In part 1 of this essay I used Benjamin Constant's characterisation of the modern, individualised liberties as being dependent on the republican liberty of collective self-determination to characterise the ways in which technology can be seen to be simultaneously freedom enhancing while also dauntingly threatening. The progressive tech-topians, recognisable today as they were at the start of the industrial revolution, do not see either how hyper-individualism might lead to an atomised, dominated subjection or how the new facility for community-making might generate the tyrannies of society from which modernity promised to liberate us. This second part of the essay elaborates on these dangers. A final third part will emphasise the inescapably political and collective task of preserving liberty.
- The clouds gather
- The tyranny of society
- Modern/Libero-Phobic: Kafka and Orwell
The story told by Zittrain (2008) and Benkler (2006) is of the netowrked computer as a tool of the freedom of the ancients. It is hard not to be taken up in the enthusiasm. The wired world enhances both the freedom of the ancients and the freedom of the moderns. How can we not be at the point that Mill thought the railway and the printing press had brought? The world, and not just England, can now become that mythical Athens that Mill so desired, surely?
Zittrain (2008), however, is not a Millian optimist. His very ambiguous title tells it all: ``The Future of the Internet (and how to stop it)". Stop what? The Future or The Internet? Zittrain (2008)'s is the forecast that, soon, the Internet could end up as the first "self-closing open system". The impetus for closure comes, for Zittrain, from the wrong-headed approach to solving the problems that open systems inevitably, he argues, throw up: 2
The self-closure of generativity.
Zittrain (2008) argues that companies will be tempted to ``lock-down" open systems in the face of malware, spyware and privacy breaches. Governments will be called on to regulate, imposing privacy laws (which they will exempt themselves from, naturally). The Internet will be balkanised as telephone companies create ``safe havens", networks that bar the use of peer-to-peer file-sharing, privilege one sort of traffic over another, and generally tether the appliances and software to avoid the ills and risks of openness. The game consoles and smart phones are a Trojan horse. Sony and Microsoft have designed their gaming platforms to be walled gardens of computing. The iPhone and other ultra-portable computers are an alliance of consumer electronics with telecoms monopolies, none of whom seem to want to make the business ``mistake" IBM made in outsourcing and opening the critical pieces of the PC.
This is a story of how the Internet moves from being a privileged domain for the freedom of the ancients to being a battleground for the libero-phobia of the moderns. He describes the tendency towards a move from the bottom left quadrant to the top right. Zittrain's own schema is of a move from ``emergent hierarchy" to ``top-down polyarchy" as a principle of organisation.
Keeping my distance from a useful taxonomy.
Zittrain's solution for countering this trend is to return, as often as possible and with as much imagination as possible, to the expert self-regulation that solved the Internet's problems in the past (Curzon Price, 2008). Before government regulation, let us find ``communitarian" solutions. Zittrain, for example, suggests that we try to address real privacy concerns in the same sort of way that the early web controlled crawlers with robots.
As we travel with phones, making payments, with store cards we leave a trace through the ether as surely as any click-trace through the web. Early on in the history of the web, automated services--for example indexing services--started crawling web servers automatically for information. How could a webmaster signal that content should not be indexed? For example, on openDemocracy.net, we want our editorialised articles to be indexed, but not our free-form forum discussions. When a Google search returns an openDemocracy address, we want it to be for a piece of content that we have vetted, as part of our brand management.
The solution to the problem--the creation of the robots.txt de facto standard--is described by Zittrain (2008) :
Controlling crawlers with robots.
This could be a template to decentralised, emergent, locational privacy protection. My own personal "robots.txt" could instruct cell-phone operators, ticketing agents, credit card companies, what information about my comings and goings was sellable and what not. I could specifiy who it could be shared with, and with what degree of anonymity. I would want my medical record available to hospital emergency rooms in extremis, but not to my employer, for example.We should as much as possible try to stop the accumulation of information by inventing the right expressive means to do so. Such ``easy wins" should not be passed over.
Zittrain (2008) sees the emergence of "robots.txt"-style solutions as the victory of hierarchic bottom-up organisations versus poly-archic top-down organisations (Curzon Price, 2008). His exhortation is that we should, wherever possible, look for solutions to information processing problems that live in the ``hierarchical / emergent" corner of his taxonomy.
