Several events recently have led to demands that the internet should be better controlled. There is concern about web sites promoting paedophilia, terrorism, drug use and violence, but also concern about the stifling of free speech. Equally, there is the conviction on the part of some that governments are over-diligent in trawling email and social media, whilst others are relieved that the forces of law and order are taking steps to defeat the ‘bad guys’. Prime Minister David Cameron is now suggesting that legislation may be necessary if all voluntary measures fail.
The concern is understandable and legitimate. Yet most of the discussion on these subjects ignores certain facts, and I should like to begin by looking at those facts.
First, in essence the internet is simple. It is a network, mainly of fibre optic cable, which forms a transport mechanism for packets of data, just as the postal network is a transport mechanism for the physical equivalent. Each data packet is labelled showing where it is going and where it is from; the first allows it to arrive while the second allows the packets to be reassembled into a document. This labelling means the flow of traffic can always be monitored, unless the origin of packets is deliberately disguised by sending them via a ‘proxy server’.
Second, we need to understand certain features of the web, that massive collection of documents linked by the familiar ‘http’ protocol. Other protocols are used for email, file transfer, telephony, video and so on, which also use the internet but are not part of the web. (Confusion is caused because they may be made accessible via the web through e.g. webmail, Skype, file sharing sites or YouTube).
Third, within the web, not everything is public. A website can be invisible if it is not associated with a domain name, or if the content is generated only when called for, or if it is hidden within a ‘virtual private network’ (VPN) such as those used by businesses. The internet therefore has two parts. The public internet is made up of visible web sites (including social media) along with services like email that piggyback onto them. Sites are available via the domain name system, and search engines can catalogue them. The hidden internet is very different, it can only be accessed if you know where to look.
Fourth, even within the public internet, only text can be read easily by computers. Graphical, audio and video files generally need to be viewed or listened to by a human for their meaning to be revealed. Monitoring non-text media is therefore very difficult.
Fifth, there’s no free lunch. Once we have an internet service provider (ISP) a world of goodies is handed to us on a plate. Post pictures on Facebook, browse YouTube, find information on Google; but these things are not free, they are driven by commercial interests, and are eventually paid for in the prices of goods and services. In particular, billions of our mouse clicks are recorded every day and collated so that businesses can market to us. It follows that the public internet has to be pretty open if we want everything for nothing.
Sixth, add in the human element: we use weak passwords, put confidential information on social media sites, use insecure networks and click on rogue email attachments. Hardly anyone encrypts emails. In addition to this carelessness, there are many hackers out there. The only sensible conclusion is that anything sent by email or submitted to a web site must be considered an open book. This includes social media, blogs, chat rooms and messaging.
We are now in a better position to tackle the difficult questions, beginning with the concern about web sites that promote illegal activities such as child abuse and drug trafficking.
If a public web site is promoting illegal activity, preventing access to the site is remarkably easy. Web sites can be set up anywhere in the world, and there will always be countries willing to ask no questions; but the packet labelling that allows the internet to function also makes it easy for an ISP to block rogue sites. Some autocratic regimes use this method to exclude political criticism, but there is no technical reason why the same methods cannot be used to exclude other material, under democratic control.
The issue, then, is not technical but cultural. Do we have the will and financial commitment to counter illegal material on the public internet?
There is concern about the amount of monitoring of the internet that governments may carry out. However, the sheer quantity of data involved means that any indiscriminate monitoring must be automated. This is doomed to failure, because even on the public internet it is easy to set up a ‘front’ organisation with legitimate operations. Transfer of secret information can be done by splitting it up into emails, phone calls, files to be downloaded, graphics, video clips, tweets, and even old-fashioned post. Much of this data cannot be read electronically and all of it can be routed via intermediaries; in any case, detecting individual messages would not betray the overall content.
Even if the ‘bad guys’ were careless occasionally, any small sliver of useful information will be buried in a massive haystack of trivia. Frankly, I don’t care if some intelligence officer knows that I have a meeting with Fred Bloggs at the Holiday Inn on Friday. If I did care I would use a VPN and encryption. The commercial data holders know far more about us than any government, and again I shall return to this point.
Infiltrating the Hidden Internet
In order to find data on the hidden internet, as I have said, you have to know where to look. Those who want material on certain subjects such as terrorism or paedophilia do know where to look, because they form networks that can communicate internally by a variety of methods, including personal meetings. Once ‘in the know’ they can also use highly confidential facilities such as Freenet and Tor, which bounce packets around several proxy servers that add more and more security and cover the tracks of the originator. The technology is not expensive. Payments via these services can be handled with Bitcoin, which is anonymous. Following the money is, therefore, increasingly difficult.
In short, preventing the passage of information or trade between people who actively wish to engage with it is impossible by electronic means alone. As well as the monitoring and blocking options that can be employed on the public internet, infiltration of the various communities that promote illegality is also necessary, which is a labour intensive and costly activity. I am not sure there is a sufficient cultural mandate for it, partly because of internet mythology.
The Medium has become the Message
If you compiled information about yourself, your family, your job, past relationships, political and religious allegiances, prejudices, purchases, likes and dislikes, and published it in a newspaper, you would not complain about the newspaper’s lack of privacy if something you had written came back to haunt you. Yet that is exactly the sort of complaint made about the internet, especially the public side. We feel the internet should allow privacy as of right, so we ignore the fact that it doesn’t.
