Focus:This paper examines the nature of the problem caused by potentially offensive material on the
Internet and summarizes current efforts to regulate content, along with reactions to those efforts.
1. Introduction: what is the problem?
Earlier last year, anyone could have been forgiven for believing that the world had just discovered the
Internet, and that it had, in the process, concluded that the Internet was awash with pornographic images,
drugs information and general threats to the safety and good order of society. What was worse was that
nobody appeared to be in charge of this new phenomenon: indeed, it was proudly proclaimed that “no-one
owns the Internet”, except perhaps the millions of people throughout the world who contribute to it in
various ways. This is seen in many quarters as its major benefit – freely available information from
numerous sources – but it has also come rapidly to be regarded in some quarters as its most worrying
feature. Everyone with access to the required technology is free to make material available via the Internet,
and there appears to be no control over that material and so the ‘cyberporn debate’, amongst others, began
on the WWW, in the media, and in legislatures.
As a result, suggestions have appeared which link the use of the Internet with the Oklahoma bombing,
extremist political groups, the manufacture of ecstasy and other drugs, and with the ready availability of
pornographic images – and these are all problems which public libraries and schools are concerned about as
they move towards providing access to the Internet. There were also fears that it was too easy to find this
material inadvertently (as distinct from consciously searching it out): again it was felt that there was a
threat to the innocent and the unwary, although an investigation, reported in the Guardian newspaper,
estimated (Holderness, 1995) that “the odds against finding a random pornographic image thus seem to be
worse than 70,000:1″.
Given that a well-known British broadsheet newspaper recently listed the URL of a World Wide Web
(WWW) site which included links to eight so-called ‘top shelf’ magazines, inadvertent discovery can be
regarded as a problem, although some commentators have denied this, saying that it requires effort to find
these sites and to download images and so on. Inadvertent retrieval of offensive material is also less likely
due to the increasing use of warnings placed at the start of WWW pages: my recent research suggests that
the number of these warning signs appears to have grown considerably – of 81 sites I investigated, 45%
now have a warning notice which, amongst other things, requires users to be over the age of either 18 or 21.
In some cases, users are required first to register by quoting a credit card number: this is not used to charge
for access, simply to verify age, and a few sites have now begun to quote the Communications Decency
Act as a reason for requiring proof of age. !
Few if any of the other sites actively prevent under-age users from accessing the pages, but at least there is
no excuse for not knowing what the site contains. Of course, many would argue that such warnings will
only serve to encourage access, especially by children and young people.
It is not particularly difficult to find this material if one is consciously looking. The ‘adult’ bulletin board
systems advertise freely in many magazines, including some of the now well-established journals for the
Internet, while telephone numbers of BBS and network addresses circulate freely in the newsgroups. On the
WWW, the various search engines will retrieve Web pages with little difficulty, using keywords in
sophisticated search strategies. Professor Harold Thimbleby suggested (Thimbleby 1995) at last year’s
British Association meeting in Newcastle that “47% of the 11,000 most often repeated searches were
pornographic”, though it is very important to note that this does not indicate what proportion of the total
number of searches this represents: unfortunately, Lycos will not release this information, so it is
impossible to judge the relative extent of such searches (Whitney, 1996).
Thus, enter the word ‘sex’ on the Lycos search engine (which indexes over 130 million unique URLs, or
Internet addresses) and you will told that there are 30,976 documents containing that term. However, of the
first 50 URLs listed, 27 (54%) appeared not to contain pornographic images, but were instead either serious
discussions of sexual matters, duplicate entries of one site, lists of newsgroups on the Internet, etc.
Similarly, a search on ‘drugs’ found 92% of the first 50 sites (17,504 were found) could not be considered
as actively encouraging drug use, and of the first 50 documents listed under ‘explosives’, 60% stemmed
from organizations with a legitimate interest in explosives such as professional institutions, etc. , although a
number of the remaining sites were parts of The Terrorist’s Handbook, which does give recipes for bomb-
The problem of accurately calculating the relative volume of potentially offensive material has confused
the Internet since its growth first became noticeable. It is too easy to generate figures and then to use
generalizations and apply them to the whole of the Internet to create the wrong impression, but it is often
this false perception which generates so much concern in parents and legislators. This was a major criticism
of the study by Marty Rimm last year which featured in Time Magazine , and was also used by US senators
pushing through the Communications Decency Bill. Rimm’s study (Rimm, 1995) was interpreted by US
Senators as proving that 83.5% of the Internet consisted of pornographic images, when in fact he only
analyzed the adult sites and then extended his conclusion to the entire Internet. Examining only adult sites,
it was hardly surprising that he could claim to have found a large percentage of such images. A more
accurate figure has been calculated by Hoffman a!