The tendency of the Internet to move from ancient freedom to modern control is not the only worry with the ancient spirit of the modern networked technology. Constant was clear in identifying the tyranny of society that had also come with the freedom of collective, discursive self-determination. The counterpart to the freedom of collective determination is that the collective can call you to arms, ostracise you, even put you to death without recourse or rights. When the group is small and the community tight-knit, its power extends far beyond law, and can be a real source of oppression.
You won't need to remind Star Wars kid of this, still suffering depression (surely the modern form of exile) after the largest school-yard humiliation in history. The unfortunate boy filmed himself using the school video camera doing a Star Wars light sabre routine with a broomstick. A mixture of narcissism, hubris and, most of all, the dramatic irony for the audience of knowing that this person did not know he would be watched by millions ...He did not delete the sequence; some peers posted the video on the web; the public shaming of the notoriety may never leave him.
There are many other anecdotes which illustrate the way in which the recreation of tight community is also a loss of the protection of anonymity.3
Painfully learning the new rules.
Is this new tyranny of the wired communities a passing phase? Something that we will learn to navigate once the unfamiliarity of the new spaces passes?
When cameras became a cheap mass consumer good, just before World War I in the UK, there was a national privacy scare. Anyone was at risk of being captured in a private act. And the risk is with us still, of course, although a combination of copyright, libel and press-privacy laws allow us mostly to navigate the new contours of privacy. David Cameron, the leader of the UK's opposition, had to spend a good deal of family money to buy the copyright of a photograph taken of him during his student days as a member of Oxford's notorious Bullingdon club. Employers regularly check Facebook and MySpace to find the real person they are about to employ. In a mirror image of the Bullingdon club example, celebrities are exactly the people who cultivate their private lives as a business model: they generate monopoly private information in order to sell it to an eager public. There was an eighteenth century version of this trade: an enterprising late eighteenth century theatre entrepreneur in London took the muslin curtains off the boxes in his newly launched theatre. The mob crowded in, not so much to see the spectacle but more to rubber-neck at the much more attractive soap-opera of the wealthy in their boxes. But the business model was not right: the wealthy got nothing much from the sale of their privacy and left for more traditional venues. ``Hello" and ``OK" would perfect the model by buying monopoly access to the lives of the ``Pipol" [People] (as, in a lovely inversion of meanings, celebrities are known in France).4
The similarity between the patter on a facebook feed and the gossip of "Hello" is striking . It seems to me likely that the popularity of exhibitionism is quite likely in part due to the exhibitionism of our celebrity heroes. Their status is measured by the column-inches devoted to them, so the democratisation of print-space leads to a proliferation on ``micro-Hello" publications. Facebook is to ``Hello" as the academic blogosphere is to the opinion columns of the ``mainstream media". Doing what the stars do, be it intellectual or social, makes us all a bit more like them. The whole point of MySpace and Facebook is that they are public, and give each the opportunity to be a bit more like the role model in the abandonment of privacy. And just like the academic blogger versus the WSJ columnist, the main difference is the first do it for free.5
``Hello" is the Pipol's Facebook.
A closely related danger to freedom from the tyranny of community--possibly to the institutions of liberal democracy that have more or less upheld freedom for 150 years--is the argument made by Sunstein (2007) that the new facility for niche information provision also fragments communities, allowing each of us to live in narcissistic halls of mirrors where we face no great challenges to our views and opinions. There is the possibility of a radical communitarianisation of news and opinion that will eventually, argues Sunstein, undermine the ``broad tent" institutions that forced us as nations to seek compromise and agreement. This is a story of the Balkanisation of politics because the economics of the production of knowledge and opinion no longer forces us to share costs with those we might disagree with. Sunstein (2007) produces many examples of the danger. My colleague Felix Cohen made this short film about the community of "Vaccine Deniers" on YouTube to illustrate the danger. The power of the example, of course, comes from the potential public health externality created by the development of these credible but not belief-worthy communities.
Vaccine denial: echo chamber with externality.
Nozick (1974), in the ``Utopia" part of the book, considers what an individualistic approach to community-formation might be. He asks us, as a thought experiment, to consider a world in which each individual can will into being all other individuals; individuals so willed can either opt to stay in the world imagined by others, or exit to another world of their own willing; the process is imagined to continue until a stable configuration--if there be one--is found. If there is an equilibrium, then by construction we have a world in which ``none of the inhabitants of the world can imagine an alternative world they would rather live in" (Nozick, 1974)[page 299] which has the stability property of being arrived at in this Utopian way.
