Inspiration
College textbooks are ridiculously expensive. Even used copies online can fetch hundreds of dollars. The College Board suggests that students set aside a yearly $1,200 for books and other course materials. We wanted to help students save money, while also preventing 30 million trees from being cut down specifically for textbooks every year.
Textbook acquisition and sale is a challenge Bruins face every quarter, and it has yet to be conquered successfully. Between the expense, the environmental impact, and the space they take up, having used textbooks around is a burden. With shelf, we sought to kill both birds with one stone.
What it does
shelf is a app-based internet platform enabling students to sell, buy, exchange, and donate used books. This creates a network of university affiliates leveraging the vast trove of available books.
How we built it
shelf employs a robust yet standard Android app with a Firebase backend utilizing phone number authentication, cloud firestore, and a real-time database. This runs with a dash of Google Maps API for location services, and a sprinkle of Algolia for cutting-edge search and discovery.
Challenges we ran into
A few hours into Firebase searching, we realized that it was very limiting in nature. It only allowed exact pattern matching - which meant that only one field could be indexed at once. To resolve the issue, we had to employ Algolia for search and discovery, a tech stack we were all fairly unfamiliar with.
Accomplishments that we're proud of
Our impressive database management rodeo. Neatly organizing huge unstructured, heterogeneous data into a fast-loading, easily accessible format. While natural language processing (NLP) tools make it easier to process, we were able to devise a method to accurately and efficiently extract information.
We're also incredibly proud of the fact that we were able to combine interdisciplinary backgrounds into an equally interdisciplinary project, and complete a fully-functional version of the app with very niche functionalities aboard - all with a span of 36 hours!
What we learned
Google Maps API bad :3
Project management was very challenging because of the hybrid nature of the hackathon. We learned efficient workflows for communication like communicating data flow using entity relationship diagrams (ERDs) and delegating tasks in swim lanes to ensure parallelism.
What's next for shelf
We made an attempt at integrating computer vision and NLP parsing so people would be able to take a picture of the book and the application would autofill fields like the title and the author. While the time constraint disallowed us from making any tangible progress on this, it's surely something we're looking to implement eventually.
You can try shelf out by cloning our GitHub repository and running it with Android Studio's GUI.
Log in or sign up for Devpost to join the conversation.