nd Novak, who point out that “less than one-half of 1% (3% of 11%) of the messages on the Internet are
associated with newsgroups that contain pornographic imagery.” (Hoffman and Novak, 1995). The point
appears to be that, while this material is available over the Internet, it is not there in the quantities which
some people fear, though it is very difficult to change entrenched attitudes. A good friend who is a school
librarian, told me recently that she is being given every encouragement to provide Internet access. Her first
question was, “But how do I stop the pupils finding all the pornography?”.
Once one page with the sought after material has been found, it is usually possible to link from that page to
others on the same theme, just as one can follow citations in a journal article. Some WWW pages are
simply that: links to other related pages, like bibliographies of their subject.
The conclusion is that material which could give offense and which legislators in the UK, USA, Australia,
Canada, Germany and New Zealand (to name but some of the countries actively investigating Internet
content) are concerned about is there, but not in the quantities which some branches of the popular media
would have us think. Having said that, it does appear that a lot of use is made of these sites, especially
those providing the alleged pornographic images, though last year. The Guardian also found that the
pornographic newsgroups generated much less traffic than, say a group listing job opportunities
(Holderness, 1995). In a recent analysis, I found a high of 58,198 accesses per day and a low of 198 for 12
randomly chosen sites: the average daily figure for these sites was 14,578 accesses. Again, this figure must
be treated cautiously, as it may not, in every case, represent the number of individuals accessing the site,
but rather the number of times text or an image was!
downloaded: some accesses may result in multiple downloads. Some sites claim accesses far in excess of
this: one million plus in a few months has been seen.
This high level of demand is also a contributory factor in the rapid turnover of sites, which close down or
are closed down regularly. The most frequent reason for closure is sheer overload on the server involved,
which causes problems for other users: 41% of sites closed down in the course of one investigation and the
single most frequently cited reason (58% of cases) was system overload. Concern over legal action or
breaches of acceptable use policy tend to be less frequently cited as reasons for closure. The JANET
Acceptable use policy in the UK already prohibits the use of JANET for a range of materials, including the
obscene, libel and copyright infringement (UKERNA, 1995). We might also note that such sites have a
relatively short life-span: 87% had been closed down within 6 months of start-up and 43% lasted only 2
months or less.
So far, we have considered Internet access to the material which has caused most reported concern to
various groups, but if we are considering the question of free or fettered access to networked resources, we
might also look at a wide range of material which could be subject to criticism on completely different
grounds, if only because of the rapidly increasing demands being placed on Internet bandwidth. Given that
the original purpose of the Internet was academic research, why should such recreational material as home
pages for football clubs, popular music groups, television or a student’s interests and hobbies be allowed
space on servers? Somewhere on the Internet, I’m told, is a list of one young man’s CD collection: is this a
sensible use of the resources? In the United States at least, the continuation of such sites (as well those
containing the potentially more offensive material) are justified under the First Amendment, and they are
symptomatic of the change in the na!
ture of the Internet from academic research tool to multi-media information resource.
I want simply to raise the question here because it is part of the wider question of Internet regulation – some
universities do not permit students to maintain their own home pages because it is an ‘inappropriate use of
school resources’. A further step in this direction has been taken by one British university which logs the
thirty most popular URLs each week. The resulting list is examined and access to any sites which
contravene user regulations may be blocked. This includes those sites which it is felt constitute a waste of
university computing resources, and ‘ deliberate or reckless overloading of access links or of switching
equipment’ is prohibited under the UKERNA JANET Acceptable use policy.