When wondering how far this Utopian construction might be from a feasible realisation, Nozick points to the limitations that include the fact that groups impinge on other groups; that it is costly to discover groups that one wishes to join, and one might stay in sub-optimal groups for fear of not finding a better solution; that communities might actually try to restrict the freedom of choice of members in order to perpetuate themselves.
The first and the last of these ``failures" of Nozick's individualistic communitarianism are the most troubling in the context of the re-communitarianisation of knowledge-making. The first is a concern, in our context, that the belief-making of one group will affect another group. The vaccine deniers are a case in point. Sunstein (2007) provides evidence that the last point is a danger: that the ``hall of mirrors" entraps people into solipsistic world views and amplifies the differences between groups.
Communities create poles.
Mark Hunter, a professor of media a INSEAD, argues that the commercial future of news is represented by both Rupert Murdoch and Michael Moore. Both of them know how to give an audience a sense of itself. The quickest way to do this is through polarisation, the exaggerated invention, even demonisation of the ``other". As the fixed costs of community creation disappear (a printing press, a distribution network), society can fragment into many non-communicating shards. The return to ancient Greece was not meant to also repeat the Peleponesian wars. Posner makes the argument that this communitarianisation of news might be a good thing: it avoids the dead-hand of orthodoxy stamping out the diversity of views that it is socially best to express, even if only to better monitor.6
The argument, however, relies on there being not-too-great a feedback between the encapsulation of a community in a publication, or blog, or forum, and the path that this community takes. The strength of the worry Sunstein raises comes from this dynamic effect: the types of communities you end up with are dependent on the technology used to keep them together and define them. The Rwandan massacre, for example, would not have been possible without Radio Mille Collines. The Kenyan violence of January 2008 was made possible by SMS.
Letting off steam?
There is a resurgence in the tyranny of the group just as there is in the power of collective self-determination. The two are inextricable. Liberal Whigs have to hope that re-communitarianisation does not have to be Balkanisation. It is as if the ``National" phase of the evolution of the just society was a dry-run for its global version. Re-communitarianisation, with global communications technologies, is part of the unpicking of the unitary Nation State, with the adherence to the tribe returning to many overlapping communities, some digital, and the functions of the State being spread both above and below the old Nations. The economics of community-making are an important part of this large-scale historical process. To make it a historical evolution, and not another return, we should keep our eyes open to the the new forms of the ancient social tyrannies.
We have considered two types of danger so far--first, the privacy and Sunstein (2007) effects point to the transformation of ancient liberties into modern communo-tyrannies; second,Zittrain (2008) and Curzon Price (2008) point to the sliding from the new communo-freedoms to the modern stifling technocracies of government and other natural monopoly.
Possibly more dangerous than all of these are the forces that transform modern individualistic freedoms into the hyper-modern nightmares imagined Kafka (1925) and Orwell (1949). This is the story of the harnessing of technology into a database State. A little like nuclear power, so attractive in the 1950s and 1960s, accumulates waste that today appears to pose small risks of very large harm, so centralised databases are stockpiling a resource which has a small probability of becoming highly noxious.
The behaviour, identity, past and reputation of each us is now reproduced around large numbers of state-level and corporate databases creating dangerous concentrations of personally identifiable information (PII).7 The dangers are of two basic types, call them Kafka-esque and Orwellian.8
The Orwellian, in which the State accumulates information for direct purposes of control, is the most obvious and dramatic. Take the example of the hapless mechanic from Stoke-on-Trent .
An innocent's arrest and DNA forever more on the record
Darren's credit ratings, job prospects and life chances have been permanently affected by his record on the UK arrest and DNA databases. As technocracy would have it, this information gets shared between more and more government agencies and even in some cases to private sector subcontractors of government. Of course, the data also gets lost and may find its way into the public or criminal domain. The impacts on lives and liberty of such Orwellian events are clear, and our ability to influence through policy is also clear.
