The ‘open’ internet was supposed to give us a worldwide ‘network society’ where our communications would move from being controlled from above through broadcast models, and towards more horizontal ‘mass-self-communication’. The excitement about the use of social media in the Arab Spring and even the furore over Anonymous’ (temporary) disruptions of some minor engines of capitalism suggest that we are still tantalized by the potential that technology appears to bring. At the same time, we become worried about exploits of the networked power of the internet – that come in the form of cybercrime and widespread breach of existing laws and norms like copyright.

Increasingly, the negative and disruptive aspects of the ‘open’ internet seem to be getting more attention than the potentially positive ones.  Governments are concerned about the rise of cybercrime, the threat of filesharing to industries that depend on the control of intellectual property, and the control of dissenting speech.  Along with industries and police, they strengthen intellectual property laws, prosecute and shut down file-sharing servers, track individual activists through social networks, and arrange with Internet Service Providers to block and filter problematic internet content.

So now we are in a situation where law, policy, and architecture combine to close down aspects of the ‘open’ internet. This has the paradoxical result of driving underground some of the practices that used to take place out in the open – beginning with some of the more unsavoury actions that happen on the internet, like file-sharing, but also extending to the kind of activism celebrated as an example of the democratic potential of the ‘open’ internet.  On one hand the move away from the ‘open’ internet has inspired innovation in technologies like encryption, file-sharing and and community wireless mesh networks, but on the other, it could have longstanding impacts on our communication environment.

Yesterday, the Guardian reported that Pirate Bay, in an effort to resist a High Court decision that file-sharing sites should be blocked, has moved to a new system for filesharing, using magnet links instead of displaying torrent files on its website. Magnet files are links with no files associated with them, which avoid tracking by containing very little information apart from an indication of the content they are associated with. The attraction of magnet links, according to SoftPedia, is that they make it easier for file-sharing sites to avoid accusations of wrong-doing in court.  Other file-sharers use ‘cyberlocker’ technology where users pay for passwords to third-party file servers (often supported by advertising) where they can leave files they wish to share with others. Unlike torrents, cyberlockers (as well as magnet links) are difficult to monitor. They are also incredibly useful for benign purposes like sharing files between work and home, or collaborating with other people – the popular file storage system Dropbox is a form of cyberlocker.

These changes in practice are part of a move where some of the more unsavoury and disruptive products of the ‘open’ internet shift to dark corners where it is more difficult for governments and courts to get a clear picture of what is happening. They may respond by passing laws or enacting policies that attempt to address illegal behavior but in doing so may overreact to actions that are not illegal. For example, the UK’s Serious Organized Crime Agency (SOCA) recently took over a music sharing domain after suspecting its operator of conspiracy to defraud – but not without initially posting a message implying that people who downloaded from the site may have conducted criminal offense. SOCA eventally changed the message, but the implication was that the use of ANY music site could be a criminal offence – which might well limit the number of people who want to use legitimate music-sharing sites, and push the less legitimate ones further underground.

Activism too is moving into the dark shadows. One of the consequences of the Arab Spring has been a greater attention by governments to the communications of its citizens – and in parallel greater attention from activists to securing or encrypting their activities. The New America Foundation’s Open Technology Initiative has been working on various prototype technologies meant to help activists avoid blocking, filtering, or internet outages. These include Commotion, a project that promises to use networked devices (mobile phones, laptops) as the points of connection in a mesh network that could grow to ‘metro-scale’. Designed to be decentralized and to link devices together in ad-hoc formations when and where required, the project promises to create an alternative network as an when needed. The New York Times reported on the project, which was supported by US government funds, calling it part of a ‘stealth internet’. My old community wireless networking co-conspirator Sascha Meinrath is quoted as saying “we’re going to build a separate infrastructure where the technology is impossible to shut down”. The article also reports that other veterans of community wireless networking have moved away from creating networks that help to share internet access towards networks that are designed for secure communications – including the FunkFeur wireless networking project in Austria. This project has been building an autonomous network across the city of Vienna which is owned by its builders, a longstanding goal which, in case of threats or constraints on the commercial internet, could provide an alternative mode of communication.

Other projects go even further: ArsTechnica reports on The Darknet Project, another proposal for a worldwide meshed network, and Serval, a project to create ad-hot wireless mesh networks using regular smartphones.

At one level, these projects feel like reinventions of the internet, which a collective burst of imagination framed as a platform for horizontal, networked communications. But now that the centralization and control of that platform is becoming evident, we need something else to imagine. The problem is that in creating darknets and super-encrypted dropboxes, all of the other benefits to speech that the internet has supported can get lost. One open internet, as compared to numerous separate and encrypted darknets, suggests the opportunity for global interconnection and communication. Already, social pressure and the habits of millions of internet users conspire to create ‘echo chambers’ online. What remains is a shared imaginary of openness, of a resource to be governed by its users. The rise of super-encryption and darknets suggests that this imagined unitary resource is fracturing. As more of the unsavoury action goes underground, so might the kinds of communication we think of as ‘open’ and democratic. What do we risk when the activists go underground?