So, to summarize this first part, we can say that there is material available over the Internet which causes
concern in some quarters, either because of its content or because it is seen as a less than appropriate use of
computing resources. We can also note that some degree of regulation is already in effect, especially in the
university sector, though increasingly service providers are taking note of the problem. CompuServe, a
major Internet provider, announced that it was blocking access to some 200 newsgroups, following
complaints from the German Government about obscene materials which were being accessed from
Germany. CompuServe say that there is no way in which they can selectively prevent access by users in
one country, so their action applies to all CompuServe users. In addition, the German Land of Baden-
Wrttemburg is investigating Deutsche Telekom to see if it can be considered as assisting in the
dissemination of the views of a US-based neo-Nazi (Schofield, 1996)!
In the USA, legislation was passed at the beginning of February 1996 which makes it illegal to distribute
certain types of material over computer networks, and the UK, Canada, Australia and New Zealand are all
actively examining the problem with a view to defining its size and the nature of their response. In the UK,
for example, the CCTA’s Ethics Open Collaborative Group is looking at ethical issues in the use of the
information superhighway, and its 1995 report contains statements on censorship (CCTA Ethics
Collaborative Open Group, 1995). It is also looking at technical options, such as limiting access to
approved sites, or prohibiting access to banned addresses. Following a meeting between the UK Home
Office and Internet service providers, the UK government is looking towards a code of conduct from
providers. In Australia, a Computer Bulletin Board Task Force is examining the regulatory options for
BBSs, in the light of community concern over offensive materials online (Fe!
deral Department of Communications and the Arts, 1995), and the Canadian Government has already
issued a background paper on the topic of illegal and offensive content on the information highway
(Sansom, 1995):the final report of the Canadian Information Highway Advisory Council makes a number
of recommendations, though none goes so far as to advocate new Canadian legislation (Information
Highway Advisory Council, 1995).
2. Is the Internet so different?
If offensive material is there on the Internet, are we in fact facing a radically new situation, or is it simply a
case of old wine in new bottles? Does the fact that a new medium is involved warrant new legislation of the
type currently before the US Senate, or can existing laws deal with this material? What, in a word, is
different about the Internet? For many observers, the answer is ‘nothing’. They would claim that current
laws are perfectly adequate to control material judged to be offensive by national legislation, whether this
relates to possession or distribution or both, not least because it would appear that much of the offensive
material is in fact digitized from original print versions. The argument is that the Internet is just another
medium, and that we must not allow ourselves to be dazzled by the technology into creating a whole new
set of laws, most of which, might not be enforceable. At most, all that is required is an amendment to
existing legislation which !
will cover digital media and computer networks.
Other commentators, however, claim that there are differences in the way in which this material is
distributed which are inherent in the Internet and that it does therefore create new problems. Firstly, there is
that fact that access to this material is easier: there is no longer a need to have contacts or to go looking for
this material, the search engines will find it for you (as we saw earlier). Thus, individuals who might
otherwise have been protected from offensive material by simply not knowing how to find it, can now
actively seek it out from a desktop terminal (admittedly, one does require access to the technology, but that
is becoming less and less of a problem).
Secondly, since this is access from a desktop terminal, access is less public. There is no need to risk the
embarrassment of buying a top-shelf magazine in the local news agent, or of trying to find the Terrorist’s
Handbook in a book shop, or to be seen with known members of neo-Nazi groups: all of this can be
downloaded in the (relative) privacy of one’s room.
However, the biggest difference which the nature of the Internet makes is in its international nature, and the
potential this creates for breaches of national legislation. We have already mentioned the demands of the
German Government to CompuServe and the problems these have created, and this is only the latest
instance of the problem to surface. It is now well-known that material prohibited in one country can be
stored on a server in another and then accessed from the first and that, at present, there is little that
governments can do about this: it cannot be stopped at the frontier, as print can. This may not always be an
overt attempt to evade one country’s laws, but there are instances where information is located on a server
in one country specifically to avoid legislation in another. It is, however, accessible to all who seek it out,
even if it is forbidden to some. Can country A put pressure on country B in some way, and how would
country B react? How, for example, wou!
ld the Netherlands, with its more relaxed attitude to soft drugs, feel if the UK asked it to close down WWW
sites containing information on marijuana, because British residents were accessing the information? If it is
not illegal in the Netherlands, would the Dutch Government have any right to accede to such a request? The
Chinese government is very concerned about this aspect of Internet provision, and is attempting to
centralize control of Internet usage by requiring all users to register with the police: its concern is over both
pornographic material and information which threatens the security of the Chinese government. Singapore
has recently announced that it too will seek to control access to politically sensitive and sexually oriented
material (The Guardian, 1996).
