Is the multi-stakeholder model a viable way forward for Internet governance?
This essay was originally submitted in 2016 to the International Institute for Strategic Studies. Minor revisions have been made.
“We pay too little attention to the reserve power of the people to take care of themselves. We are too solicitous for government intervention, on the theory, first, that the people themselves are helpless, and second, that the government has superior capacity for action. Oftentimes, both of these conclusions are wrong.” – Calvin Coolidge, President of the United States (1923-1929)
The Internet as we know it today has roots in the technological developments of the Industrial Revolution and the inventions that came into being as a direct result of research and development into computational technologies by a diverse range of commercial, academic and governmental groups. In this article, we will address the role that combined technological contributions and shared governance has had on the development of the Internet since its inception, and speculate on how a workable model for the future may best develop.
Arguably, modern-day computation started with the Babbage Difference Engine, proposed by Charles Babbage in his paper of 1822 to the Royal Astronomical Society which took the often tedious and painstaking process of calculation by hand into the realm of automation. Over the next hundred years, computing developed from simple arithmetic into complex machinery with real-life applications, growth of which was driven by the twin engines of commerce and war through hundreds, if not thousands of contributors across a multitude of sectors. By the time World War II began, computation had reached a point where these applications of computing culminated in the development of the Colossus series of computers at Bletchley Park and subsequently the Bombe, used to decrypt German communications and which made a huge impact to the war effort.
Bletchley Park's NI1b and 'Room 40' government agencies were not the only organisations to develop significant advancements in computation around that time. Commercial organisations, and those affiliated to but standing independent from government, also drove progress. For example, free from Bletchley Park after the war ended, Tommy Flowers, the inventor of Colossus, rejoined the Post Office Research Station throughout the 1950s subsequently helping to develop ERNIE, one of the world's first random number generators, a successor of which is used today to pick Premium Bond winners; he and his team also developed hitherto unknown pulse modulation techniques for telephone exchanges. Across the Atlantic, IBM were busy with a multitude of projects, from the IBM Automatic Sequence Controlled Calculator (the 'Harvard Mark I') in 1944 to 1956 when IBM were busy pioneering disk drives in their San Jose laboratory. The transistor was invented in 1947 by a team at Bell Laboratories; the EDSAC, a paper-tape computer was built in 1949 at Cambridge University. From hand-held mechanical calculators to trackballs, the acceleration of progress in computing was marked, driven both by commercial and government interests, and with significant contributions from academic institutions and individuals too.
Throughout this period of early computing, evidence shows that individuals, government, academic and commercial organisations all contributed towards technological progress, and that premise holds true today. The technology 'gold rush', which has culminated in Facebook in the modern era, continued through the 1950s to 1970s with an ever-accelerating number of new insights, inventions and ideas. After the development of ARPANET by DARPA in the early 1970s, networked computing was opened to universities and consequently the general public. Open-source computing in the form of the UNIX operating system began as early as 1969, and anyone was free to develop and improve upon it. The miniaturisation of silicon chips throughout the 1970s and 1980s led to Moore's Law, named after the co-founder of Intel which states that the number of components on an IC chip would double approximately every two years (now thought to have reached a plateau).
Frameworks for these new technologies came not exclusively from central government, but from gifted individuals such as Alan Turing, John Von Neumann, Clive Sinclair and Jack Kilby. Sinclair, an entrepreneur, brought affordable consumer electronics to the masses; a latter-day James Dyson, his inventions ranged from the ill-fated Sinclair C5 electric trike to the wildly successful Sinclair ZX Spectrum. Kilby invented the integrated circuit as part of his work with Texas Instruments, an integral component of every modern computer system. Famously, Bill Gates co-founded Microsoft from a garage, a company which has grown to be a market-leading force today.
We can deduce then that the impetus for research and development into early computing was sourced from not one controlling entity, but by a vast array of experts working for a variety of causes. This understandably led to a variety of competing standards and protocols in the field. For example, TCP/IP was invented in 1974 independently from UDP in 1980, but both each able to handle Internet traffic (with key differences). Early console-type computers such as the Acorn Electron, ZX Spectrum and Commodore 64 competed in the commercial marketplace, and the rough-and-tumble of commerce drove manufacturers to produce bigger and better solutions.