The UK, together with many states since the start of the ``War on Terror" (WoT) have increased the powers of surveillance available to it. There are real threats, and some surveillance is necessary to protect citizens. But it is important to understand the Orwellian habits that very naturally come to a State. Here is an example from Scotland:
In Falkirk, which used Ripa [the Regulation of Investigatory Powers Act] 380 times, citizens could be spied on for noise nuisance, littering, if they were suspected of driving a taxi without a licence, for breaching the smoking ban and if their expense claims were thought to be exaggerated.(The Sunday Herald, July 18 2008)
Jozef K., the anti-hero of The Trial, never knows who is accusing him or what he is accused of. He has no control over the information that is disseminated about him or any means of recourse. The story of Jozef K. has become the archetypal story of alienation, one in which the modern atom has no understanding of the forces at work at him. The case of Mr Bunce , a story, with a somewhat happier ending, hints at a modern version of the story where the information and crime have escaped your control--it certainly shows a very concrete example that those who have nothing to hide might nevertheless be caught up in nightmarish information-traps. The police knock on the door. You are accused of having bought paedophilic pornography over the web. You are innocent, and unlike Joseph K. you certainly understand the accusation here, but what have you done that relates to it? Why are you being accused? What piece of information about your life has been disseminated which leads to this point?
A modern-day Job
Mr Bunce comes out of this sounding like a saint: his salary quartered, his family forgiven for disowning him ...but many others would remain more resentful.
Commercial databases are increasingly the repository for Privately Identifiable Information. The case of Yahoo's Chinese blogger debacle is an example of commercial data-gathering being put to State use.
Corporate and State databases cooperate, 1
Information flow from state to corporations is increasingly common. For example the UK government's "Tranformative Government" agenda includes offering corporate access to the National Identity database (for example, for background checks on criminality). Sub-contractors to government have ways of accessing data--where this can go from multinationals running IT systems to small operations running local authority wheel clamping services. The film of the useful UK Department of transport web page on Who we share information with and why provides a small glimpse into the complexity of the multi-way flow of information. Can every outfit identified as a worthy recipient of information be counted on here to have a careful, responsible attitude towards the data? How much will lead to stories like that of Darren, Bunce or Shi Tao?
What your driving says to whom (follow link to film)
The accumulation of ever-more personal databases creates threats that are both Orwellian and Kafkaesque. A common response in this age of the War on Terror is that a Hobbesian state, one that takes seriously its duty to ensure the safety of its citizens, needs to employ the necessary means against an enemy that has found modernity's Achilles heal. An enemy that has learned to take advantage of the open society will prompt counter-moves from the primitve, Hobbesian state, that will close some of the loop-holes. Hence the justification for extensive surveillance as well as ``collateral damage" in innocent lives destroyed.
A report by the Mail on Sunday on CCTV surveillance by police in Shenzhen and Leeds illustrates the point. In Leeds, police prove the innocence of a supposed wife-beater with CCTV footage ...a petty thief is caught after the police thought they would have to let him go ...in Shenzhen, crime rates are down 10% and detection rates are up thanks to a state-of-the-art surveillance system. Here comes the downside:
anti-government demonstrators are identified and picked up for questioning -- and images of those still ``at large'' are posted on public information pillars ...9
The Hobbesian state has its attractions. But we have a sort of ``surveillance trap'': how to ensure that information is used only for the good purposes of fighting crime, and not for the bad purposes of manipulation and discrimination?
It is very hard for a modern mass state to carry out extensive Hobbesian information processing without also, to some groups, becoming Kafkaesque. Take the (hypothetical) example of a young male Muslim rounded-up in a terrorism prevention operation. He is innocent (known to him, unknown to us), and has been arrested because he is a friend of someone who is not. During interrogation, it becomes clear that the State has a huge amount of information about this person. He is eventually released without charge. He returns to his community knowing that the authorities follow him closely and that he is now on a national DNA database, and that his arrest is a fact viewable by all manner of official departments (and even private sector companies). When he does not get a job; when his mortgage is refused; when he is burgled ...a suspicion has to fall on the involvement of the state. Is this a state he is likely to cooperate with? is he likely to inform against a genuine threat that he knows about? The Kafkaesque State is the one that creates sub-cultures of mistrust in which real threats can find protection and thrive.10
We can view the Kafka effect as an externality of information acquisition. As long as the process of acquiring information generates "false positives"--cases where the informational inference is incorrect--then it will invite false positives in the other direction: inferences by the citizen that the state is involved in shaping outcomes even when it is not. So the often well-intentioned process of collecting information for Hobbesian purposes produces a side-product, the false positive, which creates a paranoid sensitivity. Philip K Dick offers a particularly gruesome twist to the basic plot in Dick (1956), where information processing by the state has got the point of forecasting and preventing crime. The Kafkaesque impact of information collection is pernicious: if a state gives you reason to believe that it is falsely and obscurely accusing you, then your trust in that state evaporates. Any Hobbesian justification for freedom-reducing measures should always account for the ultimately freedom-destroying loss of faith in the State that the Kafka effect produces.
