First Monday
Read related articles on Business and the Internet, Internet communities and Internet economics

Personal boundaries/global stage by Harold Thimbleby

The Internet will apparently solve many problems, but it also creates new ones. We are simply not used to thinking on the scale of the Internet, and our personal views can easily be played out on an unfamiliar global stage, where they may have unintended consequences. Computers, the most logical things we have ever created, are raising serious ethical issues - especially in areas that traditionally have had a low priority in the modern world. We are not used to the global stage.


Tensions and the Red Flag Analog
There are Problems
Platform for Internet Content Selection
Ethics is Based on Physical Reality
Homogeneity - "Tribalism"
Scale has Consequences
The Tragedy of the Commons
Henry George and "Common Good"
Omissions and Simplifications

The world has a communications infrastructure [1]. The most obvious part of this structure is the Internet [2], itself the largest single structure humans have ever built. The Internet can convey telephone calls, radio and television broadcasts, faxes, financial dealings, as well as a variety of new forms of communication between people. There is increasing convergence between communications technologies. Today, shops are selling televisions that can also surf the world wide web. Today, a desk top computer eclipses what NASA had to send man to the moon.

We could be "technological determinists." The technology is amazing (and it certainly is). Therefore it will do amazing things. It certainly will, but the amazement will not be that technology does new things, but that humans do new things together.

If your desk top computer could send someone to the moon, what else have you done with it recently?

The Association for Computing Machinery (ACM), the largest computer society in the world, recently celebrated its fiftieth anniversary. Almost everything it has done to celebrate has raised the issues of human impact, social change, global economics, privacy, and fundamental human rights such as freedom of speech. Their special anniversary issue of the Communications of the ACM included articles like "Education and jobs in the digital world," "Better democracy through technology," and "A new social contract for research."

There have been some exciting changes brought about by communication. The Berlin Wall coming down in November 1989 happened because communication respected no national barriers. Human rights abuses around the world have been exposed because of the ease of communicating out of oppressive regimes.

Ethics is about right and wrong, and how we ought to live. It is fascinating that computers and communications, the most logical things we have ever created, are raising serious issues in ethics. The issues raised, such as free speech, impact the most important human rights and duties, yet are ones that traditionally have had a low priority in the modern commercial world.

Tensions and the Red Flag Analogy

The Internet has been called the information superhighway. It's interesting to compare one highway with another. Children don't play on motorways. Lorries don't drive on pavements. Cars use indicators when they turn. The Highway Code and the underlying Road Traffic Acts in Great Britain are enormous documents, but society has accommodated the rules and considers them quite sensible. Yet when cars first came on the roads there were tensions.

The Locomotives on Highways Act of 1865 enforced a speed limit of 4 miles per hour all over Britain, except in towns where it was 2 miles per hour. Cars had to be preceded on public roads by a person carrying a red flag. Looking back, we can see this response to the perceived problem as rather quaint. Equally, we can imagine Victorians would think our modern traffic regulations completely unreasonable and a major threat to their way of life.

Nowadays, we don't have red flags (we have lots of road deaths instead). We live with the problems in a different light, because we can see the advantages of mobility, freedom and so forth. These advantages seem to us to be worth the troubles that dominated the Victorians' thinking.

The lesson is that social tensions do need addressing, but that red flags are not necessarily appropriate. On the Internet, I can see many areas where there are tensions; I shall describe some shortly. Easy as they are to wave, I am not sure red flags are what are required.

Roads also illustrate a very interesting connection between safety and social convention. In Great Britain, we drive on the left. That is because everybody else here does, and therefore it is safest to do so. Unfortunately some other countries are not so enlightened, and this sensible British custom can, if used abroad, cause accidents.

It seems likely that the Internet will, in time, accumulate a long list of regulations and conventions. Some will be rational, some will be arbitrary conventions that smooth relations. The framework that our children grow up to use will be incomprehensible to us, even unjustifiable to us, but that framework will be the glue holding the global community together.

There are Problems

It is tempting to focus narrowly on ethical issues. There has been a furor on children's access to pornography. Such issues are important, but they are part of larger issues that need tackling. For example, there are other forms of information, such as how to construct explosives, to manufacture drugs (things which are legitimate in some circumstances) to instructions on torture (which probably aren't). We may prefer certain classes of people to have restricted access to such kinds of information. Or we may prefer certain sorts of generalised rights, such as free speech, over the problems of some people saying things we consider counter-productive. These are standard ethical dilemmas; but, moreover, the problems as we see them may not be where they are most obvious. It is easy to point to information we think is destructive. It is much harder to change our attitudes.

For example, on the question of nasty stuff (not just pornography) and children, perhaps our notion of childhood is faulty. Children have a period of innocence, which traditionally ended around late teenage years when they could leave home and move into the big world out there. Now, young children in the privacy of their own bedrooms can surf around the world, and the big world enters their private spaces while they are still innocent. Under this view, we either have to keep the big bad world out, or we have to revise our sentimental ideas of childhood.

This is an important debate. How society works out these issues will have ramifications in, say, how schools operate, and how society labels and deals with offenders and victims. Meanwhile the reach of the Internet is extending, and such problems are prominent and motivate change.

Platform for Internet Content Selection

We are progressing by developments in technology, vying with regulation. There is a Platform for Internet Content Selection, or PICS.

PICS claims to be a value-free way of classifying information on the Internet. That is, rather like a library classification system, it does not make a judgement about what is good or bad, it just lets people label information, and use those labels to select or reject what they do or don't want.

How do the labels get attached to the information? Either the writers of the information attach the labels, or labels can be put in independent directories, typically by concerned organisations.

Let's consider two sorts of labels we might want.