Even that ‘should’ is ambivalent. Suppose that you met some friends in a pub and were severely injured when violence broke out. You later learn that many people in authority knew this was likely to happen because rumours had been picked up from gossip and local contacts, yet no one was able to dig deeper because of the law on privacy. Suddenly, privacy does not seem such a good idea.
The problem here is that the internet as a medium is unique in the freedom it conveys through its technical innovation. However, a mythology has developed that we can roam at will, at practically no cost, and with no risk to our privacy; we tend to think that what we do on the internet should be private, but what others do should be capable of scrutiny so that the ‘bad guys’ can be caught. When spelled out like that the fallacy is obvious, but the myth of freedom has a strong hold. The Medium has become the Message.
The key to our future sanity is to break out of the mythology. As individuals or organisations we are faced with a simple choice – either accept the limitations (including loss of privacy) of the public internet or pay for private facilities. Having bitten that bullet we can then recognise that all the other principles historically applied to public media can and should be applied to the internet. The key principles are as follows.
Censorship occurs in all civilised societies. I think most would agree that hard-core porn, instructions for making terrorist devices and images of torture should not be available in the daily newspapers. Why, then, should it cause cries of outrage when steps are taken to keep them from the internet? The outrage often focuses on civil liberties – but what has happened to the civil liberties of children who have been groomed by paedophiles, who will always live under that shadow? Or those who die young because of the drug culture? Or the victims of terror attacks?
Freedom of speech is a key to democracy, but it has long been recognised that incitement to criminality is not subject to that principle. Neither is grossly offensive discourse, slander, libel or defamation. Yet loud voices claim that the internet should have a unique respect for anonymity, which prevents all sanctions against such abuses of free speech. The vital concept of the publisher is forgotten. Publishers of books are obliged to identify themselves. In principle, internet publisher identity can be found from registrars’ databases – but internet publishers are allowed to use a third-party proxy service to preserve anonymity.
Responsibility should apply to the internet just as to all other aspects of life. Lawmakers are responsible for scrutinising those things that happen in our society that affect the common good, yet when it comes to the internet there are fragmentary efforts but often with little conviction. Until, of course, yet another high-profile child abuse case arises or a terrorist attack takes place.
In order to apply these principles within a sovereign state such as the UK all that is needed is the enactment of minimal new legislation, or enforcement of existing law, along the following lines:
1. The ‘level playing field’ principle should apply: if publication of any material is illegal in print or broadcast media, it should also be illegal if carried out on the internet.
2. The registered owner of a domain should be legally regarded as the publisher of any material hosted on that domain, except on social media sites where the person posting the material should be regarded as the publisher.
3. Registrars should be legally obliged to provide the name and contact details of publishers to lawful authorities on request. Owners of social media sites should have the same responsibility to disclose publishers of posts on their sites. In neither case is it necessary for publisher information to be publicly available, only an obligation to disclose to lawful authority is necessary.
4. Owners of web domains have absolute discretion as to whom they use as registrars, and if they use a proxy as the visible registrant they can choose one who will disclose their identity to lawful authority. If publisher information cannot be obtained, therefore, that fact should be taken as justification for blocking the site. The owner had other options available.
5. ISPs should have an obligation to block any web site when requested to do so by lawful authority. Further, it should not be a criminal or civil offence for anyone to block a web site for which it is impossible to trace a publisher. Action could then be taken against such sites before any illegal material is placed on them.
That last proposal is bound to be controversial. Does it not imply a presumption of guilt, contrary to the principles of our judicial system? No, it doesn’t. In most countries a car must carry number plates so that the owner can be identified in the event of misdemeanour. This is not seen as an infringement of civil liberties, merely a sensible precaution to allow enforcement of democratically agreed law if necessary. A cloaked anonymous web site is simply the cyber equivalent of a car without any number plates.
My proposals are no more draconian than those applied to every other sphere of life by most sovereign states. Yet they are bound to arouse cries of ‘neo-prohibitionism’ and the like, which in turn will cause hesitancy on the part of political and cultural leaders. Why?
The first reason for such hesitancy is that since the 1960s individual freedom has become increasingly elevated as a first principle, as against any kind of conformity to the community or nation. The huge benefits of the internet have included the opening up of individual freedom to millions of people, since communication is a prime weapon against oppression. That is all to the good. Unfortunately, the apostles of freedom, under the influence of the internet myth I outlined above, are reluctant to draw any boundary, even though this has always been necessary for other media.
The second reason for hesitancy is that many people who are no more than averagely wicked nevertheless have a great emotional involvement in the internet as a gateway to pornography, drugs for social use such as marijuana and cocaine, information on sexual activities, extreme political views and so on. Some of these activities are legal, some are not, but a substantial proportion of the population partake in them. There is, therefore, a fear that the freedom to walk on the wild side in these grey areas will be lost.
The third factor that blunts action is the cost of taking effective measures. I am glad that David Cameron is taking an interest and he makes some excellent points, but bludgeoning the search engine providers about search terms cannot work, and if carried too far would clash with their commercial interests. The subcultures under attack will simply generate more and more coded ways of describing their activities, ensure that information about their activities is contained in non-text documents, and add to the number of (apparently innocent) gateway sites within the grey areas mentioned above. Issuing diktats, however, is cheap, even legislation is not financially demanding. Enforcement is, because it requires large numbers of people to take part in monitoring exercises, respond to complaints from the public, and institute the necessary legal proceedings.
The issues raised by the internet are, therefore, fundamental ethical ones concerning the rights and responsibilities of the individual in relation to society, and the obligation of our society to shoulder the financial burden necessary for the good of the whole. This is a great challenge, not only to our politicians but to us all.