In a new article in the BBC titled, "How a computer that 'drunk dials' videos is exposing YouTube's secrets," Ethan Zuckerman discusses how he and a team of researchers at the Initiative for Digital Public Infrastructure (IDPI) at UMass Amherst have developed a new tool for compiling YouTube metrics. Despite being one of the world's largest social video sharing platforms, YouTube and its parent company Google have been unwilling to release data on very basic questions about the platform, including information on its size, content, number of languages spoken, and more. Additionally, it has never allowed outside researchers access to the internal recommendation algorithms which drive content to viewers. In order to understand the platform's scope and scale, Zuckerman and IDPI researchers developed a program that scrapes information from YouTube by generating random URLs in order to create a random sample, allowing researchers to better answer critical questions about YouTube's form and function.
IDPI research led by Ethan Zuckerman on compiling YouTube metrics covered by BBC