Of course, the same piece of technology in Leeds is less of a worry than it is in Shenzhen, because civil liberties are more respected in England than in China. Trustworthy states should be allowed to take Hobbesian advantage of technology, and if we want more surveillance, we should start by making, through politics, more trustworthy states.
We are building corporate databases of personal information on an unprecedented scale. Much of this is coming from the advertising-financed "free lunch" of Web2.0 technologies and services. The advertising model on which these services rely for information might in itself be benign despite the possibility of deep but hidden effects of advertising. However, there is the clear possibility that highly targeted advertising may contribute to a background belief in a Kafkaesque state. When my email offers me a link clearly suitable to my current state, I become habituated to the notion of being watched.
However, the databases it creates will themselves be irresistible to State agents. The psychological irresistibility of the free lunch is accumulating a large and risky liability -- like the promise of nuclear power "too cheap to meter" has created a stockpile of nuclear waste whose risk we will always have with us. It creates a datamine that acts a a magnet to states and organised crime. As I will argue in the conclusion, the worst of it is that the model that accumulates personal information in exchange for web services is not even necessary to provide the services that we are now enjoying for free.
To understand the process by which these databases are being accumulated, consider the ill-fated case of Gator . Gator was an application that offered to fill in web forms for you. Very nice functionality. But Gator also--and not very transparently--relayed all information about web surfing back to HQ. HQ knew the browsing history, purchase history, inferred sexual preferences, film tastes and much else about its users. The Gator client watched and told all that happened between keyboard and screen. Gator could have been much more malicious than it was with all this information. In fact, its team of crack statisticians applied themselves to the simple task of predicting which of an available number of advertisements a given user at a given browsing moment was most likely to click through. So, if I was booking a holiday to Hawaii--as I remember doing when I last had Gator installed in 2003--as soon as my flight was booked, Gator served me an offer for a condo and a car; Gator's statisticians had built up quite a profile of me, and could offer sun-suits for the children, prescription goggles for the snorkelling, a guide book and a bird-watching tour for my wife (how had they figured that one out?).
Gator was achieving click-through rates with its profiling that were ten times what Google could offer. Gator's accounts managers could offer corporate customers highly complex advertising campaigns: ``if a middle class mother has been looking at car web-sites and arrives at GM, make the home page from soft green tones and emphasise safety; if it is a rugged young male, serve up the Hummer ..." (pre-crunch, all this).
All inoffensive, even helpful, you might think. But it is easy to imagine some more problematic cases.11
The corporate databases that Web2.0 is creating can be agents of Kafka and Orwell effects just as much as government databases. Already today, the large web-sites have entire divisions devoted to dealing with the subpoenas served by courts. The information exists, and will be used in accordance with the laws of the land ...as Chinese bloggers have found to their cost.
The battle field
If our old liberties and rights did a good--though restricted--job in the late 18th Century, with State power where it was, they certainly need reinforcing with state power where it is today. In the UK, for example, Habeas Corpus has been restricted to a huge extent--police can hold suspects without charge for 28 days, moving to 42. And this in an environment where discovering information about suspects can be done at the cross-tabulation of a handful of databases. If it took 24 hours to charge a suspect in the past, the increased efficiency of information gathering would suggest that it should now take less time to charge, not more. It is hard to avoid the conclusion that the War on Terror has been a boon to the institutional interests of technocracy (see (Barnett, 2008)).
Barnett, A.: 2008, 42 days: ``an abundance of caution'', openDemocracy .
Benkler, Y.: 2006, The wealth of networks, Yale University Press.
Boiteux, M.: 1956, Sur la gestion des monmopoles publics astreints a
l'équilibre budgétaire, Econometrica .
Curzon Price, T.: 2008, From zittrain to aristotle in 600 words, openDemocracy .
Dick, P. K.: 1956, Minority report, Fantastic Universe .
Hayek, F. A. v.: 1982, Law, Legislation and Liberty - A new statement of
the liberal principles of Justice and Political Economy, Routledge and Kegan
Kafka, F.: 1925, The Trial, Project Gutenberg EBook.
Nozick, R.: 1974, Anarchy, State and Utopia, Basic Books.
Orwell, G.: 1949, 1984, Secker and Warburg.
Solove, D. J.: 2007, “i’ve got nothing to hide” and other
misunderstandings of privacy, San Diego Law Review 44.
Sunstein, C.: 2007, Republic.com 2.0, Princeton University Press.
