Government blackout

A government internet blackout is the deliberate shut down of civilian internet access by a government for a small area or many large areas of its country. Such a shut down is typically used as a means of information control in a brief period of upheaval or transition. In autocracies, internet shutdowns have appeared especially in the context of contested elections and post-electoral violence.^[4] It can impede the ability of protesters or insurgent forces to mobilize and organize. It also serves to prevent real-time information access for foreign people or entities. Reactions from leaders, journalists, observers and others in foreign countries can be delayed.

Weather or natural disaster

Extreme weather events and natural disasters can lead to internet outages by either directly destroying local ICT infrastructure or indirectly damaging the local electricity grid. The Monash IP Observatory and KASPR Datahaus have tracked the impact of Hurricane Florence 2018,^[5] Cyclone Fani 2018,^[6] and Hurricane Laura in 2020.^[7]

Cyberattacks

Accidents

Expert Analysis

Several organizations use expert analysis to identify internet shutdowns.^[16] Some, such as the Digital Society Project (DSP), send out surveys to experts around the world, and then aggregate the results into a single score. For internet shutdowns, the DSP asks "How often does the government shut down domestic access to the Internet?" where answers range from "Extremely Often" to "Never or almost Never."^[19] Freedom House's Freedom on the Net report also uses expert analysis to assess whether internet shutdowns have occurred, but instead of surveying multiple experts, Freedom House identifies and partners with a single expert to conduct an analysis.^[12] Freedom House asks the question "Does the government intentionally disrupt the internet or cellphone networks in response to political or social events, whether temporary or long term, localized or nationwide?"^[12] Generally expert analyses are more prone to false positives and fewer false negatives (i.e. identifying shutdowns that other sources cannot confirm), than remote sensing methods with manual oversight.^[16]

Remote Sensing

Other organizations use various remote sensing techniques to identify shutdowns.^[16] Some organizations, such as the Open Observatory of Network Interference, the Internet Censorship Lab and the Monash IP-Observatory use automated remote sensing methods to detect internet shutdowns.^[11][20][21] The Open Observatory of Network Interference uses software installed on computers of volunteers around the world to detect shutdowns. However these methods are prone to false positives, false negatives, and various technical challenges.^[22][23][24]

In order to address these concerns, some organizations have implemented various methods of oversight.^[16] Organizations such as Access Now and the OpenNet Initiative use such methods.^[15] Access Now uses technological methods to detect shutdowns, but then confirms those shutdowns using news reports, reports from local activists, official government statements, and statements from ISPs.^[15] The OpenNet Initiative has volunteers install software on their computers to check websites from access points around the world, then confirms those results with manual observations.^[25] These methods are prone to more false negatives and fewer false positives (i.e. all shutdowns that these sources identify can be confirmed by other sources) than expert analyses.^[16]

A comparatively new method for detecting internet shutdowns is remote sensing with automated oversight. These methods have been praised as more ethical and efficient as they do not endanger in-country volunteers.^{[26][27][28][23]} However these methods have yet to produce regular datasets.^[16]

Temporary alternative forms of communication

This section is empty. You can help by adding to it. (April 2020)