First, a label for personal use. My daughter wants a guinea pig, and I worry about what she may find out. I don't know where it is going to lead, so I want to restrict her access to guinea pig sites on the web. If so, I have set myself an impossible task. I estimate that there are 30,000 places I need to classify [3], and there are new ones arriving daily. You may well think this is a silly example (as I do!) but it is not difficult to think of examples that most people think are silly but which seriously affect other people. Those few, concerned people, are not going to be able to use PICS to help them. My guinea pigs are a fun example, but they make a serious point.

Secondly, let's consider a label for general use. PICS can already be used to classify sex and violence, and many sites have been so classified. The lists of classifications, because lots of people are interested in them, have become commercialised. There are several companies operating in this area. Some have classified their competitors as unsuitable for accessing. It is indeed ironic that a system designed to promote one sort of morality encourages another sort of immorality.

There are social and technical problems too. The people who write material that others find objectionable are not always co-operative. My search for guinea pigs would not have found anywhere that put its message in a picture. There are many more techniques that could be used mischievously, and unfortunately the sorts of minds that would create information we may want to control are often the sorts of minds that would resort to such techniques to misrepresent themselves. And there are many costly activities, such as on-line gambling, stores that sell alcoholic drinks or cigarettes, or fraudulent "charities," that cannot be blocked by PICS (the categories are unlimited) but which most parents would be reluctant to allow their children to be addicted to.

Because PICS tries to be politically correct, it offers no agreement on classification. There is no agreed standard on how to label guinea pigs. Had PICS gone down any of the many established classification routes, for example by using Dewey decimals, I could easily exclude undesirable rodents. Moreover, even the second version of PICS doesn't support any library classification system, standard or otherwise.

PICS emerged because the U.S. government was considering restrictive regulations on the Internet. The regulations would have made corporations responsible for the information they carried. PICS very clearly shifts the judgement of content from them to the senders and recipients (and classification bureaux). PICS is a politically-motivated red flag [4].

The obvious use of PICS is to help a child who needs moral safeguards in their selection of material. It is not clear to me that a computer can do this in principle. The effect of encouraging parents and other authorities to believe that PICS is a solution will be to make them less involved, in the limit this will atrophy everyones' moral skills, just when understanding moral issues is starting to have global significance.

To be more positive, classification systems do work in some areas, for instance in helping chose appropriate films to watch - U, PG, X and so on. Classified films are mostly entertainment. Unfortunately the Internet is not mostly entertainment. Clearly the issues are immensely complex. Not only are they complex, but they are not likely to be addressed cross-culturally.

Despite my earlier tongue-in-cheek comments, my daughter has already read enough to know guinea pigs are social animals, and she should get two females, rather than a boy and girl. Even she knows where that would lead!


PICS is an alternative to censorship, or rather, it is an alternative to having censorship imposed. Cultures other than the U.S. have acted differently. Singapore currently employs people to monitor traffic on the Internet entering - and no doubt leaving - the island. At the present rate of growth of Internet traffic, Singapore will need to employ more censors. Soon the state could be paying everyone to censor the material it does not want its citizens to access. It is a policy that can only work while some other approach is developed, or until there is a change in attitude to information. We may be amused at the irony, but the first case of Internet censorship in the UK was a pro-life web site that was required to remove accurate pictures of foetuses some people found objectionable. Thus is censorship used to suppress truth.

Just because criminals use roads, our democracy doesn't tolerate road blocks at every junction. We tolerate road blocks used judiciously and to catch specific criminals; the idea that everyone is a potential crook is abhorrent [5]. That reasoning applies to the Internet: it is abhorrent to set up censors everywhere regardless of how easy computers make it seem to do so.

Censorship comes in different forms, and it helps to distinguish them. First, there is gagging - removal of a person's ability to speak out. Second, there is hiding - making unwanted speech disappear. Third, there is blocking - stopping the thoughts reaching particular people. There are many variations, including methods of distorting what people say to gain more deviously the desired effect. These are all speaker-centred views; there are converse issues for listeners, which are easy to forget. The Internet legend (due to John Gilmore), that the Internet interprets censorship as damage and routes around it sounds appealing until it is spelt out. Suppose there are three people, A, B and C; and that B is censored. "Routing around it" would then mean A and C can communicate freely, because the Internet finds a route around B, who is being censored. Gilmore really seems to be saying the rest of the Internet will work even though censorship happens; that's not quite the same as saying the Internet - as a community - resists censorship. True, the censors who disconnect B to supress B's ideas will likely find those ideas replicated in too many places, in too many nations, to control. (An attempt to suppress information because it is copyright, as might happen with confidential material, could flounder because some nations are not party to the Berne Convention on copyright, and expedients like B giving material to the Swedish parliament makes it public domain.) But false, consider what happens when we take B's, as a listener's, point of view: B has no idea what they are missing, B does not get information, and they are disenfranchised. In fact, that the Internet has found a route around B means that everyone else can carry on without B at all. B has become a non-person.

In summary, the appealing idea that the Internet resists censorship is true - but only just: it is a technical view, not a human view. People on the Internet do not "resist" censorship in the same way. Whether or not the Internet routes around censorship is a technical issue; that people are affected is certain. Even the censors themselves are not immune from being affected.

Ethics is Based on Physical Reality

People have been thinking about ethical issues for thousands of years. Most of our Western views come from consideration of physical objects, including people's bodies as physical objects of special significance. We have well-established ethical frameworks about property and possession of objects, and indeed about the possession of people.

We have a huge tradition of what might be called "pure" ethics: of philosophers like Plato, Hume, Kant. These have created a huge industry of disputing and classifying ethical issues, giving us the impression we know more about natural ethics than is the case. We also have a bias to Western ethics, and it is very easy to lose sight of conflicting cross-cultural ethical issues: we don't encounter them very often, we don't understand them, and we don't have the conceptual tools - the cultural heritage - to think about them clearly. Furthermore, most people operate - and most people on the Internet operate - without reference to any standard texts and their distinctions! So our ethical ideals, even if we know what they are, and unless very carefully circumscribed, have rather limited relevance in the face of the varieties of views on the Internet.

