Ciolek, T. Matthew. 2003. The Internet and its users: The physical dimensions of cyberpolitics in Eastern Asia. A paper presented at the conference "From the Book to the Internet: Communications Technologies, Human Motions, and Cultural Formations in Eastern Asia," University of Oregon, October 16-18, 2003.

The Internet and its users:
The physical dimensions of cyberpolitics in Eastern Asia

Dr T. Matthew Ciolek,
Research School of Pacific and Asian Studies,
The Australian National University, Canberra ACT 0200, Australia

A paper presented at the conference "From the Book to the Internet: Communications Technologies, Human Motions, and Cultural Formations in Eastern Asia,"
University of Oregon, October 16-18, 2003.

Document created: 6 Oct 2003. Last revised: 13 Oct 2003.

0. Table of Contents

  1. Abstract
  2. Online information and its circulation in society
  3. Two notions: "Internet skills" and "Online users"
  4. The methodology
  5. The model: the 2003 scenario
  6. The model extrapolated: the 2009 scenario
  7. Magnitude and intensity of Internet skills in Eastern Asia
  8. Conclusions
  9. Tables with statistics
  10. References
  11. Acknowledgements
  12. Version and change history
  13. About the Author

1. Abstract

The energetic uptake of Internet technologies across Eastern Asia has placed fresh social, cultural and political demands on the region. Regional responses to these new pressures vary in both direction and strength. A new hierarchy of leaders (and strugglers) has emerged. Japan, Taiwan, and South Korea are found to lead the region in terms of Internet hardware, as well as their networked communication and publishing skills. They are closely followed by Hong Kong and Singapore. Some distance behind are China and Malaysia. Finally, Thailand, the Philippines, Indonesia, Vietnam, and Mongolia are under-equipped and under-skilled. These developments may also herald a possible bifurcation of cultures into online spectators and performers; a new form of a digital divide; and adjustments to the political power in some countries.

2. Online information and its circulation in society

Three distinct yet dovetailing realms of publicly circulated information shape contemporary societies and cultures.

First, there is oral tradition of communication, which exists in the form of face-to-face interactions, story telling, live TV broadcasts and other live performances.

Second, there is the recorded tradition of communication: the information that people depend on throughout their everyday comings and goings is stored, transposed and exchanged by means of various objects. For example, written texts, paintings, films, microfilms, or LP records.

Lastly, in more recent years we have witnessed the birth of a networked or online tradition. Within this third realm exchanges between participants - texts, numbers, images, movies, and audio recordings - form a fluid mosaic of electronic bits and bytes. These bits are then circulated in the shape of digital documents and files that are sent across a computer network.

All societies, places and historical periods can be analysed in terms of their range of communication traditions and the meanings and channels they routinely co-employ. They can also be profitably discussed in terms of the overall amount and energy with which those various communication modalities are put to daily use. In doing so one is able to discern the salient features of these societies or periods and develop a way for interoperable comparative analyses of their attainments and failures.

This paper's will maintain a narrow focus, examining only the developments in the world of online information.

Networked information and its vessels and channels (for instance: Internet hosts, hard- disks, telecommunication lines, databases, online directories, web-sites, e-books, chat-rooms, listserves, applets, and email software) are in great demand in the modern world. Their very existence carries several psychological and social consequences to those who provide them, as well as to those to rely on them. The more information is consumed on a regular basis, the greater becomes the need for it. The immense popularity of information exchange in the form of email, chat-rooms, web-browsing, web-cams and text-messaging in today's wired societies proves that information in its online guise is especially addictive.

This addiction has many roots. Networked digital information offers immense portability, minuscule cost per access, ease of copying, ease of transformation, stupendous speed of distribution on both the local and global scales, ease of documents annotation and search, automation of repetitive and over-demanding tasks. In short it offers to its aficionados an unprecedented swiftness, flexibility and convenience of action.

Consequently, the role of those individuals and groups who are able to provide their clients with periodic and ever more frequent "information fixes" cannot help but grow in stature. At the same time, their potential for exercising coercive and administrative power also keeps growing. Any denial of access to information inevitably disrupts deeply entrenched routines and expectations among its habitual users. Even a temporary interruption to the regular flow of information can put daily decisions on hold and create a sense of disorientation and loss.

All this leads to the inescapable conclusion that those with the ability to control or manipulate the content or the flow of online information necessarily accrue power and influence. That power and influence sooner or later will become invoked and exercised over those who voluntarily and involuntarily came to depend on online information.

