censorship-course

Internet Censorship Course

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

Throttling and Rate Limiting

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.

Throttling is the book’s cleanest example of Molly Roberts’s friction: the site is not blocked, it is just slow enough that you give up. The chapter is honest that detection is genuinely hard — Wehe results contradict themselves on the same network — and that’s precisely why throttling is politically attractive. These breakouts push on accountability and on where “throttling” ends and “everyday network management” begins.


Breakout A: Plausibly Deniable Censorship

Motion. “Throttling is the most dangerous form of Internet censorship precisely because it is deniable — and detection tools like Wehe are too unreliable to hold anyone accountable.”

Prep reads (5–10 min).

Discussion prompts.

Bring back. One accountability mechanism your group thinks could actually work against throttling — measurement, regulation, transparency, market pressure, or something else — and the strongest reason it might fail.


Breakout B: Platform Throttling vs. ISP Throttling

Motion. “When a platform (YouTube, TikTok, X) throttles or down-ranks content, that is worse than an ISP throttling the same content — because there is no regulatory backstop and no exit.”

Prep reads (5–10 min).

Discussion prompts.

Bring back. Your group’s ranking (worst to least-bad) of: ISP throttling, platform down-ranking, OS-level app friction, CDN de-platforming.


Instructor notes

These map to §2.3.1’s core arguments: throttling is a form of friction (Molly Roberts), detection is genuinely hard, and the line between intentional shaping and congestion is porous. Breakout A is the natural pairing with the Wehe discussion — students who took the technical detour on token buckets will engage with the evidentiary-standard question. Breakout B is looser and more current-events-driven; it works well if the class has been circling net-neutrality debates or platform regulation. If time is tight, A is the tighter fit to the book chapter; B is better if you want to bridge forward to later platform-power material.