by Alif Ibrahim
Inquiries & comments:

What happens when the internet dies? When it’s taken away?

It’s surely something we’ve taken for granted at this point.

Recently, it’s become apparent that governments have an increasing awareness of the impact and methods of supression and seizure of internet infrastructures in moments of political turmoil.

Internet Death Support Tools is the first of a three-part project that will research, compile and distribute tools for journalists and activists who need to be increasingly prepared to face the suppression of traditional and mainstream internet in times of political crisis.

Check back the page throughout the residency for a live and up-to-date development of this research.

== Protests & Internet in 2019 ==


MAY 2019 - JUNE 2019

Hong Kong

JUNE 2019

 .           .             


MAY 2019



- Wiranto,  Coordinating Ministry for Political, Legal, and Security Affairs

A Very Long Preface

In a project to create an alternative internet, we need to start by being specific in what we are actually building for. For one, “the Internet” is not just one singular entity. It is a series of interlocking pieces of technology, one that spreads across its physical, protocol and political layers. A segment of today’s calls for an alternative internet, especially those built on blockchain technologies, are aimed to fix issues around data privacy, data ownership and limiting surveillance. Some tools developed for this revolve around the decentralisation of data that tries to break the power that large private sector actors (FAANG, FAGA, FANG, the acronyms go on and on) has over the platform-centric mode of the internet today, but even fewer look at how to access this information in the first place when access is shut.

However, in the case of a national internet shutdown, the needs that need to be met are vastly different than common internet usage outside the moment of political crises (although, as theorist Wendy Chun argues, new media and neoliberalism is both built upon the creation of constant habit-forming crises). What’s important in this case will be detailed later below.

Internet governance scholars, for instance, usually divides the governance of the web into three distinct actors: the private sector, governments and civic society. These three segments often have conflicting interests, but it’s also that these conflicts exist within these categories as well.

The solutions, therefore, need to go beyond just the technical aspects but also take advantage of the political and social collectivity that an alternative internet needs to have. Of course, at its core, the technical reality of networking needs to be understood. Technology has always been shrouded with mysticism and rhetorical collapses that terms today don’t mean the same thing they did ten years ago, and even the word AI has lost its technical meaning. 

I am of course not alone, not the first and not the most powerful person invested in this sort of project. For one, this project addresses the first two of nine principles of Tim Berners-Lee’s new Contract for the Web, which also divides the principles based on the three actors in managing the internet governance. Principle 1 says “Ensure everyone can connect to the internet” and Principle 2 states “Keep all of the internet available, all of the time”, aimed at governments worldwide. As we’ve seen, that’s not always the case, and it certainly is something to strive for.

So what do we need for Internet Death Support Tools?
  1. Preparation: Internet users and communities worldwide need to realise that the internet can be taken away at any moment, so monitoring network outages and setting up infrastructure before the shutdown is critical.
  2. Sharing intranationally: Often, internet shutdowns are implemented to disrupt local mobilisation and organisation. Fast, lightweight and internet-less ways of disseminating information is needed to support mobilising efforts.
  3. Sharing internationally: As raising awareness to the international community can often impact the outcomes of local political crisis, sending information abroad when local networks have been shut down is also important and is the more technically difficult problem to solve. Creating direct links to civic society actors and reliable media establishments can serve as an important tool as common messaging platforms like Twitter or Whatsapp get filtered.

From this, it becomes evident that questions around privacy and data ownership takes a backseat to reliable and fast information dissemination. Before looking at the existing, available tools that might come in handy, it’s important to demystify some theoretical concepts about networking so as not to drown in the sea of jargon.

“New Media” is not always new

An important concept in new media theory is the concept of remediation, championed by Jay David Bolter and Richard Grusin in Remediation: Understanding New Media (1999). In it, they say that emerging media is usually a remediation of existing media. They argue that new media “wants to erase its media in the very act of multiplying them”, where the “twin logics of immediacy and hypermediacy” are invoked “to remake themselves and each other”. Although Bolter and Grusin focuses on the desire for immediacy in this seemingly contradictory and temporal feature of new media, the concept of remediation is always a useful tool to evaluate new technologies that appear or when formulating new technologies.

Instead of looking forward, it is then perhaps more useful to look at existing methods of alternative networking that exists and reformulating them to fit our needs. In the larger project of forming alternative technics and technical metaphors, championed by the likes of Yuk Hui and Phil Agre, it is just as tunnel-sighted to think that there currently exists only one form of interconnectivity. 