Zittrain, J.: 2008, The Future of the Internet (and how to stop it),
Princeton University Press.
- ... 2)1
- Many thanks to all the people who have commented on early drafts of this paper--Selina O'Grady, Graeme Mitchison, Victoria Curzon Price, Anthony Barnett, Jonathan Zittrain, David Hayes, Jeremy O'Grady, Stefaan Verhultz. This paper owes a great deal to a seminar funded by the MacArthur foundation in March 2008, "Credibility in the New News" in London. Many thanks to Kathy Im and Elspeth Revere for making that gathering and space for thinking possible. I presented a version of this paper to the Annual Meeting of the Mont Pellerin Society in Tokyo in September 2009. It was a personally emotionally charged occasion, being the child of two members of the society while feeling uncomfortable with most of the positions taken by its members. On the question of the authoritarian state, however, we were on common ground - at least at some level of abstraction.
- The parallel between Zittrain's worries about the Internet and Hayek's in the ``Road to Serfdom" are clear. In both cases, there is a call to action against specified dynamics from within that threaten all that is best about the system. The Great Depression provided the impetus for the sorts of government actions that Hayek thought would undermine the market economy, just as Zittrain thinks malware, spyware and the abuse of private information threatens the sorts of control that will undermine the open, creative, re-usable Internet.
The relationship between anonymity and the freedom of the moderns is very strong. It has its counterparts in the modern notions of anomie and alienation. Hayek picks up on the relationship in his discussion of the exercise of market power through price discrimination:
It is probably not much of an exaggeration to say that almost all really harmful power [...] rests on this power of discrimination because it alone, short of violence, gives them [firms] power over potential competitors [...] Though the majority of people may still be better off for the existence of such a [discriminating] monopolist, anyone may be at his mercy in so far as the nature of the product of service makes aimed discrimination possible and the monopolist chooses to practice it in order to make the buyer behave in some respect in a manner that suits the monopolist [...] Since the power of the monopolist to discriminate can be used to coerce particular individuals or firms, [...] it clearly ought to be curbed by appropriate rules of conduct. (Hayek, 1982, Vol.IIl, page 84)
- However did the celebrities become ``les Pipols"? There is a real cunning of language in this appropriation--a recognition that there is nothing between celebrities and people except for celebrity.
- It brings to mind Keynes' joke that GDP falls when a bachelor marries his maid--what was previously recorded as a monetary transaction is now subsumed in the un-measured domestic economy. Similarly, every journalist replaced by an academic blogger lowers GDP without necessarily reducing welfare.
There are real success stories of citizen journalism bringing to public attention stories that might otherwise have been covered or buried. The great pet food scare of 2007 may be the clearest example, (especially because it is an example so devoid of politics). All over North America, pet owner forums started reporting that their cats and dogs were ill. Forum and blog members thought this was all too much for coincidence, and traced the problem to a single Canadian manufacturer. The pet-owners did all the investigation and coordinated their work through their blogs. As Jay Rosen says, they moved from the demand side of news to the supply side. The old-media caught up with the story once the journalistic work had been done at the grass-roots, by these consumers turned producers.
In terms of political positioning, the technology here is a tool of negative freedom: consumers can aggregate information and protect their rights as purchasers against the previously superior informational power of the producer. This is an enhancement to the normal freedoms of civil, contractual relations.
- And quasi-PII--it is often possible to combine databases none of which have PII to, in the overlap of data, personally identify individuals.
- The important distinctions between the Orwellian and Kafkaesque abuses of information is made by Solove (2007)
- Just as Jozef K. was Jewish and felt anyway on the edges of Viennese society, a ready target of radicalisation, so those who come into contact with the most Kafkaesque aspects of our states are those likely to be close to today's radicalisable edges of society.
- Obviously, the profiling could be used for old-fashioned price discrimination. The old headache for tariff setters trying to implement Boiteux-Ramsey pricing (Boiteux (1956)) for utilities was how to achieve a ``separating equilibrium"--how to make sure that the simple set of tariff choices on offer would distinguish customers by their elasticity of demand. No such constraint with Gator's profiles, where the identity, or at least identity-type of a customer could be determined by the logs of browsing history. ``TCP is likely to click through for car hire, so don't offer the special deal ..."
(For the story, Gator was hugely profitable. It was nearly sold to Microsoft for large amounts, until Redmond, in its due diligence, broke off discussions. Gator changed its name, and now trades honestly as Claria).