In particular, we have very little ethics about information as such. There are two main exceptions: the area of contracts, which are promises between people, and the more recent area of intellectual property rights. Intellectual property rights, IPR, arise because technology makes it trivial copying information yet manufacturers think they need prices comparable to the costs of creating information.

It is more complex than this, of course, for protecting intellectual property rights requires lawyers, and their costs have to be factored into the royalties. If I copy some company's program, they [insert name of a software company here] lose nothing - they still have their program - other than the opportunity to charge me. Nobody, least of all the company, will pass up an opportunity to make easy money. Conversely, because technology has made it trivial to copy, consumers, too, have less on their conscience when they copy information. People who might wonder at the legitimacy of copying a paper book may not feel restrained at copying an electronic book, because it is so easy to do so. One wonders whether worries about IPR are more to do with physical effort and tradition, rather than something intrinsic in information itself.

Whatever, we now have a society that just hums along on IPR (if all fees and royalties are paid for the tunes).

Except it doesn't. I have a considerable amount of intellectual property, over which I have no rights whatsoever. My financial habits, my spending habits are intellectual property that are owned by corporations who watched what I did, and then copied it into data banks. I have no effective rights over this information that I clearly created.

One interesting aspect of information that makes it ethically very different from physical objects is cryptography. It is now possible to make information more secure than is possible by physical means. Banks don't use strong cryptography because if they did, and lost the key, they would be in trouble that no box of dynamite would solve. Similarly, a criminal who wants to store information on a computer can do so in a way that no law enforcement officer can decode, ever. Using these methods, it is possible to sign information (rather like you sign a cheque) to guarantee it is your information. The signature can be done in such a way that it is unforgeable. You can also send secret information to people, and only they can read it. The technology to do this is freely available (though it is banned in some countries).

It is claimed that these methods put power back in individual's hands. Indeed they do, and there are many cases where people have been able to speak out confidently about human rights abuses, knowing that the authorities in their countries cannot read what they are saying.

Unfortunately there are snags.

Governments everywhere are a bit edgy about their subjects keeping secrets that the state has no access to. Various schemes have been proposed that would give authorities a "backdoor" into private information. Obviously there would be safeguards, like requiring warrants. So far nobody has come up with a satisfactory scheme, other than the libertarian view that the state has no rights over individuals: but that works both ways! The state can be as secret as it likes.

Another snag is that if there are facilities like digital signatures, then you need a system to recognise whose signatures are whose. I might sign my will, but if nobody can interpret the squiggle, or know whose it is, then the will is useless. Apparently, there need to be authoritative repositories of signatures (although humans have managed to cooperate for a long time without such authorities!).

It is then but a short step to national - if not global - databases and national IDs. Actually, we already have the same thing. There is nothing to stop the state correlating the information on driving licences and benefits for the registered blind, to clamp down on people who are fraudulent claimants - or very dangerous drivers. And it is a short step from where we are now to requiring an identity to be carried by the person, perhaps permanently. There will be so many advantages to having such an ID that we may willingly have smartchips embedded in us to ease our daily lives.

Recently, a baby was stolen from a UK hospital. Given the immediate publicity and depth of feeling aroused by the media, it was easy to make a persuasive argument for all babies in hospitals to be tagged so that they would not be stolen [6]. Next year we may have arguments that if the babies keep their tags on, then health checks will be easier. A few years later, the children's education may be improved with tags - who knows? If adults had such tags, then the first priority would be to implement participatory democracy. Some hope! We could have regular referenda, and governments would know what people thought. The tags would put an end to double voting and all the usual problems.

We may feel queasy about tags, but there are other cultures which would have no reservations. We have to think through the possibilities and the principles. Is Europe's Nazi memory relevant in a world of Asian tiger economies?

Homogeneity - "Tribalism"

The scale of the Internet makes a difference to personal relationships. There are enough people on the Internet for a large number to be interested in whatever you are interested in. Suppose you are interested in guinea pigs, then there are lots of other people out there to share your thoughts and photographs with. More seriously, if your concern is hereditary fructose intolerance - or any other important issue - finding a support group can be a life-saver. This appealing side of self-selected groups has led to the metaphor of the "global village," a social structure that is world-wide with the conviviality of an agreeable neighbourhood. A more objective term is "glocal" - the local is global, and the global is local.

Now suppose you are interested in torture. There are a lot of people out there who would share your interest and would, just as with guinea pigs, share techniques and provide advice. Of course, there are always going to be sad humans and the Internet won't change that. What it has changed is that it can provide more than enough contact for it to seem like that is all there is. A sadist can now live a life segregated from people who might want to reason with them. This is the problem of homogenisation.

If you can select the community you live in, you tend to select one you get on with. That is useful for progressing science or business [7], but if that community becomes socially all there is to you, your view of reality is distorted.

Jewish groups say that they prefer to see where revisionists and hate groups are, and what they are thinking. Unfortunately, this is a temporary benefit. As encryption gets more widely used, we will no longer have any chance of knowing who is in what group or what they think - if they want to keep their thoughts to themselves. The benefit of that is one could no longer come across their ideas by accident.

If this line of thought is pursued, we find we are able to give people free speech in every area of their lives exactly to the extent that it has no impact on anyone who disagrees with them. This is not free speech as we know it. Since it is easy to see that hate groups wouldn't have effective free speech in our society, it is easy to see that human rights activists somewhere else in the world would be equally powerless.

