Is this real measurement data?

Yes — real flagged tests from OONI's own public measurement API.

No. The measurements API is fully key-less.

Does an anomaly prove censorship?

No — it is a real automated signal, not a confirmed finding.

Yes — free, no download, in your browser.

Why does a country with more measurements rank higher?

The counts depend on how many tests are run, so places with active testers can produce more anomalies even if the underlying blocking is similar.

What is a confirmed measurement?

OONI marks a result as confirmed when it is highly certain a resource is blocked, for example when an ISP serves a block page or DNS returns an IP address linked to censorship. A bare anomaly is weaker evidence.

Who can see the measurements?

OONI publishes measurements in near real time through OONI Explorer and its API, so researchers and journalists can inspect individual results.

Internet Censorship Watch Screensaver

The Internet Censorship Watch screensaver ranks real recent internet-interference measurements. It is free, runs in your browser, and ESC exits.

How the Internet Censorship Watch screensaver works

The saver reads real flagged measurements from OONI's own public API.

Countries are ranked by a real recent count of anomalous test results.

The pipeline refreshes on OONI's own continuous measurement schedule.

No key is required for this key-less public endpoint.

What OONI measures

The Open Observatory of Network Interference, or OONI, is a free software project that aims to increase transparency of internet censorship worldwide, and is registered as an independent nonprofit, the OONI Foundation ETS. It builds OONI Probe, free and open source software that tests whether websites, messaging apps and circumvention tools can be reached from a given network, and publishes the resulting measurements openly.

A test result that shows possible signs of network interference, such as a blocked website or a middlebox, is flagged as an anomaly. A result is marked confirmed when OONI is highly certain a resource is blocked; for websites, under current heuristics, that means an ISP serving a block page or DNS returning an IP address linked to censorship. The sample ranks countries by a count of recent anomalous results.

An anomaly is a signal, not a verdict. OONI notes that a flagged result can be a false positive caused by factors such as unstable networks or unreliable servers, so anomalies should be reviewed over time before drawing conclusions. Counts also depend on how many measurements are collected in each country, so a high count may partly reflect heavy testing.

A worked example

A real country where recent tests flagged blocked or interfered connections rises toward the top of the board.

Settings & tips

Counts are real, unedited OONI anomaly measurements, not accusations.
Accent recolours the bars and header text.
OONI is the real Open Observatory of Network Interference project.

Frequently asked questions

Is this real measurement data?: Yes — real flagged tests from OONI's own public measurement API.
Is a key needed?: No. The measurements API is fully key-less.
Does an anomaly prove censorship?: No — it is a real automated signal, not a confirmed finding.
Is it free?: Yes — free, no download, in your browser.
Why does a country with more measurements rank higher?: The counts depend on how many tests are run, so places with active testers can produce more anomalies even if the underlying blocking is similar.
What is a confirmed measurement?: OONI marks a result as confirmed when it is highly certain a resource is blocked, for example when an ISP serves a block page or DNS returns an IP address linked to censorship. A bare anomaly is weaker evidence.
Who can see the measurements?: OONI publishes measurements in near real time through OONI Explorer and its API, so researchers and journalists can inspect individual results.

← All World Live screensavers