The goal of this application is to measure the sentiment in online media and related discussions.

There are technical and conceptual reasons for using Reddit as a source for sentiment analysis. Reddit has a huge user base and a culture of hard discussions. And its API allows to retrieve data with a high free trier and reasonable performance.

It's clear that this application won't make the world a better place - but I hope it won't do any harm either.

It's also clear that the code/methodology used to calculate the indices is very simple and most likely does not meet standards considered to be state of the art. While simplification is a strategy to deal with complexity, it's not an excuse not to improve. In this sense, feedback or collaboration with coders or researchers is welcome .

Thanks to Oli for the initial idea and Pete for inputs and testing!