As this paper will try to show, the amount and intensity of this power (and, simultaneously, the amount and intensity of this addiction) are quantifiable. The degree to which a society or a group of people is infused with power and addiction can be unambiguously stated and compared with similar measurements of other societies or groups.

Apart from the question of skill, the informational dominance over others - whether an individual or a whole group - appears to be a function of several additional factors.

Firstly, there is a question of evenness or parity in the distribution of online skills and knowledge among those who form a recurrent communicational arrangement. This is so because the degree of power and influence that one party has over others is not governed solely by the extent to which frustrated consumers can bypass the existing restrictions on their use of networked information. Such power also depends on the extent to which technical attempts to evade those restrictions can be detected and neutralised by those who have introduced them in the first place. To put it differently, those with similar online skills will be peers and equals, those with widely discrepant skills will not.

Furthermore, the overall calculus of online power and influence depends on two additional variables. One of them is the depth of one's need for the particular type of online information service, or the urgency with which a party wants to consult such a resource. The other variable is the degree to which the pursued online content or vessel is unique and irreplaceable (i.e. it is the question of monopolies operating on the whole or a subsection of networked resources).

In sum, in societies, which make regular use of the Internet, the are three factors which matter:

  1. demand for online information,
  2. degree of redundancy among online resources, and
  3. actual distribution of Internet skills across the population.

These factors can vary substantially. The hypothesis here is that it is the interplay of these three factors which accounts for some, at least, of the differences in distribution of informal and formal power in societies.

This hypothesis needs additional reflection and exploration. They will be provided in the next section of this paper.

3. Two notions: "Internet skills" and "Online users"

Statistics such as the following are common:

"Internet. Total users: 600 million. Penetration rate: 10pc of global population. Forecast users (2005): 1 billion. Forecast penetration rate (2005): 15 pc. Asia- Pacific users: 194 million. North American users: 176 million." Aylmer (2003).

Table 1 (Populations and the Internet in Eastern Asia in 2002-2003) offers similar summary information.

----------- Insert TABLE 1 here -----------

The highly generalised categories of "Online user", or "User of the Internet" are inadequate when they are deployed without an indication of what a person might be able to accomplish once he or she is placed in front of a PC hooked up to the Net. In fact, these categories are as clumsy and obfuscating as "user of printed paper" (i.e. a category which indiscriminately lumps together writers, editors, and readers, as well as those who will reach for printed sheets only to wrap a fish).

Internet literacy - that is, where an individual or a group has a good command of "Internet skills" - is a multi-faceted phenomenon. It is a differentiated set of concrete, everyday skills:

For these reasons variegated concepts "Internet literacy" and "Online users" constitute a far more hopeful departure point for research into social relations in cyberspace.

So, if we are keen to move our research beyond the mere restatement of and commentary on existing statistical data - i.e. beyond the approach taken in my study of electronic infrastructures of Eastern Asia (Ciolek 2002) - we need to go beyond the surface level at which these data were originally collected.

This could be done in several ways. For instance, one could try to fragment the general statistics on "online users" into demographic or occupational or socio- economic cohorts. Another way might be via a breakdown of that general notion into its component parts according to the criteria of geography, ethnicity, religion, political orientations, or life-styles. This paper will take that second path. It will delineate various types of "online users² in terms of the related levels of their "Internet literacy.²

The paper will also attempt to discern the possible social and political implications of patterns apparent in that disaggregated material.

4. The methodology

4.1. In praise of models

There are many reasons why one should feel gratified by the possibility of building and playing with models. By models I mean those intellectual artefacts that establish "a much simpler universe, [by] including some features of the actual universe and neglecting others." (Freeman 1999:xiv, cited in Staley 2003:58).

Models are useful to a researcher because they confer specific advantages.

Unlike grand theories (see for instance theories of societal and cultural affairs presented and reviewed in Skinner 1985), models by their very nature are not over-ambitious. They are built to do a specific task, and do their work only in the area of their competence.

Secondly, models (again, unlike large-scale theories) are modifiable, extensible, and gregarious. They can be broken apart, fiddled with, modified and enlarged. They may also be put to work in the context of other models, with a view to eventual integration with the operations of one or more additional models. One of the most endearing features of models is their ever-present incompleteness, their malleability and potential for endless change and enhancement. Indeed, perfect models do not need to be perfect to start with. In the course of research perfect models reach adequacy and usefulness only gradually.