The Ku Klux Klan (which was originally founded around the same time as the Red Flag Act) has Web sites. America Online has come under pressure to remove them. I found that the aggression on the Internet expressed against the Klan is itself pugnacious. The U.S. has an Act to suspend habeus corpus to help fight the Klan. Clearly there is a very fine line between opposing and oppressing. Should you suspend justice to act justly?

These are now global questions.

Should the U.S. publish sub judice details of Canadian trials? Should Salman Rushdie have published his book Satanic Verses if it caused a fatwa against him?

There are many serious issues. Moreover, they are bound up with national security, cryptography, and with mechanisms for processing money.

Thus it is hardly likely that free speech will triumph - not that it should. The conflicting set of values we too-conveniently call "free speech" require universal literacy, confidence, the ability to give and take constructive criticism.

Fortunately these more fundamental things can be developed in Internet sub-communities. Interestingly, many Internet communities have explicit conventions, having worked out for themselves a balance between privacy and free speech.

If Rushdie moved to Northern Ireland, he could use his literary talents exploring the tensions there. Northern Ireland could do with a high-profile novel exploring the issues. Moreover, Muslims might see Rushdie in another light [8]. The point is: Rushdie can already go to Ireland and join its communities on the Internet.


There is no Star Trek transporter. Everything on the Internet is information, and any physical appearance it has is imaginary. Our brains were designed a long time ago, and we are simply not very good at distinguishing reality from virtuality. The flickering light coming from a television stimulates us as much as real events do.

The media industry benefits enormously from this primitive response.

How we understand media has, in a sense, not changed: for our limbic systems, it is reality. But how we create media has changed dramatically. Technology has made it unbelievably cheap to create images, sounds, and texts. The media is then available on a global scale with no extra effort. The leverage between my tiny efforts as an author and their potential impact on the world scale is phenomenal.

The well-known problems with "flames" are a small example of this: it is all too easy to write email that the recipient finds extraordinarily offensive. Flames, fortunately, are mostly one-to-one. But if someone sent a flame to everyone in another country, well, wars have been started by less.

The Ring of Gyges lets a person make themselves invisible. Plato, in The Republic, argued that nobody wearing the ring could be so strong willed they'd be able to resist and not behave like a God. They would not be able to stand fast and behave justly. It's an interesting thought because the Internet is a Gygean Ring.

Being present yet invisible does have advantages. For people who have impediments to social interaction in the physical world, the Internet gives a tremendous boost. They are no longer disabled.

Consider two interesting consequences of this.

First, as the saying goes, on the Internet, nobody knows you are a dog. It turns out that pretending to be other than you are is very attractive, and it is routine for netizens to present themselves as other than they are. A man might pretend to be a woman, because it gives them a sense of freedom, or whatever. They might do it to be deceptive. Anyone could pretend to be a child and befriend, and hence mislead, a real child. A lot of this is going on, and people don't just explore different human traits, but also try being animals and avatars, pixies and spirits [9].

Secondly, our society economically marginalises older people. When people can pretend to be other than they are, and since the Internet is making money, then it follows some old people will make money on the Internet. Which is fine, you say. But demographic trends - lots of old people - plus Great Britain's (and every other countrys') hope of being a knowledge economy suggests the impact will be a lot more significant than Granny topping up her state pension. We already have enough disenfranchised youth without them having to compete for work with 70-year olds.

Actually it will soon be us, not our grannies. In that day, when tags are needed to jack in and join up, will we be too old to register to be allowed to get them? If we are optimistic about the potential for social change, if older people contribute economically, then rather than old people being retired we might have young people - when they are having children, buying first-time houses, getting established - being freed of having to be economically productive and paying the pensions of the older generation.

Scale has Consequences

The markets are global and we disdain local produce in favour of what is apparently much better. We use supermarkets because we have more choice than in corner shops. Result: town centres decay, and we have inner city deprivation. The same pattern is being repeated on the Internet. I can buy books from the largest book shop in the world, and more cheaply than I can from my local store. Who can compete when only the best can survive? Competition is too fast, and the benefits of scale too slanted in favour of the leaders. Microsoft can get bigger because nobody can compete in any area it chooses to operate in.

A company like, probably the world's largest online bookseller with sales of $38 million in the most recent quarter, would be silly if they did not locate themselves in several places. If one of their computers crash, it'll be mirrored elsewhere so they can carry on. Their mirrors might as well be in, say, North Korea as in different locations across the U.S.. Some places will be cheaper to locate in or to do business in. There is already huge mobility in the financial markets, but this is mobility in conventional markets. The consequence is that no nation can charge high taxes on corporations: they would simply move elsewhere - not that it's obvious where they are to start with.

Information can move around, but people can't. That is one reason why governments tax people rather than information. A special case of information is money. My building society is raising its rates again, because it could lend the money it lent me elsewhere at a higher rate. If moving money around was costed, as my physical movement is, the building society would be more reluctant to hike its prices.

There is an important difference between the mode of operations of and Microsoft - to take them as typical of large corporations. I can buy books from anyone else if I want to. The Internet relies on computers running systems to compatible standards, and just any old programs will not do. Hence "lock in." Lock in is best understood by the contrast with the Oxford English Dictionary, which is the best dictionary of English, and five times larger than its nearest rival (note that Microsoft isn't the largest software company in the world). Nevertheless, using the Oxford dictionary does not lock its users in to Oxford University Press or to any of its products. In contrast, Microsoft's software does, as do Internet standards like PICS. (It is too early to say whether English will become locked in.) Curiously, one Ku Klux Klan Web site apologises for the inconvenience of lock in (not that it used the term) to proprietary software [10].