The answer, it may appear obvious, is to create new international legislation, but as we shall see, this too is
fraught with problems. In many cases, it appears that all that can be done is to request cooperation between
states to establish which nationals have accessed the prohibited material and then to take action internally,
but that too may involve breaches of constitutional rights of privacy and/or freedom of speech.
3. What should we control?
Can we then look at the problem of what should be controlled on an international basis, so that
some level of multilateral agreement is hammered out? What Internet content could the international
community agree to prohibit universally, so that there is no uncertainty?
I asked this question of almost 20 students during a seminar I attended just over a year ago. Many
suggestions were forthcoming from individuals, but we could not agree on one single topic which should be
universally prohibited. In many cases, the attitude was ‘I don’t want to look at this material, but I appreciate
that others will’. Others pointed out that no-one was forced to look at this material: as with television, one
could always use the equivalent of the ‘off’ switch. Now, while a straw poll like this cannot be said to be
representative, it does indicate the problem, perhaps best summarized as ‘one man’s terrorist is another’s
freedom fighter’ – what offends me may not offend you: have I the right to say you should not view it or
read it? Attitudes to material which could be deemed offensive are so often part of a nation’s culture, and so
will differ widely: consider the example of the Netherlands quoted earlier, or the difference between
Scandinavian and Middle Easte!
rn attitudes to nudity. Jack Schofield has recently raised similar questions: could fundamentalists ban
discussion on theories of evolution, or could the Vatican seek to ban sites on birth control? (Schofield,
1996). Can we hope to find common ground on which all could agree, and if we did, would the result be so
bland and anodyne that it would, in effect regulate nothing?
Somewhere in the vast amount of information currently circulating on the Internet I have seen it suggested
that daily traffic is around 30 terabits or 30 million bytes of information (Merel, 1995) – there will be
something to offend someone or some group: how can we hope to stop it all, any more than we have been
able to stop all of the same material circulating in print form? Let us not forget that, thanks to its Cold War
origins, the Internet is designed to overcome ‘blockages’: it interprets them as damage and finds another
way round (the newsgroups are a different case, and can be stopped if necessary). So would international
legislation be effective, even if there was agreement? Indeed, should we try to regulate all of it, or should
our main concern be the freedom of information and of ‘speech’, even in an electronic environment?
Societies do not generally try to control everything which is published just because it offends one group or
another: we are free to argue the !
case for and against abortion, the legalization of soft drugs, and so on, and the Internet provides us with
another medium in which to carry on that discussion.
Many commentators fear that legislation of any sort against offensive material like pornography or drugs
information will simply be the thin end of the wedge: it is, they feel, only a short step from there to
controlling or censoring other types of material which upset national governments (witness the efforts of
China and Singapore) or big business and commercial interests, a fear which will, I believe, only grow with
the progressive privatization of the network. Once commercial interests operate the Internet, will they use
their power to prohibit access to, for example, environmental groups critical of their operations?
There is a further danger that, in seeking to control the Internet, any national or international legislation
would be so blunt that it would prohibit access to genuine, serious material (on admittedly controversial
topics) along with the offensive. Let me give a personal example. I maintain a set of WWW pages on
Internet regulation, and, since pornography is the main topic of concern at present, the word ‘pornography’
occurs frequently in the annotations, etc. As a result, if you search for ‘pornography’ you will find my
pages, whether you are looking for serious discussions or pin-up pictures. The corollary of this is, of
course, that if you seek to ban sites containing the word pornography, my site would disappear as well!
This is one of the problems encountered by the various software programs (discussed below) which are
intended to control access to the Internet.
Of course, any such legislation would be doomed to failure from the start, because all that would be needed
is to remove the offending words, and legislators would be back to trying somehow to examine content, an
equally impossible task given the huge volume of material flowing daily over the network.
The Internet is clearly in a state of flux at present, and we will have to wait and see what develops, but for
many Internet users that too has is fraught with danger. They argue that we must take action now to prevent
governments legislating on Internet content, otherwise it will be too late.
4. How can we control the Internet?
With that in mind, let us turn to ways in which some form of control can be exercised over the
material which is currently found on the Internet.