An instantaneously successful model teaches the researcher nothing. It is far better to learn about improvements and refinements to a model in a truly Popperian fashion, that is, through an iterative series of careful conclusions and inferences. The most fertile way to learn about the world at large is not so much from the successes of our models or theories, but primarily from repeated confrontations (Popper 1994) with our methodological mistakes, biases and conceptual blunders.

Thirdly, models are crystal clear. Their initial "set-up" is always explicit, and their outcome, that is their resultant state/condition/value/shape after the "n" number of cycles/steps/iterations, is also plainly stated. Models are the ultimate champions of two pragmatic principles.

  1. "What you see is what you get" and
  2. "What you get is exactly what you put into it only transformed in accordance with your original explicit (and implicit) specifications."

Having said that, it is time that I proceeded directly to the heart of today's exercise.

4.2. The purpose of this model

I propose a model which attempts to reveal and describe the most likely distribution of Internet communication tools and skills among populations of the countries of Eastern Asia.

4.3. The boundaries of this model

The model focuses exclusively on developments involving the Internet (i.e. the digital communication network based on the TCP/IP communication protocols). It excludes until some unspecified future time any discussion of developments taking place amongst the other digital, but non-Internet communication networks as well as among those networks which are established for circulation of information in oral and recorded formats.

The model deals with data collected at the level of individual countries. It does not attempt a formulation of a trans-regional view. Nor does it offer a granular view in which data would be linked to individual provinces and cities, or specific ethnic groups or socio-economic classes.

The paper uses published statistics for three groups of countries only.

The paper examines the Internet in a broadly-defined Eastern Asia. For the purposes of our analyses that large-scale region comprises China, Hong Kong, Indonesia, Japan, Malaysia, Mongolia, the Philippines, South Korea, Singapore, Taiwan, Thailand, and Vietnam.

For the sake of comparison and to provide context, the East Asian data are put side-by side with those on the Internet in Scandinavia (i.e. Denmark, Finland, Iceland, Norway, and Sweden). Then, both Asian and Scandinavian materials are placed in the context of data stemming from Australia and USA.

The model has a circumscribed time span. Despite the fact that the Internet has a long and rich history - one which harks back to at least 1983, if not 1969 - our historical investigations will reach no more than 8 years back (i.e. till 1995), and our projections will not extend more than 6 years into the future (i.e. till 2009).

4.4. Variables comprising the model

4.4.1 Format
The collected data is presented, as much as possible, in the form of numbers and descriptive statistics. Each table will form a composite picture derived from several sources at once.

4.4.2 Data sourcing
The Population data are taken from the "World Factbook" (CIA 2003). In the case of all countries of all three regions all values pertain to the CIA's estimates for July 2002.

The data on the Internet hosts (i.e. the Internet hardware) are based on the results of the "Internet Domain Survey." The Internet Software Consortium, aka the ISC (2003) conducts this survey twice a year (in January and July). In this paper the 2003 data pertain to the values collected for January 2003. (At the time of writing, July-September 2003, no ISC data were available for July 2003).

The data on so-called ²users of the Internet² relies on statistics reported in the "World Factbook" (CIA 2003). Most of the material presented in Table 1 (Populations and the Internet in Eastern Asia in 2002-2003) comes from 2002. It is only in the case of two countries, Taiwan and Thailand, that the data refer to the situation in 2001.

In addition to "contemporary" materials on online users as provided by the CIA, the paper also uses "historical" data published by the Nua Internet (2003). That historical material, as presented by the Nua web site, has been gathered by various agencies and organisations according to several, often dissimilar methodologies, and according to many schedules. Thus the earliest available information on the number of online users varies from country to country: Australia - September 1997, China - an average from values reported for June and August 1997, Denmark - September 1997, Finland - September 1997, Hong Kong - August 1997, Iceland - February 1998, Indonesia - December 1997, Japan - September 1997, Malaysia - an average from values reported for March and October 1997, Mongolia - December 1999, Norway - September 1997, Philippines - August 1997, S. Korea - February 1997, Singapore - September 1997, Sweden - July 1997, Taiwan - June 1996, Thailand - January 1998, USA sometime in 1995, Vietnam - June 1998.

4.4.3 Terminology/Taxonomy and its genesis
As it has been already noted, the general term "online user" or "user of the Internet" is inadequate for our purposes. Therefore, the term will be systematically broken into smaller component parts.