Technological progress up to the 1970s and 1980s demonstrated that governments, corporations and individuals could work alongside each other developing new and exciting changes to the computing landscape. With each organisation having their own motivations – whether these were financial, philanthropic, militaristic or educational – progress in the field was at an apogee, with new developments arriving faster than the general public could consume them, and, tellingly, faster than the government could regulate them.
This lack of regulation arguably contributed to the explosive arrival of the Internet in the late 1980s. First operating via dial-up modem, connection to a World Wide Web of shared, unregulated, free material for the cost of a phone call was an attractive consumer proposition. Leading the way was AOL, with their 'GameLine' product – a service that linked an Atari 2600 via phone line to a server, allowing consumers to rent games for as little as $1.00. AOL transformed quickly into a generalist consumer Internet Service Provider, and alongside the likes of Yahoo they cornered the market in user-friendly Internet connectivity. Corporations driven by the survivalist need for growth and the requirement to please their shareholders with financial success were outpacing national governments.
Nevertheless, national governments tried hard to catch up, both in research and development and in introducing regulation obstensibly to protect Internet users and to safeguard national security. The appearance of the world's first computer worm, the Morris worm by a person who admitted they had created it to 'see how big the Internet is' in 1988 frightened the US Government and led to the very first conviction under the newly-minted US Computer Fraud and Abuse Act. It also provided a demonstration of how the Internet was never under governmental control, despite much of their funding originating from federal budgets.
IANA, the Internet Assigned Numbers Authority, who administer top-level DNS servers for the Internet, came into being after calls for regulation from key players in the development of the new World Wide Web such as Vint Cerf and Jon Postel. IANA is a not-for-profit American corporation, however it was financed by DARPA funds that originated from the US Government. When the U.S. Science Foundation tried to hand over control of IANA to a private corporation, Network Solutions, there was a backlash from Internet users who felt that the Internet would be in jeopardy of losing its freedom and independence, a period referred to as the 'DNS Wars' and which lasted until the creation of ICANN.
Today, there exists a tension and ongoing argument surrounding ICANN, who administer top-level domains on the Internet. This tension arises from the American nature and location of the ICANN organisation – with the Internet today being truly global, many nations feel that top-level Internet administration should rest with a global organisation not so closely tied to one particular country.
This power struggle at the top of the Internet 'chain' is a reflection of the nature of the true globalisation of the Internet. With regulatory oversight undertaken by a variety of stakeholders that are not exclusively governmental, the same kind of innovative fast-paced development of technology established in the late 1800s can continue into the future. With control of the upper echelons of the Internet held by a global board of members rather than a single state with specific aims, the Internet can be governed by a broader democracy.
Let us look at the alternative. Governments have a track record of interfering with or prohibiting new technology to progress their own goals or ideals. This has early historical precedents; on the invention of the printing press early in the 15th century, Arabic imams in the Ottoman Empire banned the printing press on religious grounds, with the first press in the region not established until 1729. More recently, the United Arab Emirates threatened to ban Blackberry mobile phones unless the encryption keys were provided to the government by the manufacturer. Today, Google Street View remains banned in Greece and Austria on privacy and national security grounds, and there is an ongoing argument playing out in the media between Apple and the U.S. Government about provision of technology to break the previously-uncrackable iPhone on iOS 9.1 in the name of fighting terror.
In historical cases, resistance to this kind of government instruction led in many events not only to violence but to capitulation or a reformation of the stance taken by the governmental organisation to bend to the will of society. As an example, think of the recent row in the United Kingdom about the 'Snooper's Charter' – the Investigatory Powers Bill – initially quashed by Parliament due to large opposition from both MPs and the general public, and now being put through Parliament for a second time by the Home Secretary under cover of the ongoing EU referendum debate. Suspiciously attracting zero coverage in mainstream media, it is likely this privacy-invading Orwellesque bill will result in legislation, which will have a profound effect on the ability of the UK government to spy on its citizens.