One consequence of lock in is that customers are effectively pressurised to stay locked in. If I wish to communicate with most people, then I had better invest in Microsoft products, simply because most people use these products, and they have unique quirks that I can only use effectively if I use the same products. Microsoft, in its turn, wants to make a living by not selling products that do a job so much as selling a subscription service, so I have to pay them regularly. The same problem occurs with Internet service providers - America Online is the recipient of Internet antagonism - except that it is perhaps more obvious to its users, as AOL is by no means a majority service provider.

The result is that computers have become a fashion industry. As consumers become fashion victims, everybody feels that they cannot "keep up" - the fashion markets are always providing enough temptations to keep most people ahead of us, which in turn pressurises us to catch up to remain in touch.

In areas of life that are free from lock in, the cost of fashion is easy to resist. When I am cycling along a road and overtaken by a sports car, however much I might covet the car, I don't bother pedalling faster. My bicycle does cheaply what I need it to do. Even though I have a good bike, sports cars cost thousands of times more, whereas powerful computers seem to cost very little more than older versions. Thus even the hardware runs on a subscription basis.

Most companies that rush to upgrade - in order to stay in business - ignore the human costs of retraining, and the costs of wasting high technology. Discarded computers are of very little use anywhere in the world. Few companies treat their software investment as an asset, so upgrading is under-costed. Moreover, software suppliers play to this by giving customers no choice: to fix bugs they have to buy upgrades. In turn, by paying to get bugs fixed, people buy themselves into the next level of fashion, and therefore encourage more people to catch up to remain compatible. It is a cycle that stimulates short-term economic activity, but which is non-productive.

To make the Internet succeed we have to break out of this trap. There are millions of people on the Internet, and while there is a perceived lock in, there will always be a lag between leaders and followers - so the race will persist. To break out, advanced technology will have to remain compatible with earlier versions, and early versions will not have to accrete features. The old version of my word processor cannot read the files a new version of the same word processor produces. This is completely unnecessary, and can only have happened because it suits the manufacturer to lock me in, to encourage me to pay to upgrade my software.

It will not be easy to do so, because humans like each part of the process (see also the discussion on homeostasis below). It is only when we can stand back a bit that we see that the positive feedback lacks any overall productivity.

There are markets in information itself. The news is bought and sold. Sifted by multinational media companies, items from around the globe are brought into our personal awareness. This is both good and bad. It's good because we learn about Bosnia. It's bad because of the scale: we lose sight of the significance of our local geography. If there is one murder somewhere in Britain, it is made to seem as significant as one next door. We are frightened to go outside - which no doubt reinforces the consumption of the media because we have nothing else to do inside!

Has anyone heard about the 'nation' that unilaterally expelled 40,000 of its citizens? It was a multinational company, and I expect their share value went up as a result. This has been called "virtual feudalism." Some virtual fiefdoms are already larger than many nation states. It is becoming clear that whereas nation states are obsessed with borders, corporations have embraced the virtual mobility the Internet gives them. Local unemployment will be one consequence.

Restoring a sense of community may help the unemployed, and this is something the Internet can provide. Except that computer networking is only available to people who can afford it - which immediately eliminates most of the unemployed from trying it. Serfs don't surf!

The cost of information is an increasing social issue; we have free public libraries, but they charge for access to the Internet. This opposes whatever principles we might have had underlying free access to information and culture. (Perhaps if we can retain traditional democracies, there will soon be enough unemployed to elect a sensible system!)

Most people who have a house have to have building insurance. Last summer, a house fell into the ground and the insurance company said it was an act of God and therefore not covered. It makes you wonder; insurance is there to provide protection, so when your house is swallowed up in a hole, isn't that just the sort of thing any house insurance should cover you for? Apparently not.

The reason for this is that insurance companies protect us against certain sorts of risk, namely ones whose risks are quantifiable, and therefore necessarily happen more often. If you want protection against nuclear war, earthquakes, and a wide variety of other disasters, you had better start paying your premiums to the Red Cross.

The Red Cross operates on a different scale and for a different reason. It is not likely that any corporation would want to, or even exist long enough to, work on that time-scale.

The Year 2000 problem is said to be the most costly single problem ever faced. In Britain, it is estimated to cost billions. There isn't that much money around, and there aren't that many programmers. Not surprisingly, insurance companies no longer cover the millennium bug, because the risks are unquantifiable.

In December 1997 it was reported that the Prudential Corporation had not yet sorted out its part in the pensions mis-selling scandal [11]. I think the Prudential used computers to support a customer base that was too large to handle when things went wrong, as they did. They still have a backlog of 71,358 cases. They claim that victims are not returning important information to be processed. Nearly half of their victims fall into a "high risk category" which includes customers who are dead! When things work well, a corporation can lay off workers, because few are needed. When things go wrong, they can go wrong on a scale that is beyond the few that are still employed are able to handle.

Computers allow a "free" scale up in operations. Provided nothing goes wrong, a company can operate with a huge customer base. If something goes wrong, the company does not have the human resources available to sort problems out. When I talk to my building society, I get the distinct impression that you don't need to wait for things to go wrong for the company to be out of control of what it is doing. Another example: I recently applied for a Visa to visit Australia. In the "old days" I would have been able to talk to somebody; but the High Commission has now automated its handling of enquiries. I had to ring several telephone numbers, and the automated service that provided the answers was a very slow menu-driven scheme that cost me £10 to get just one answer. It has become easy for them, anonymous - and the technology bills everyone who uses it. They benefit in every sense from the scale, at the literal expense of individualism.

There is another problem of scale, complexity. When a corporation scales up its operations, each customer is treated much the same, so everything is ideal for computerisation. However, when information is different, the number of ways in which things can combine and interact quickly gets overwhelming. Putting information on the Internet is not just a matter of publishing a web page, but has an organisational overhead, as anyone who has tried to maintain several pages will know. The complexity of saying something significant rapidly becomes prohibitive to all but the most well-resourced publishers.

