The Future of Detecting Internet Disruptions

When it comes to the internet, we live in a world of contradictions. While global internet connectivity is skyrocketing, governments are increasingly attempting to control their citizens’ access to the internet by enacting policies and investing in infrastructure that allows them to restrict the information that their citizens can access online. In recent years, researchers, journalists, and civil society have documented an increase in global internet censorship. Just last month, both India and the Ukraine were accused of censoring social networks. In response to this growing trend of government control of information, diverse groups of stakeholders are coming together to advocate against disruptions. Other examples of network shutdowns, including in several African countries during elections and periods of political unrest, have demonstrated that network measurement data can be an important tool to show a broad audience, including other governments and international organizations, that the harm caused by these network shutdowns can be severe.

The act of interrupting or preventing access to the internet has many different names: censorship, shutdowns, disruptions, blackouts, to name just a few.  In spite of its many names, it usually refers to “an intentional disruption of internet or electronic communications, rendering them inaccessible or effectively unusable, for a specific population or within a location, often to exert control over the flow of information.” Diverse groups of stakeholders use these different labels for a variety of reasons: for some groups, certain terms have taken on a political meaning, and others are weighted with particular nuance given the terminology used within different research, engineering, or academic communities. However, internet shutdowns by any name can cause harm to the local economy, personal and community safety, as well as interrupting the ability to share information in order to create safe places, plan protests, or just share basic knowledge. Platform developers have been working for years to develop tools that identify internet shutdowns: they include tools such as the Open Observatory of Network Interference (OONI) and ICLab, just to name a few.

In addition to platform developers that work on the technology to track internet shutdowns, there are many other communities that explore the harms of internet shutdowns: academics, researchers, advocates, internet measurement experts, and more. It can take days or months for news of an internet disruption to emerge from a country or region, and the stakeholders involved in censorship measurement often struggle to identify, capture, and find the source of a disruption in real time. These communities also collect data in different formats, making it difficult to compare and use shutdown data in meaningful ways to shine a spotlight on this issue. In order to understand different stakeholders that make up the censorship measurement community, and identify commonly-held best practices and community needs, OTI conducted interviews with many of the stakeholders that are involved in internet measurement and released its findings in the recent report, Ensuring a Future for Detecting Internet Shutdowns: A Field Survey of the Ecosystem Around Internet Censorship, Disruptions, and Shutdowns. The paper is intended to provide an overview of current efforts in the censorship measurement community and focuses on recommendations to support the availability and accessibility of data on internet censorship.

We found that, in reality, there is no quick fix to ensure that censorship measurement stakeholders are on the same page when it comes to gathering and sharing data. Many of those interviewed for OTI’s report noted that a cross-platform dashboard where censorship measurement data could be collected and shared would be an ideal product for the community, but there are many steps, compromises, and challenges that need to be addressed before the creation of such a dashboard could even be considered. To begin that process, our recommendations include creating the following:

• Collaborative structures for data-sharing and coordination;
• Processes that allow the sharing of private-sector datasets and proprietary information within trusted communities;
• Funding and resources for long-term maintenance of measurement systems, and targeted support for rapid response interventions; and,
• Shared resources for comparison and presentation of measurement data.

A strong, collaborative community of censorship measurement researchers, developers, and non-technical stakeholders will help to document all abuses of government censorship and improve people’s ability to access the internet around the world.