One example of a nation where this level of control, and more, has been successfully implemented on the population is North Korea. In this totalitarian state, the opposite of a multi-stakeholder model has been developed – with the government in full control of North Korean's Internet access, the dominant network is an intranet-like closed domestic system called Kwangmyong. Wider Internet access is prohibited except by special permission and only to a small group, under close government audit. This is reflective of the culture in this state, essential to allow the government to control and limit the knowledge and views of its people. Freely available Internet websites are not allowed on Kwangmyong unless pre-reviewed, approved and when necessary censored by government. Other states such as Cuba and Myanmar have similar systems.
Is this the kind of Internet we would like to strive towards in the West? The obvious answer is no. Freedom remains a founding principle of the constitution of the United States; while not enshrined in a UK bill of rights, it remains a foundation of UK culture; in Europe, freedom is a given in most, if not all member states. Single ownership evidently leads to unitary control; unitary control leads to censorship and limitations on individual freedom. The current multi-stakeholder governance and development models must be kept and nurtured then, both to allow people the freedom to express and share their opinions without fear of reprisal and to keep up the pace of technological progress.
Let us imagine an Internet governed entirely by states. How would such a model be funded? In the past, websites would be created on a non-profit basis by enthusiasts and experts, and many 'webmasters' would keep their websites operational from love of their specialism rather than the prospect of monetary gain. Even now, there are examples across the web of free services funded by donation, but these are dwindling in favour of 'freemium' services driven by commercial interests. At present, advertising revenue provides the momentum that keeps the Internet moving. Under a governmental mandate, such an Internet could not exist. Anyone who has worked for local or national government could imagine how such a quick-thinking, dynamic community would fit into bureaucratic, Kafkaesque models of governance. What if the controlling government forced users to fill out forms to seek approval for new websites? Or imposed caps on the number of posts a user could make citing resource constraints? Or, more seriously impose further restrictions on the freedom of expressions of users? Censorship of views, however moderate or extremist? What if a paranoid government, desperate to dispel anti-government protests, decided to prohibit criticism of the ruling party?
Such an Internet would be unthinkable, yet as described earlier, this model exists already in totalitarian countries. The West must be extremely careful not to allow the freedom of the Internet to be further compromised in the name of ideological causes, whether these are driven by paranoid dictatorships or the ever-increasing wish of so-called democratic countries to surveil their citizens.
Let's now look at one company that is established today, and try to understand whether we have preserved the multi-stakeholder progress model on the Internet that started with a handful of entrepreneurs, scientists and engineers two hundred years ago, or whether the commercial sphere is moving towards single-handed control. Facebook is a large, well-known social media platform with dominant market share. With the effective death of MySpace by 2011 (despite a relaunch in 2013 with mixed results) and the closure of Friends Reunited in February 2016, Facebook now dominates the social media landscape with over a billion active monthly users. Facebook is an example of where the multi-stakeholder model has converged into the single-stakeholder model.
Does this mean Facebook is an example of a successful single-stakeholder model for Internet use? Its dubious recent history casts long shadows. Facebook spent over US$10m on lobbying efforts in the United States in 2015, exerting influence on lawmakers and other political entities over Internet-related legislation. To emphasise this point – this is a non-elected commercial entity exerting pressure on elected political leaders to shape certain aspects of the Internet to suit the interests of the company – not the interests of the general public, nor atruistic or philanthropic groups, but the company, and their shareholders.
Up until 2013, it was impossible to permanently delete a Facebook account. Information posted on Facebook such as political views, personal communications, identity information and more would remain in the possession and ownership of Facebook with no right for the user to demand deletion. Even after 2013, permanent deletion of accounts is possible but it is doubtful whether the data is actually removed from Facebook servers – it is far more likely that data is simply rendered inaccessible, since this presents the most cost-effective option. From a societal perspective, Facebook has also been blamed for high divorce rates – in a survey in 2009, it was estimated that approximately 20 percent of all divorces included some reference to the social platform.
