Server-Side Blocking: Watching provider discrimination

Internet balkanization is an emerging threat to global connectivity. There is not yet a solid data-backed resource to track and understand the extent of the phenomenon. We propose to: (1) develop scientifically rigorous measurement methods to detect server-side blocking, (2) build a system to continuously and globally collect data on the phenomenon, (3) curate tools for mapping and comparative analyses of the data, and (4) provide the raw data and relevant analysis in a centralized and easily accessible repository. This resource will help raise awareness of Internet balkanization and insure Internet freedom tools account for it.
The goal of this project is not to generally advocate for working around geographical
prohibitions enacted by individual private companies; rather, we aim to develop systematic ways to differentiate server-side blocking from nation-state censorship, so that other researchers can avoid having false positives in their censorship studies — and consequently avoid criticism for misidentification. Without better data with which to identify server-side blocking by content providers, measurements of censorship will sometimes be faulty, and we can see this caveat already noted in some analyses of censorship (e.g, https://ooni.torproject.org/post/iran-internet-censorship/#non-governmental-forms-of-internetcensorship).
Furthermore, by quantifying the extent of service-providers’ balkanization, and
raising awareness of this issue, we hope to motivate service providers to choose better ways to limit access rather than blocking whole regions. In summary, this project aims to construct and operate a system for measuring server-side blocking, to develop strategies for identifying such blocking, and to publish analyzes of where and how services are blocking their content from users in particular geographic regions.