censorship-course

Internet Censorship Course

View the Project on GitHub noise-lab/censorship-course

Measuring Technical Controls

Format. Break into groups of 4–5. Each group picks one of the two breakouts below (or takes both if time allows). Spend ~5 minutes skimming the prep reads, then ~10 minutes debating. A designated reporter brings the group’s position — and any dissents — back to the full class for a ~3-minute report-back.

Chapter 5.2 argues that measurement is hard precisely because censorship is hidden, intermittent, and hyper-localized — and every vantage-point choice trades off representativeness against risk to the humans on whose networks the probes run. Both breakouts today press on that tradeoff.


Motion. “Distributing censorship-measurement software (OONI Probe, VPN reachability testers, side-channel probes) to volunteers in repressive jurisdictions is ethical only if the researchers can guarantee the probe’s traffic is legally indistinguishable from ordinary browsing — otherwise informed consent alone is not enough.”

Prep reads (5–10 min).

Discussion prompts.

Bring back. Your group’s minimum ethical bar for shipping a censorship probe to volunteers in a Freedom-on-the-Net “Not Free” country — one concrete requirement, and one thing you decided not to require and why.


Breakout B: The Right to a Blocklist

Motion. “Governments that filter Internet traffic should be legally required to publish, in machine-readable form, the current list of blocked domains and the legal authority for each entry — the way Turkey has at times done, and the way Germany’s NetzDG orders are logged.”

Prep reads (5–10 min).

Discussion prompts.

Bring back. A one-sentence rule for what a blocklist-disclosure law should require, plus one specific loophole your group predicts governments would exploit.


Instructor notes

These breakouts map to the section takeaways on vantage-point tradeoffs (5.2.1–5.2.3) and to the ethics chapter (5.5) — specifically the “paradox of consent” and “co-opted devices” arguments. Breakout A tends to produce sharper disagreement (students split on whether consent is a floor or a ceiling); Breakout B tends to surface the tension between measurement researchers (who benefit from opacity being expensive) and civil-society groups (who benefit from opacity being illegal). If time is tight, A is the more directly technical debate and pairs better with the walkthrough in section 5.2.