Those are a few simple examples of the problem of scale. The Internet is the biggest human collaboration ever. So far, the most obvious (or the most complained about) problem of scale has been "spamming" - sending trivial email in huge quantities (for marketing, or for overloading people to put them out of action). One effect of spamming is to make some people fear to send email: if you send a message to a public place (say, a user group), then your email may get noticed by someone collecting emails for other purposes, such as spamming. The other problems are more subtle, and we have survival mechanisms so we don't notice them.


Humans are homeostatic. We have a standard operating temperature, and moderate changes in our environment don't affect our core temperatures. We sweat or shiver to compensate. Homeostasis applies in many other areas too. Cars made it much easier to travel, reducing the time it takes to undertake journies. The result is that we now live further apart than we used to - we have maintained the mental distance between ourselves.

When things go wrong, we have a fever if we are too hot, or exposure if we are too cold. Stretching the metaphor: we are over-exposed with the new challenges of the Internet and we are suffering feverish response mechanisms.

An optimistic reading is that the tensions we are under (which result in red flag waiving) are symptoms of homeostasis trying to revert us to a stable condition. Red flags can keep us in one stable configuration, progress without red flags is perhaps safer than we envisage, because we will end up in some other stable condition that is equally acceptable.

On the other hand, if Britain had retained red flags and slow traffic, perhaps we would be living closer together. Journey times would, on average, be much the same as they are today. But we would have been much less likely to have put up with inner city decay and many of the other problems that are blamed on personal transport. In any case, Ivan Illich argues we still move at walking speed - when we more honestly measure the miles per hour to include the hours we spent working to pay for the car, depreciation and for the journey.

Quite possibly, then, symptoms such as "information overload" and the censorship issues are all temporary while society re-establishes its equilibrium. Quite possibly, developments in micro-charging for services will make the market stabilise supply and demand. On the other hand, the size and speed of the world is much bigger than Adam Smith had in mind, and the result may be chaotic. If so, regulation could dampen the system and make it more stable - but might reduce some of the opportunities.

Homeostasis sounds like a reassuring analogy, but it fails to relate to the complexity of the possibilities opened by the Internet, and in particular, the global scale and speed of the Internet. Any homeostasis of the Internet will be an emergent feature of it, and will not relate (not in any planned way) to the preferences of individuals. Technology allows operations to be scaled-up and cheaper; corporations may respond with compensatory actions to restore their power or profits. For example, British music companies make a profit selling CDs at a street price of about £15 when they cost about £0.05 to make. When consumers can buy their own CD writing equipment, the industry's recent profits are threatened (CDs haven't been around very long). The industry could either move with the technology and promote some other form of profiteering, or - as they have chosen to do - agitate for a state-imposed tax on blank CDs to restore homeostasis.

What homeostasis does imply is that people involved with the Internet will under-estimate its scale. The global stage is too vast to comprehend. We will keep buying upgrades because it suits the small-scale relations we do comprehend - almost as if we have homogenised our software versions to live in self-sufficient isolation. Surely there is some global responsibility here?

The Tragedy of the Commons

The tragedy of the commons is a problem about shared resources, like over-grazed commons. Everybody can see that putting fewer cows on the common would halt over-grazing. Yet it is not to my advantage to starve my own cows. When each person thinks like this, the commons are over-grazed, and then everybody loses out.

While we stand to one side and watch successful trawlers coming back from fishing we can think about market forces - until there are no fish. It may be that the Internet is similar. We are currently seeing a few big trawlers doing extremely well, with their big nets, and they are clearly doing so well because everybody else pays them a little bit, and governments subsidise the infrastructure, and so on. These big, flashy, trawlers catch our attention.

In the same way, each individual improvement to the technology appears to be progress for all. A year ago it would have been easy to argue that the Internet provided a platform for everyone. Better than printing, if anyone had a message for the world, here was the way to get it noticed. Things are changing. Have you Java, cascading style sheets - or the new extensible markup language, XML, that obsoletes style sheets and all your HTML software?

The technology is now far more complex, and consumers' expectations have gone up. A site with three dimensional graphics on it has 1000 times as many hits as a site with two dimensional graphics. Web directories (like Alta Vista) now only handle a small fraction of the web. It would be easy for your message to get lost in the surf.

There is a view that the Internet automatically leads to democracy. This view arose from a mix of technological determinism, plus early experiences with on-line communities, mostly ones steeped in North American culture and so-called First Amendment absolutism. (Some Americans think Europeans have a strong personal socialisation, so they imagine that we don't see as much value in computerised networking [12].)

But it is not realistic. Somehow, virtual communities have to get real - perhaps like co-operative societies. Unfortunately the web is not run like this. The World Wide Web Consortium says of its own members that they are "not altruistic." It says its members "drive the future of the web," and the benefit of membership is "a marketing advantage lever to promote your company." Perhaps if it didn't say this it would have no influence.

The Internet once had a dominant culture like the Red Cross (as a metaphor of coping with scale): full of highly motivated, committed and co-operative people. But it is now rapidly moving into a culture of corporatism, where if anyone can make money, then everyone should. This transition, which is everywhere evident in Britain (most visible in privatisation), is a tension.

Democracies of influence have to be in the physical world. Votes have to be transformed into physical consequences. And democracy requires equal access; in the case of the Internet, the increasing standards are lifting the entry threshold beyond most individuals. Already, few people can contribute information properly classified by PICS - so, ironically, whatever they want to say will not be heard by people who care about what they read.

Raising the standard of entry may be a minor hindrance to people in London, but if we recall the Internet is world wide, then just where its advantages could be of greatest value - in the third world - the thresholds may be insurmountable. Why use a modem when you have no clean water?

