Talking Point Tracker
Bad Idea Factory is proud to present the talking point tracker, a tool that watches the content of news television and attempts to figure out the zeitgeist of a given segment of time.
This project leverages Opened Captions, a project that exposes TV captions in real time for programmers to use in non-consumptive applications.
This project requires Python 3 and node. The project is written in node, but the NLP magic is all from the spaCy project (which is python).
Installation and setup
- ESLint and Babel for making sure that strangers don't judge use
- GraphQL + Sequelize + Postgresql for storing and exposing our deepest secrets
- React + Express for helping us overcomplicate our interfaces
I feel like I'm forgetting something.
This project is modeled after our Truth Goggles project in structure. Here is an overview of the main directories:
/configcontains configurations that go beyond the kinds of simple values that would normally be put inside of
/docscontains project documentation, mostly for use by human beings who want to contribute to the project or otherwise understand our diabolical plans for the future.
/srccontains the actual code.
/src/servercontains code that runs on the server. This includes APIs, worker scripts, models, data, and logic for how to present things to the user. It basically contains everything that doesn't get loaded into a browser. There are subdirectories in here but when I started listing out what they were for I realized I was just repeating the names of the folders in my descriptions (e.g.
/src/clientcontains code and content related to the user experience. This is where react and it's fellow demons get written, and ultimately compiled from.
/src/client/publiccontains static files. This is where things get compiled to, and is the only directory that the server code actually serves.
How to contribute
Be sure to take a look at our contribution guidelines as well.