The most time-honored method is that of self-regulation: individuals or systems managers and
administrators control what is made available from the systems under their control. It is they who have
been responsible for most of the closures of sites which I mentioned earlier, primarily because of the load
such sites place on the server, or because the material does not accord with an acceptable use policy. Many
system administrators admit, however, that it is technically very difficult to control which sites their users
access (as distinct from the material which users place online), though acceptable use policies are similarly
relevant and there is some evidence of various monitoring procedures (which are made known to users) to
ensure that use policies are adhered to. It would appear from a 1992 study, however, that obscene or
harassing material is not the primary concern of systems administrators: the most frequently occurring
incidents are virus infection and the printing o!
f non-academic materials on university-owned printers (Stager, 1992). Complaints that, for example, the
contents of BBS were obscene were mentioned by less than one third of the computer centre directors in
this survey. There is also, as Stager has pointed out, a potential source of confusion in the minds of
computer centre directors. More than two-thirds of those asked agreed that university-operated bulletin
boards are subject to restrictions if their content is clearly libelous or obscene, but less than half agreed that
they had a ‘general understanding of what my campus community considers obscene’. Again, we have the
problem of defining our terms.
I have already mentioned the growing use of warning signs as an example of self-regulation: they do not
prohibit access, but no-one can be under any illusion about the content of such sites. Acceptable use
policies on the part of organizations (such as universities and private sector companies) and of service
providers must also count as self-regulation, and there are instances where these have been used to prohibit
access to, and provision of, material judged to be offensive and/or illegal. The UKERNA guidelines are an
example, since it as a result of these that many UK universities block access to the more controversial
newsgroups, but many such institutions and providers have developed similar policies: many of these are
available on the WWW. The British Computer Society’s Schools Committee has produced guidelines to
prevent computer misuse which recognize the existence of potentially offensive material and give advice
on the development of a suitable policy (British Comput!
er Society Schools Committee).
For many Internet and WWW users, self-regulation is the only way in which to control content, anything
else being seen as an infringement of freedom of speech or civil liberty (especially in the United States). In
the United Kingdom, this seems, for the moment, to be the route which will be taken by Government,
which feels that a code of conduct is the best route forward, and presumably the Internet industry will now
sit down to establish that code, preventing the storage and distribution of offensive material.
4.2. Parental/Teacher control
As Internet access becomes a growing feature of schools and homes, parents worry about what
their children are seeing and reading in class or in the privacy of their rooms, and teachers are concerned
about the implications if, during one of their classes, a pupil stumbles across or goes looking for a
pornographic image, a discussion on abortion rights, or where to buy marijuana.
Many have argued that, rather than introduce sweeping legislation which will, it is claimed, be the end of
the Internet, parents and teachers should take responsibility for their children’s actions. The analogy is often
drawn with teaching our children to cross the road safely and to deal with traffic: we don’t let them wander
off on their own, but show them the dangers and how to avoid them. Gradually, as they learn and mature,
they are allowed to do things on their own. Similarly, it is argued, parents should accompany their children
on their first Internet surfing, pointing out the dangers in the process.
To help parents and teachers in this, a number of software packages have been developed to shut out access
to the more offensive material. Programs like NetNanny, CyberSitter, SurfWatch and Internet Filter block
access to sites containing certain listed words: this stop list comes ‘built-in’ to the software, but it can be
added to by parents to suit specific requirements. In some cases, attempts to access prohibited sites are
logged and can be inspected by parent or teacher, which could, if it was known, help to stop such attempts.
There is even a version for business use: presumably it could also be used to prevent staff from wasting
time surfing the Net!
4.3 Government legislation
However, anyone who has kept an eye on the scene over the last twelve months will be aware that the big
(and now rather complicated) story in controlling the Internet is the passage of the Communications
Decency Act (CDA) through the US Senate, initially accompanied by two other broadly similar bills, plus a
fourth which is believed to take a slightly different line and could be more acceptable to the majority of
Internet users. The amended CDA was passed by a single vote in a House/Senate conference committee
and became effectively an amalgam of two bills passed by the House and the Senate respectively. The three
most restrictive bills, which have come in for the most criticism, all seek to make it illegal to transmit
indecent material and would make both online service providers and content providers liable. The fourth
bill was thought to avoid such action and to empower parents and teachers to control Internet access.
