On the Intrusiveness of JavaScript on the Web

Various web components and JavaScripts have been used for collecting personal identifiable information resulting in privacy concerns. Although several privacy preserving tools have been proposed to limit online advertising and tracking their use has been limited and mostly limited to tech-savvy audience. In addition to poor and manual filtering-list maintenance and confusing settings, these privacy preserving tools have, arguably, usability and intrusiveness issues. Among others, their brute-force blockage of all JavaScripts on a website, may result in broken functionalities thus effecting user's web-experience. In this work, we propose a framework to quantify the intrusiveness of JavaScripts with ultimate objective of measuring the usability of privacy preserving tools. We postulate that intrusive JavaScripts carry distinct characteristics that could be used to differentiate them from functional JavaScripts i.e., scripts that are genuinely used for enhancing the user's web experience. We propose a measurement methodology that can automatically separate tracking and privacy intrusive JavaScripts from the functional JavaScripts. Our methodology assumes only partial knowledge of the privacy intrusive JavaScripts.