Project: Writing on the Wall
Began in June 2021
This project is a social media game with a similar layout to Pokémon Go. You can walk around to discover notes and chests on the current-location-based map. You can then leave notes for other players to find with different charms attached to them, which change how the note interacts with the world.
It was a very extensive project with a robust backend, and it took a year of development (not full-time development) to release the beta version.
I created this web app using Node.js, React, and Sequelize ORM. It has a secure backend built with Express coupled with a React frontend to create a gamified social media app.
This project featured a comprehensive administration portal for authorized users of varying permission levels. In this portal, administrators could manage users, approve moderator-flagged items, send messages and gifts to players, check logs, and manage the items available in the game.
The administration pages used a CRUD React component that I made, which enables easy Create, Read, Update, and Delete options as well as an aggregate table view with little setup required, making it a reusable and versatile component for interacting with databases.
One specific point which I took care to consider is how to store and retrieve nearby geolocated items at scale. I considered different latitude/longitude algorithms, including taking a measured number of digits from the latitude and longitude components such that every matching component would be in a defined rectangle corresponding to the block being searched. I resolved instead to query the database by calculating the distance between latitude marks at the current longitude (as this varies with distance from the equator) and then determining the maximum distance away needed to be checked. I then refined this smaller chunk of data with a search that calculated exactly how far away each item was from the user's current location.
The live website is currently not available due to a decline in demand.
Check out a demo and walkthrough of implementation details in the link below!
