YouTube's recommendation algorithm is designed to increase the time that people spend online. Many people, including me, had the experience of being dragged by YouTube's algorithm into a rabbit hole of videos, spending more time than desired, and coming out deeply unsatisfied with the quality of videos. We built Steadfast to help users overcome this unaligned interest between the YouTube algorithm and you. You can follow your favorite content creators, dig into the archive to find great videos that you missed, and keep track of your viewing progress.
At the current state, Steadfast is essentially a specialized RSS reader. Comparing to using generic RSS readers to read from YouTube RSS feeds, Steadfast gives you extra information, like the duration of the video, likes, and views, and a timeline interface that makes it easier to browse the archive of a channel.
We also intend to create a tunable exploration feature into Steadfast. Because YouTube API's quota creates some scalability issues, we are currently evaluating ways to use fewer and cheaper API calls to achieve the goal.
The technology used in this project:
- Database
- Firebase (user authentication)
- SQLite (channel and video information)
- Redis (user information)
- Web Interface