The CDA was passed at the beginning of February 1996, and signed into law on 8 February. It was
immediately the subject of a law suit to the US Attorney General by the American Civil Liberties Union
and 19 other complainants, who have succeeded in having some parts of the act declared unconstitutional:
as a result, it is in abeyance at the time of writing. A similar suit has been filed by a coalition of industry
and online organizations, including the American Library Association.
The chief criticism directed at the act relates to its attempts to make service providers liable and to the fact
that, at the last minute, the word ‘indecent’ was substituted for the word ‘obscene’. The argument against this
development is that the concept of ‘indecent material’ is vague but more wide-ranging than ‘obscene’, and
that it would include not only sexually explicit material but texts which are now regarded as classics of
literature. The oft-quoted example is that this bill would prohibit the distribution of such classic works of
fiction as Catcher in the Rye or Lady Chatterley’s Lover , when most thought that this was a battle already
won. The bill would also give the US Federal Communication Commission some authority to regulate
It should be noted that opposition to the CDA is not limited to the online community and such
organizations as Computer Professionals for Social Responsibility, the American Civil Liberties Union, or
the Center for Democracy and Technology. Newt Gingrich has already suggested that such efforts are
unconstitutional and the US Justice Department has said that the legislation is unnecessary (it believes it
has all the legislation they need to prosecute where necessary). Other opponents have said that such
legislation will only result in the creation of an underground Internet, where the material objected to will
still circulate freely and could still be accessed by those in the know.
Not only would this legislation prohibit literary works, but also the serious discussions of many topics and
problems to do with sexual matters. There would be a threat to AIDS information, mutually supportive
discussions of abuse and many other online forums which, enthusiasts claim, can be very important to those
affected. Its effect on discussions of abortion rights have already been challenged as a breach of the US
This is currently the only impending legislation which seeks to control content on the Internet: many other
countries are investigating the question, and the UK government, for example, would prefer to establish a
code of conduct, as I indicated earlier. However, I think we can be sure that, if the USA passes this or
similar legislation, many other countries will face considerable pressure to do likewise: the pressure is
already there in the French efforts to introduce European legislation.
The question of a free or fettered Internet is not an easy one: so many variables come into play, and in
many instances we can only have a personal reaction to the problem of offensive material on computer
networks. The overwhelming majority of Internet users agree on the need to protect our young people from
this material, but the way to do so, they argue, is to involve parents and teachers and to educate our
children, just as we educate them about the dangers of road traffic. The problem, it is argued, is not as
serious as some would have us think, and legislation is too big a hammer for this particular nut: it would
destroy more than it would protect. The next few months are going to be both interesting and crucial to the
future shape of the Internet.
British Computer Society Schools Committee (1995) Preventing the misuse of computers in schools,
Swindon, British Computer Society.
CCTA Ethics Collaborative Open Group (1995) Superhighways Ethics Project: Report
Federal Department of Communications and the Arts, Australia (1995). Consultation paper on the
regulation of on-line information services, 7 July 1995.
The Guardian (1996). Singapor to censor Internet. 6 March, 12.
Hoffman, D. and Novak, T. (1995) A detailed analysis of the conceptual, logical, and methodological flaws
in the article: marketing pornography on the information superhighway
Holderness, M. (1995) In search of a sea of sex, The Guardian, 2 August,19. (Also available through the
Guardian Online Archives search facility)
Information Highway Advisory Council (1995) Connection, community, content: the challenge of the
information highway. Final report of the Information Highway Advisory Council, Ottawa, Industry
Merel, P. (1995) Response to consultation paper on the regulation of on-line information services,
Electronic Frontiers Australia
Rimm, M. (1995) Marketing pornography on the information superhighway: a survey of 917,410 images,
description, short stories and animations downloaded 8.5 million times by consumers in over 2000 cities in
forty countries, provinces and territories
Sansom, G. (1995) Illegal and offensive content on the information highway: a background paper. Ottawa,
Industry Canada; Spectrum, Information Technologies and Telecommunications Sector.
Schofield, J. (1996) This comment is free for now, The Guardian Online, 26 January
Stager, S.F. (1992) Computer Ethics violations: more questions than answers. EDUCOM Review , 27 (4),
Thimbleby, H. (1995) Problems in the global village
UKERNA (1995) JANET acceptable use policy, London, UKERNA,
Whitney, T. (1996) Personal communication, 18 January.