Henry George and "Common Good"

Thousands of years ago when our physical existence relied on manual agriculture, nations had to be very careful how they shared land. In Sumerian and Biblical times, there had to be a regular equalising of ownership because if too few people owned the land, there would not be enough food to go round. As technology advanced, the egalitarian views withered against the concentration of ownership and physical power it permitted. Most of us are beholden to great landowners, because technology has given us enough freedom to survive without being tied to the land.

We are also in an age when information can be handled so efficiently that we are becoming beholden to great information owners. The corporation that mortgages my house knows a lot about me. The Australian High Commission knows information that restricts my physical movements, and it apparently has no intention of telling me for my convenience - and it has a financial incentive to behave like this (because computers give it the leverage to be impersonal).

Physical technology briefly promised freedom from oppression: everybody would be able to work less, and so on. It was a utopia never-to-be. What happened was a greater concentration of wealth, and a greater proportion of poverty and ill-health. Rather than people working less, fewer people worked. These issues were explored by the nineteenth-century social reformer Henry George, and I have become persuaded by his arguments about land, subsidies, and taxation on people. There are profound parallels with the virtual land, its virtual population, and the consequences - if not temptations - of scale for those in a position to benefit most from it.

We currently have no notion of "common good" that applies to information. We do not require information-based corporations to act in everyone's best interest. Britain might soon have a freedom of information act, and it will be interesting to see what "common good" it actually promotes.

If the true costs of operations, such as including the costs of cleaning up waste, were charged to corporations, market forces would reduce pollution and unemployment. However, corporations do not bear such burdens. Nor do they straight-forwardly bear the complex costs of information use, concealment, or collection. Instead, subsidies encourage corporations to exploit national intellectual capacity, which they can then export anywhere in the world. What we now see as physical deprivation in inner-cities will become intellectual poverty and intellectual ill-health in nation states, for cities have the relation to nations that nations have to multinationals.

Already corporations are trading in cultural values. Microsoft's Office'97 suite defers to Western ideas of sexual equality. But the version sold in Asian countries has an "office lady" who has a variety of submissive behaviours [13]. What does this say about Microsoft's view of Asia, and what does it say about its priorities? Presumably if negative sexual stereotyping is a bad idea, it's a bad idea anywhere, and perhaps especially so in countries that haven't noticed - except that you can profit by exploiting the cultural differences.

Reversibility is one of the standard ethical priorities. Intellectual poverty is irreversible for those trapped in it. If nations do not unite, the rest of the world will always be bigger; markets will encourage an outward flow of resources. The standard complaint is that everything we invent is made abroad: this is no longer a peculiarly British ailment, but a statistical inevitability for every nation.

I don't want to be (just) a doom-monger. I want to make the point that there are serious, really important issues that need thinking through very, very carefully.

Plato's Republic gives us the famous analogy of escaping the bondage of the cave for the "reality of forms" - his phrase for cyberspace. But his republic was non-democratic, authoritarian, and relied on strict censorship. If we do descend into such intellectual ill-health, then William Gibson's Neuromancer (the dystopian science fiction that coined cyberspace) will look decidedly utopian.

Omissions and Simplifications

In any discussion about the Internet, there are bound to be omissions. It is important to briefly summarise some things that have not been mentioned here, and to point out some simplifications in the things that have been said.

To make things briefer and easier to follow, I assumed an anglophile point of view. In fact, anglophiles are a minority in the world. The current English and American bias of the Internet is an issue we in Britain find difficult to appreciate. (Our popular culture is already so dominated by the U.S. that there is nothing new to worry about.) To us, the French, where the World Wide Web was invented, seem isolationist. But China, which out-numbers all English-speaking nations by 3 to 1, will find our point of view as inconsequential as the French view might seem to us! This is a serious issue, of the relevance of cultural values, of language, and of ethical frameworks. Our - if I can call it "our" - core notions of individuality, privacy, sexuality, and so on - make no sense in many other countries.

If I had mentioned human factors, I would have pointed out the Internet has provided many examples of companies imposing essentially untested products on millions of users. Things become "standardised" before they have been thought through - this is another problem of scale: it suits manufacturers to release dodgy software, and waste consumers' time, provided the software is just good-enough to ship. It is not only manufacturers who are to blame; we live in a culture of fashion, and consumers are willing fashion victims, locked in, and eager to upgrade and buy fixes to poor design. (I hope that the millennium bug, which will affect the Internet just like all other computer-based systems, will make everyone realise that computer problems are not always the users' fault.)

Communications are not the only area that is advancing. What use is an electronic self-help community for arthritic people if they still can't get jam jar lids off? Soon developments in robotics will probably make many physical chores easier (for those who can afford the gadgets). Telerobotics, telemedicine, and so on are already here in specialised areas. As my arguments about IPR make clear, cheap and ubiquitous robots will create tensions in even more areas because humans will be unaccustomed to working with the physical - even ethical - leverage.

A point (I made back in 1976) is that computers eliminate intermediary crafts. This is why computers are difficult to use - because the engineers who design them build systems that are not interfaced to humans, because the craft knowledge to do so seems superfluous. The same problem is now happening in corporations: they undertake business, removing intermediaries who would have understood personal values.

Limited rationality is the view that there are some problems that in principle we cannot solve head-on. Like Odysseus, who wanted to hear the Sirens, yet knew he did not have the will power to resist their call, we may need to limit our options so that we can solve the problems. (Odysseus had himself tied to his ship's mast.) Restricting free speech and other rights may be necessary in order to have a simple enough world in which to make sensible decisions. Later, when we have sailed pased the Sirens, we may remove the restrictions - though, if you read the story, you'll find that Odysseus had to have a plan to get himself untied!

If information can be sold, then there is a tendency for any information to made profitable. This raises the issue of commodification. This is a worry because many forms of information that democracies rely on being free, such as education, museums and libraries, are already being costed.