There are many more criticisms that the company has drawn in the last decade, many of which have striking parallels to North Korea – issues relating to censorship, influence, control, privacy and surveillance. With increasing polarisation of the Internet around large, flagship corporations such as Facebook and consequently the reduction of the number of stakeholders in Internet development and governance, basic human rights are at risk of violation. This underlines the need to have a diverse range of controlling entities in charge of the planet's most prolific network instead of moving to a system with a unified controlling party.
Effective governance of the Internet in the future is a doubtful proposition. While social media is thriving, greater tectonic shifts are underway and the future of the Internet looks to be moving at an accelerated pace not just towards mobile but towards wearable and embedded devices – the so-called 'Internet of Things' (IoT). With devices such as the Fitbit monitoring heart rate, exercise, sleep patterns, food intake and even sexual activity, and this data sharable to a mobile app, traditional website and via social media to friends and family, how does a government effectively regulate such a device?
Current tools such as the UK Computer Misuse Act 1990 (with revisions) are beginning to look hopelessly traditional and outdated. Section 3A of the Act, for example, prohibits the supply of articles that may be used to commit offences as defined in the other sections of the Act. Does this mean computer manufacturers may be committing an offence? What if a wearable device such as a smartwatch was used to record video that was later used to commit an offence – does this render the manufacturer and retailer liable to prosecution? The Computer Misuse Act has no reference to IoT concepts such as wearables, or embedded domestic environmental controls such as the NestCam.
With encryption now commonplace in e-commerce and a move towards encryption as a standard protocol, governments are left without means to enact surveillance or control of Internet users. Even with the advanced tools available to specialist government departments (such as GCHQ and MI5 in the UK and the NSA in the USA), as the current Apple debate shows, governments are a long way off maintaining control of Internet activity.
This increasing lack of ability to govern the Internet is a welcome move. Historical examples have shown that hobbling free-market technological progress through state-driven paranoid megalomania simply doesn't work. Printing presses came about despite the government; the early Internet, funded by DARPA, exploded into the public domain. Despite the USA banning the export of encryption software, so did encryption, and progress into technologies such as the IoT in the future will happen despite, not because, of state intervention. Regulatory and governance powers as they exist today cannot be effectively exercised when governments fail to keep up with the pace of technology.
Through examining the history of computing, the present state of governance and the attractiveness of the multi-stakeholder model, it is clear that the only viable path forwards is to maintain the multi-stakeholder model.
As it has always done, commerce will provide the infrastructure. End users will continue to provide the advertising revenue that powers the engine of growth. Although the content of Internet may be tainted by commercial interests, it should remain a domain where ordinary people are free to speak, create, criticise, trial, explore and imagine. It should remain an invention that we can use to communicate with each other, building bridges, doing business, exploring new places. Ceding control of Internet governance to single organisations, whether governmental or commercial, would be disastrous – North Korea provides the example of poor state governance, and the negative impact that Facebook has had on Internet impartiality provides the negative commercial example.
Government does, however have a role. State governance should guide development, in much the same way that banks of a river guide the flow. Governments should step in only when necessary to curb the worst excesses of complete freedom; where existing laws are broken online, the government should support the prosecution of offenders under existing legislation. If commercial interests get too big for their boots, governments could introduce rules or caps to limit the influence that such commercial giants have on the rest of the online population. If governments want to encourage technological progress, then from a governance perspective they would be well advised to stand aside and let today's entrepreneurs, scientists, tinkerers, academics, hobbyists, experimenters and ordinary users provide the driving force.
We are some way from the Internet transforming from a free-market commercial Utopia into a philanthropic reflection of the best aspects of humanity, driven by altruism, enthusiasm, compassion and a wish to improve the human condition. It may never happen. But over-regulation through consolidation and seizure of Internet control is not only the wrong answer to a question that should never have been asked; it would be disastrous, severely limiting the economic, social and cultural outlook of the entire world.