The subdivision of the term "user" is based on the analysis of pilot data on the relative complexity of online tasks. These materials come from an interview conducted by this author with seven computer programmers and navigators of the Net (Burkey 2003, T. Ciolek 2003, Johnson 2003, Hurle 2003, Payne 2003, Walker 2003, Young 2003).

The subjects' experience with computers, computing, networks and the Internet is considerable. For example, Mr Darrell Burkey, RSPAS, ANU has worked for 15 years with computers and telecommunications. Other respondents have similar or better exposure to those technologies. The numbers in round brackets indicate the length of their professional experience in those two areas: Mr Thomas Ciolek, Dreamcraft Pty Ltd. (12 years), Dr Ian Johnson, Archaeological Computing Laboratory, University of Sydney (26 years), Mr Rob Hurle, RSPAS, ANU (39 years), Ms Lyn Payne, RSPAS, ANU (15 years), Ms Helen Walker, RSSS, ANU (13 years), and Mr Greg Young, Asian Studies, ANU (20 years).

Each respondent was separately asked to rank the relative richness and complexity of computer and network skills necessary for the successful completion of some 22 online tasks. Scores given to each of those tasks could range from 1-10 points. The sequence of tasks to be ranked was presented to them in random order. The task list is quite arbitrary, and is far from being comprehensive. It was developed in early July 2003 in the course of a brief, informal, brainstorming session between this author and Mr T. Ciolek. While the list of online tasks and activities could be extended, it seems to cover the whole spectrum of a person¹s major involvements with the Net. Also, a line plot of cumulative scores "earned" by these tasks shows that, when the tasks are sorted according to their apparent complexity, they form a continuous and smoothly rising line. This indicates that there are no major discontinuities on our ordinal scale.

The results of this preliminary investigation are reported in Table 2.

----------- Insert TABLE 2 here -----------

Table 2 postulates that there are at least three distinct levels of Internet skills, or the three phases of Internet literacy. Each of those levels, namely "Basic skills"(1-33%), "Moderate skills" (34-66%), "Advanced skills" (67-100%) is defined as the respective tercile on the scale listing the cumulative "skill value" for a sequence of increasingly complex online activities.

Closer inspection of Table 2 indicates that the postulated three phases of Internet literacy correspond also to three different attitudes and online roles:

Naturally, the allocated points represent, at the moment, the opinion of seven respondents only. Future interviews with additional experts in the field of computing and networking are bound to shift the boundaries between various levels of online competence. However, even at this early stage it is obvious that: (a) the conceptual, technical and navigational skills necessary for an extensive and trouble-free use of the Internet are wide-ranging. This pilot investigation has looked at 22 components of Internet literacy. More ambitious studies can easily develop a more comprehensive catalogue of Internet skills; (b) Importantly, each of those top-level skills can easily be further subdivided into separate series of component sub-skills. The questionnaire can, therefore, be expanded to resemble a collection of nested "Russian dolls"; (c) Skills which are a part of overall Internet literacy constitute a natural hierarchy based on their inherent complexity, difficulty and associated know-how. All in all, the pilot methodology proved entirely workable. It can be built on, refined and incorporated into similar future investigations.

4.5. Relationships between the variables

The model outlined in this paper assumes the existence of a couple of positive feedback loops. More precisely, it assumes that:

4.6. Variables and their possible transformations

The model also assumes that the constituent variables are locked into a cycle of fast and prolonged growth. More specifically, it assumes that:

The number of Internet hosts will to continue to grow for a very long time. As a consequence, the final number of Internet hosts in operation in a given country may come to equal or even exceed the number of people in that country.

The number of "Internet users" in a given country may also reach very high numbers. In fact, this growth appears to have three parallel components: (a) the number of people with sporadic access (say, a few hours a month or a week) to the Net; (b) the number of people with regular access (say, a few hours a day) to the Net; (c) the number of people with permanent/pervasive and high speed access (i.e. at any moment, from anywhere, in a matter of seconds) to the Net. Therefore, we can guess that once a society reaches a level where 100% of its population is said to be "users of the Internet", such a situation will mean that all those people are said to have been assured at least sporadic access to the Net.

As new Internet-based resources, services and technologies appear online the overall gamut of Internet skills is bound to evolve. It is bound to become increasingly broader and more complex. Therefore, with the passage of time the difference between entry-level Internet skills and high-end skills will become increasingly more pronounced. This gap will also continue to grow because of the steady introduction of both streamlined and "idiot-proofed" online information services and complicated high-end online operations which will make firm demands on the technological know-how, attentiveness and time of its clientele.