Internet access is distributed

Although power in the tech industry is concentrated in the hands of a few powerful companies that have stretched its reach worldwide, spreading their various iterations of tech-utopianism in what seems like an inescapable digital overreach, the access infrastructure of the internet is distributed and rhizomatic. This was deeply explored in Alex Galloway’s Protocol (2005). Rather than a centralised model (one hub with many nodes, like a command center) or a decentralised model (multiple hub with multiple nodes, like the global airport network), internet connectivity works as a distributed model, where each point of the network can communicate with each other even if one point is shut down. A real-life example that Galloway provides is the US interstate highway system, where if one part of the highway is shut down, it is still possible to travel to other parts by backing up and taking up a new route.

Ultimately, the development of internet protocols allowed the various existing networks to communicate with each other in a standardised technical language. The term internet, after all, is short for interconnected networks, joining together various local networks. Protocol, for Galloway, was the way that control exists after this decentralisation.

Networking is not solely a technical problem

Solving a networking problem using only technological solutions plays into the usual techno-determinist stream of thinking where technology solely determines the society’s cultural and political outcomes. We’ve always been involved in various forms of networking - the book is one of the oldest forms of information sharing, and the Sneakernet is one way that information gets shared without network connectivity. Creating solutions for journalists and activists facing an internet shutdown, therefore, also requires community-building, interpersonal and inter-organisation efforts to provide a structure for these technological solutions to be effective. Without a significant and meaningful mobilisation and adoption of alternative networking methods, any solution created will become yet another technological gimmick.

Building blocks of Alternative Networking

Alternative networking will be an incredibly context-specific project to take on, but these are some building blocks that can be used. These building blocks serve as the culmination of the residency’s research, but are also just the early stages of what is possible. I encourage researchers, journalists and activists to start building communities and simple tools below to prepare for future shutdowns.


Tracking tools, such as the suite developed by NetBlocks and AccessNow’s #keepiton campaigns are great resources to keep track of when shutdowns are happening. An international network of journalists and activists, when regularly keeping track of when and where shutdowns are happening, can start to anticipate and prepare resources to try to establish a connection with those located in the shutdown country. Being connected means being prepared.

Secure networks

Often, rather than a complete case of an internet shutdown, governments might just be filtering certain apps and websites. Having a reliable VPN or connecting through TOR routers can be the first step to test whether this is the case. However, government filtering have become more sophisticated lately, blocking certain kinds of media instead of entire websites like in the case of Indonesia.

Mesh networking

One of the most promising existing technologies is Mesh Networks, a technology that connects devices directly with one another without needing a router. One recent example is the use of San Fransisco-based Bridgefy, a bluetooth-based messaging system, in the protests in Hong Kong. In this case, messages ‘hop’ between one phone and another, so if you’re sending a message to someone on the other side of the city, it will jump between bluetooth-connected phones until it reaches the desired destination. Since it doesn’t use internet infrastructure - no telephone cables or routers involved - a shutdown wouldn’t stop mesh networks from working.

Some commonly cited and relatively established mesh networks are the NYC mesh in New York City and Guifi in Spain. The concerns regarding mesh networking is as follows: it can still be monitored (privacy and encrypted networks are a key concern), it requires widespread adoption to work effectively and international connections need to be pre-planned to be available once the shutdown occurs.

Analog distribution

From USB smuggling to and from North Korea, local distribution of curated content in Cuba, also known as El Paquete Semanal. Especially interesting in the latter case was that it was set up in an environment of extended internet dis-connectivity. The linked paper cites that Cuba only had 11% internet connectivity in 2011, and home-based connection is even rarer. The weekly document delivers about 1TB of data with various media content like films, music videos and the likes.

Two of the most important things to learn from El Paquete has to do with the established networks of people that is required for the smooth operation of the program. The first is that although this wasn’t a government-led program, they certainly didn’t do anything to stop it, which could very much happen an a temporary moment of communication suppression. The second is the presence of some curators in areas with high bandwidth, like in tourist hotels or universities (while others wait for smuggled harddrives or record satellite content). In the Democratic Republic of Congo this year, some businessmen travelled to neighboring Rwanda to check their emails.

Combined with the fact that the internet shutdown in Iran still left 5% of the population connected, establishing contacts in locations with potential connectivity in times of a shutdown and then relaying information through mesh networking / analog distribution then becomes a potential way to overcome a national shutdown.

Delay-tolerant networks

An architecture that is used in remote areas or those not well-served by TCP/IP. As stated in Vigil-Hayes et al (2018), one established example of this is Daknet, deployed in Cambodia and India. Initially, using bus-mounted routers, emails are sent and downloaded asynchronously, meaning that although it is not immediate, the emails still get sent, somewhat like a postman delivering mail, but instead, it was a bus. A prerequisite of this, of course, is that this bus would need to travel without hindrance in an area that are already facing political turmoil, and that interactive forms of communication is not possible.