that collects this data. This exercise will allow us to explore data that may be valuable.
What is a Connected Data Sheet?
Imagine a page that pulled useful information in attractive ways. This was the promise of networked computing.
What is an informational API?
Companies that have access to information are willing to serve as the information suppliers to your applications - for a price. There are API's that get stock prices (Alpaca), weather reports (OpenWeather), bitcoin prices (Coinbase), even for getting Star Wars information (SWAPI).
Consider: nearly every single weather app in the App Store depends on the source of weather data from the National Weather Service so every app is essentially a nice design or funny jokes overlaid on the state of the sky.*
The value of connected data sheets
Any data can be enhanced by your own insights and this may be what your customers value. The data sheet should be timely, useful, or insightful. You can even draw insights through combining APIs. Just this week we learned about a newsletter that automates 99% of their work through running the news through basic summary APIs.
Example: Here we run a script that access the APIs of two cryptocurrency exchanges, Coinbase and Kraken. We can see that that there is price discrepancy. This could be very valuable to customers who want to perform arbitrage trades.
It's up to your imagination, what we can do with this data. We could create:
The Weather + Bitcoin Price app that tracks weather conditions and pairs them to bitcoin prices to autogenerate day activities - "Suns out, Bitcoin is at $10k! Put on some sunscreen and hit the beach!" This app can even hook into the Coinbase api and purchase bitcoin. (we explore interacting with API's in the next newsletter)
"Bitcoin up and weather sunny = Life is good, let's party!"
Why do this?
We are exploring the limits of useful information available through API’s. If all of the worlds data is not enough and we find holes in it, we can be the ones to create and provide it*. Later we can go beyond information and explore the idea that API’s that can perform actions like send a text, create a reoccurring payment transaction or order pizza.
Grab Postman or your favorite programming language and explore!
Note on speed: Remember that throughout this process you are building up your capacity for speed. It took once me a couple days to build a reusable template that hits the Open Weather API, saves the weather data to a database and returns a JSON response. I have this in my toolbox - the next similar API integration will take minutes.
* Dark Sky is a weather app that went beyond the stock weather data and built machine learning models to create their own predictions.