Underlying my discussion have been various ethical principles, such as reciprocity, reversibility, and others. I did not make these principles explicit.

For simplicity I assumed a rather conventional view of democracy (though I did mention regular referenda). A more creative approach would be to argue that the Internet permits homogeneous groups: why not find ways to make those groups effective democracies - "micro-democracies." Such micro-democracies would be lateral, across society, and glocal: they could be concerned with local issues (such as local transport) or with global issues (such as pollution). They could have a role in disciplines, such as medicine, helping to find consensus in many areas.

Finally, if we agree that liberal democracy is a good thing, what do we do when most of the world disagrees with us? I hope that, first, we will get our own house in order: we don't have a democracy so much as professional lobbying punctuated by ineffectual and infrequent elections, and decorated by media fashions. Like many other issues, the pressure of the Internet, and other developments in computers and communications, are going to force us to think more clearly about what we really want to be as humans, and as human communities that can handle human diversity with all its illogical opinions.


I have four conclusions, and some recommendations for action.

First, there are new ethical issues to be confronted. But these issues are not so new we should lose our heads. Recognise red flags! Every action should have a net benefit - pun intended.

Secondly, if we do nothing, then the default future will be corporatism and will reduce democracy as we know it. Despite all the trends to the contrary, democracy is not a marketplace.

Thirdly, we have been surfing a wave of excitement. The promises and problems of the Internet so far, which are obviously the ones we know most about, are transient. We've had email for just 25 years. Compare that rate of progress with cars: it took 31 years from the first car to the first regular bus service (in Paris), and then another 165 years before Ralph Nader raised awareness of the consumer issues.

Fourthly, and most important, we need to find a glocal notion of "common good" that can be applied to the Internet, which makes sense both within personal boundaries and on the global stage. We currently pay taxes to benefit everybody in the country, and I think we are the better for that. A purely unregulated free-for-all Internet would not be free for all. The Internet is currently very heavily state subsidised, and this distorts its value. A "byte tax" is a possibility.

What can be done?

When the telegraph was new in the last century, someone asked, "what would Maine want to say to Texas?" Today's questions are not tomorrow's.

It's not obvious what we can do to have a better future. I suggest some points that may help:

We should try doing some good things deliberately.

About the Author

Harold Thimbleby is Professor of Computing Research at Middlesex University, London. He gained his PhD in 1981 in user interface design, and much of his research since then has been concerned with making complex devices easier to use. He was awarded the British Computer Society's Wilkes Medal, and is a member of the UK Engineering and Physical Sciences Research Council's Computing College. He has over 200 refereed publications. He wrote User Interface Design, published by Addison-Wesley in 1990. He is currently a member of IFIP 9.2.2 (Framework on Ethics) and on the Church of England's Working Party in IT. He has been widely reported on TV and radio. His email is


  1. This paper is based on a talk originally presented at the Institution of Electrical Engineers 2020 Vision Meeting on Morality in the Information Society, held on 11 December 1997, in London.
  2. This article is not long enough to make distinctions between the different sorts of net. The Internet is the "net of nets."
  3. Alta Vista, which covers about 60% of the Internet , returns 17554 sites with the words "guinea pig" adjacent.
  4. See Jim Miller, "Promoting the PICS Fix," World Wide Web Journal, 1(3), p11, 1996.
  5. Democracy must assume everyone is not a potential crook.
  6. The Times, 8 December 1997.
  7. One of the main reasons that Internet technology develops so fast is because there are homogeneous groups of technicians working on it. The conventions of requests for comments, RFCs, exploits this.
  8. See Simon Lee, in further reading.
  9. Being anonymous is a special case, where there is no pretence to have a particular identity. Anonymous remailers are systems to ensure anonymous communication.
  10. One example: on January 30, 1998, sa id "This is a Ku Klux Klan Website! If you are offended by our views then GO AWAY. [...] This site is optimized for Internet Explorer 3.0+ and Netscape Navigator 3.0+. HOWEVER..... I have been having problems with Netscape 4.03 hanging badly on some of the pages that use Java."
  11. The Times, 5 December 1997.
  12. See A. M. Cohill & A. L. Kavanaugh, eds., Community Networks: Lessons from Blacksburg, Virginia, Norwood, MA: Artech House, 1997.
  13. New Scientist, 156(2111), p92, 1997.
  14. A not-for-profit organisation has an annual fee of 5000$U.S..

Background Reading

50th Anniversary issue, Communications of the ACM, February 1997.

W. Gibson, Neuromancer, Glasgow: HarperCollins, 1993.

W. M. Grossman, Net.Wars, London: New York University Press, 1998.

M. Hudson, G. J. Miller & K. Feder, A Philosophy for a Fair Society, London: Shepheard-Walwyn, 1994.

H. Küng, Global Responsibility: In Search of a New World Ethic, London: SCM Press, 1996.

S. Lee, The Cost of Free Speech, London: Faber and Faber, 1990.

D. C. Lynch & L. Lundquist, Digital Money, New York: John Wiley & Sons, 1996.

Plato, The Republic (Gyges's Ring is in Book II).

B. Reeves & C. Nass, The Media Equation, Cambridge: Cambridge University Press, 1996.

J. R. Saul, The Unconscious Civilization, Concord, Ontario: House of Anansi Press, 1995.

S. Turkle, Life on the Screen: Identity in the Age of the Internet, London: Weidenfeld & Nicolson, 1995.

N. Wiener, Invention: The Care and Feeding of Ideas, Cambridge, MA: MIT Press, 1993.

L. Winner, "Cyberlibertarian myths and the prospects for community," Computers & Society, pp.14-21, September 1997.

Contents Index

Copyright © 1998, ƒ ¡ ® s † - m ¤ ñ d @ ¥