4.7. Variables and their initial values

The notion of an initial value of any variable refers to two distinct possibilities. Firstly, it may be a value, or a state assumed by a given variable at a particular point in time. Secondly, it may be a specific value taken together with its trend, namely the direction and rate of change of that value. Such a trend, naturally, can be determined only by looking at a series of values assumed by the variable through time. This paper will specify the initial values of both kinds. A set of initial values understood as depicted at a particular point in time is offered in Table 1. (Populations and the Internet in East Asia in 2002-2003), That table presents information on Internet-related developments in Eastern Asia and Scandinavia (and, for the sake of additional context, in Australia and USA).

Most of the data presented by Table 1 is self-explanatory.

Naturally, it is useful for some of those values to be observed over time. Accordingly a temporal view is offered by Table 3, "Growth in the number of Internet hosts in Eastern Asia 1995-2003."

----------- Insert TABLE 3 here -----------

In the table above, data about numbers of networked computers are compiled for the following five years: 1995, 1997, 1999, 2001 and 2003. In addition, for each country and each broader geographical region the table calculates the overall size of growth in installed Internet hardware. Table 3 also calculates the changing rates of growth for the four consecutive two-year periods 1995-1997, 1997-1999, 1999-2001, and finally 2001-2003. The resultant picture is very interesting. Within 8 years and in terms of absolute numbers of Internet hosts, the block of Eastern Asian countries (over 150 thousands hosts in 1995, over 13 million hosts in 2003), has decisively overtaken the five Scandinavian countries (a total of over 220 thousands hosts in 1995, and over 4 million hosts in 2003). The East Asian growth rates are astounding too. They range from growth of no less than 559 times (Vietnam) to a 275-fold growth in China, and a 23-fold growth in South Korea. The comparable values for Scandinavia are far more tame. They range from growth of about 44 times (Denmark) to growth of about 15 times (the remaining four Scandinavian countries).

Other figures of interest describe the rate of growth expressed over time. See Table 4 (Internet hosts in Eastern Asia 1995-2003: changing growth rates)

----------- Insert TABLE 4 here -----------

Data from Table 4 reveals that with the passage of time, the rates of growth in all three groups of countries gradually slow down. Naturally, these are large-scale regional trends, which are not necessarily repeated at the level of individual countries. For instance, Taiwan, Hong Kong, China and Mongolia are examples of countries where the rate of uptake of Internet hardware strongly oscillates from year to year.

The next table, Table 5, deals with details of "Recorded Internet users in Eastern Asia 1996-2002."

----------- Insert TABLE 5 here -----------

Table 5 looks at the speed with which the number of people with access to the Internet has grown in the three studied areas. The speediest annual increases in numbers of "Internet users" occurred in China (a 57- fold increase from year to year), Vietnam (19-fold annual increase), and Malaysia (15-fold annual increase). By contrast, the countries of Scandinavia, as well as Japan, Hong Kong, Singapore, Australia and USA have been gaining fresh "users of the Net" at much slower rates.

Armed with the above information, we are now ready to estimate the number of "online users" in 2003. This is accomplished in Table 6 (Estimated Internet users in Eastern Asia in 2003).

----------- Insert TABLE 6 here -----------

Table 6 deals with people and so dovetails with host information offered by Table 1. Together, the 2003 values from both tables enable us to construct a simple model that proposes a structural link between these two variables.

5. The model: the 2003 scenario

Firstly, we need to agree that the data regarding numbers of the Internet hardware installed in various countries of the world carries information not only about the amount of the hardware itself, but also about the amount of its most immediate users, and that most those users are likely to be highly skilled ones. Each of the installed machines needs one or more person who knows how to set it up, install an operating system, fix problems, install software, connect to the network, test all the systems and subsystems, establish mechanisms for backups and graceful degradation of services, provide firewalls and other protection against external disturbances, and set-up the administration and support of the computer's day-to- day operations. In short, the very presence of a piece of Internet machinery in a given institution, place or country is indicative of the presence of capable, highly trained technical staff.

The exact ratio of the number of hosts in existence to the number of informed and skilled users is, at present, uncertain. However, an inspection of the Taiwanese data in Table 2 suggests that such a ratio could very well be close to one-to-one. If we accept that value (i.e. one Internet host equals one highly skilled user of the Internet) then, on the basis of data about hosts and numbers of all users of the Internet in a given country we can calculate the approximate number of those users who are not so well skilled. This operation is carried out in Table 7 (Estimated advanced and other online users in Eastern Asia in 2003).

