OONI

OONI: How Internet Censorship Measurement Became a Blocked Tool

What OONI Is

OONI (Open Observatory of Network Interference) is a free software project launched in 2012 by the Tor Project, later spun into an independent organization backed by the Open Technology Fund and other civil society funders. It provides command-line tools and mobile apps that measure network interference—detecting DNS blocking, IP blocking, HTTP blocking, and DPI-based filtering. OONI's user base spans digital rights researchers, journalists, engineers in censored regions, and academics studying network control. The project publishes anonymized measurement data in a public database, creating the most extensive empirical record of internet censorship globally.

Why It's Censored

China blocks OONI access primarily because the tool's measurements directly expose the mechanics of the Great Firewall. When users run OONI probes, they generate concrete evidence of which domains are blocked, which protocols are throttled, and how DPI systems identify and filter traffic. This transparency contradicts official narratives that China's internet controls target only "illegal content." The Cyberspace Administration of China (CAC), which oversees content regulation and infrastructure control, has not published explicit legal prohibitions against OONI, but blocking it aligns with broader restrictions on transparency tools and foreign-controlled network measurement infrastructure under the Cybersecurity Law (2017) and Data Security Law (2021).

Technical Blocking Methods

China's blocking of OONI operates on multiple layers. DNS filtering prevents resolution of OONI's primary domain to its legitimate IP address; queries return bogus IPs or timeout. Simultaneously, IP-level blocking targets OONI's known server infrastructure, rejecting packets to measurement endpoints from within Chinese networks. The method is blunt by design: OONI's architecture relies on reaching external measurement targets, so blocking major exit points is effective. For OONI's mobile apps, secondary blocking occurs at the app distribution level—neither Apple's App Store nor Google Play (where accessible via VPNs) hosts the app in mainland China. Additionally, DPI systems identify and throttle OONI protocol signatures when users attempt to run probes over standard connections.

User Workarounds

Users in China seeking to run OONI measurements typically establish encrypted tunnels that mask traffic origin and destination, preventing ISP-level DPI identification of OONI probes. Since OONI is measurement-focused rather than user-privacy-focused, users combine it with complementary tools: encrypted DNS (DoH, DoT) prevents ISP DNS interception, and circumvention tools handle the IP and SNI blocking layers. Some researchers use OONI on isolated networks or behind institutional infrastructure with less aggressive filtering. Building OONI from source code rather than using distributed binaries can reduce fingerprinting by DPI systems, though this requires technical literacy.

Alternatives

Citizenlabs' NetAlertX performs similar censorship measurement with lighter detection footprint, though it is also restricted in China at the network level. Censored Planet, a measurement platform operated by University of Michigan researchers, aggregates data on blocking patterns through a hybrid model that reduces individual probe exposure; it remains accessible in more regions than OONI but faces increasing restrictions. Neither alternative fundamentally solves the core problem: any active measurement generates detectable traffic signatures that centralized filtering systems can intercept.

Outlook

Restrictions on OONI are tightening, not easing. As China's internet control infrastructure matures under the CAC, blocking mechanisms shift from crude IP filtering to behavioral detection and protocol analysis. OONI's public database remains accessible outside China, preserving historical measurement data; the threat is to *collecting* new data. Absent significant political changes in internet governance, users in China will face escalating technical and regulatory barriers to running probes. The project's openness is its weakness in censored environments: transparency can be weaponized by filtering systems that learn from published blocking methods.