Table of Contents
Fetching ...

Browsing without Third-Party Cookies: What Do You See?

Maxwell Lin, Shihan Lin, Helen Wu, Karen Wang, Xiaowei Yang

TL;DR

It is found that disabling third-party cookies has no substantial effect on website appearance including layouts, text, and images, validates the industry-wide shift towards cookieless browsing as a way to protect user privacy without compromising on the user experience.

Abstract

Third-party web cookies are often used for privacy-invasive behavior tracking. Partly due to privacy concerns, browser vendors have started to block all third-party cookies in recent years. To understand the effects of such third-party cookieless browsing, we crawled and measured the top 10,000 Tranco websites. We developed a framework to remove third-party cookies and analyze the differences between the appearance of web pages with and without these cookies. We find that disabling third-party cookies has no substantial effect on website appearance including layouts, text, and images. This validates the industry-wide shift towards cookieless browsing as a way to protect user privacy without compromising on the user experience.

Browsing without Third-Party Cookies: What Do You See?

TL;DR

It is found that disabling third-party cookies has no substantial effect on website appearance including layouts, text, and images, validates the industry-wide shift towards cookieless browsing as a way to protect user privacy without compromising on the user experience.

Abstract

Third-party web cookies are often used for privacy-invasive behavior tracking. Partly due to privacy concerns, browser vendors have started to block all third-party cookies in recent years. To understand the effects of such third-party cookieless browsing, we crawled and measured the top 10,000 Tranco websites. We developed a framework to remove third-party cookies and analyze the differences between the appearance of web pages with and without these cookies. We find that disabling third-party cookies has no substantial effect on website appearance including layouts, text, and images. This validates the industry-wide shift towards cookieless browsing as a way to protect user privacy without compromising on the user experience.

Paper Structure

This paper contains 18 sections, 2 equations, 13 figures, 2 tables, 1 algorithm.

Figures (13)

  • Figure 1: Distribution of cookie counts after cookie notice interaction. Note that the curves in Figure \ref{['fig:accept_cookies']} and Figure \ref{['fig:reject_cookies']} are very similar, suggesting that most websites do not respect user choices in cookie notices.
  • Figure 2: Percentage of each element type present within the set of all successful and all failed clickable elements. A clickstream is generated by constructing a sequence of successful clickable elements. When the generated clickstream is later traversed, a clickable element fails when its CSS selector can no longer be resolved.
  • Figure 3: Normalized histogram of clickstream length across crawl groups. Clickstream generation occurs in the baseline group while clickstream traversal occurs in the control and experimental groups.
  • Figure 4: CDF of BCE screenshot differences (Algorithm \ref{['algo:bce_comparison']}). A clickstream length of 0 means taking a screenshot of the landing page without any clicks.
  • Figure 5: CDF of frequency vector DiDs (Equation \ref{['eq:did']}).
  • ...and 8 more figures