----------- Insert TABLE 7 here -----------

Once the data from Table 7 are available it is a simple operation to calculate approximate numbers of the three different types of "Internet users" for each country. Also we are able to determine the degree to which they permeate the societies of each country

In our model the Network skills among people who are less skilled than those who establish and manage Internet hosts are assumed to be distributed very unevenly. Having assumed that persons directly related to the operation and well-being of Internet hosts are the same as those with "Advanced" Internet skills (see again Table 2), then how do we divide the residual category (i.e. "Other" users) between those with "Moderate" skills and those with "Basic" skills? Several scenarios can be envisaged here. For instance, Rob Hurle, a person with over 39 years experience with mainframe computers, suggested (Hurle 2003) in a conversation with this author that the proportion of those with "Moderate skills" to those with basic "Basic skills" could be as low as 1:9. However, in this paper, I shall make a less conservative guess, and set the ratio as being 3:7. In short, once we discount the number of people directly involved in the establishment and operation of Internet hosts, for every ten "Other" online users seven of them are assumed to have "Basic" Internet skills, and the remaining three - "Moderate" skills.

Therefore, their actual numbers are estimated to be as follows (see Table 8):

----------- Insert TABLE 8 here -----------

If those values are expressed as a fraction of the total population of every country in question, then we are ready to generate Table 9.

----------- Insert TABLE 9 here -----------

In fact Tables 8 and 9, constructed from bits and pieces of data provided by all the previous tables and charts, represent two sides of the same model.

Naturally, data in the shape of mere numbers are not effective communicators. Therefore, information offered by Table 9 will be slightly re-ordered and republished here as Chart 9a (Estimated distribution of Internet skills in Eastern Asia, percentages), and again in the form of a close-up (Chart 9b), which shows in detail how Internet skills appear to be distributed within the first 10% of each population. Finally, for the sake of context the East Asian data are contrasted with data about parallel developments in Scandinavia, Australia and USA (see Chart 9c).

----------- Insert CHART 9a here -----------

----------- Insert CHART 9b here -----------

----------- Insert CHART 9c here -----------

The most important conclusion to be drawn from Tables 8 and 9, as well as Charts 9a, 9b and 9c is that in terms of Internet usage, Eastern Asia is made of two very distinct groups of countries.

In Taiwan, Singapore, Japan, and (to some extent) Hong Kong, "Advanced" and "Moderate" Internet skills are entrenched. These four countries could be justifiably nicknamed regional "info-Tigers." (The latter term was first used in Ciolek 2001).

In Vietnam, Mongolia, Indonesia, China, the Philippines, and Thailand, on the other hand, the per capita number of Internet users abysmally low. And more significantly, the relative numbers per capita of those who have more than "Basic" skills are so low that they are nearly non-existent. These six countries could be called regional "info-Sparrows."

Furthermore, comparisons of these East Asian data with materials describing developments in Australia, US and Scandinavia quickly reveal that Taiwan, Singapore, Japan, and Hong Kong resemble these seven "Western" countries far more than their six Asian neighbours.

Finally, it is worth noting that in 2003 Malaysia and South Korea represent borderline cases. In terms of the percentage of population represented by people with any Internet skills those two countries were situated closer to the "info-Tigers" than to the "info-Sparrows." However, the converse was true when one looked at their relative numbers of users with "advanced" skills.

Naturally, an immediate question arises. Is what we observe, a purely incidental situation, or a longer-term structural feature?

6. The model extrapolated: the 2009 scenario

Tables 10, 11, 12 and 13, provide the answer to this last question. Table 10 (Projected growth in the number of Internet hosts in countries of Eastern Asia 2003-2009) calculates the likely number of Internet hosts in years 2005, 2007 and 2009.

----------- Insert TABLE 10 here -----------

Table 10 extrapolates from the host values in 2003. It subjects them to three consecutive multiplications, each time by the average biennial growth rate, and each time by the factor of 80% of that growth (see again Table 4). This later operation reflects the ever-slowing pace with which countries acquire new hardware.

Table 10 which deals with hosts servers, is naturally a close colleague of Table 11 (Projected growth in the number of online users in countries of Eastern Asia 2003-2009).

----------- Insert TABLE 11 here -----------

As before, both tables can be used to construct Table 12 (Estimated distribution of Internet skills in Eastern Asia in 2009, numbers) and Table 13 (Estimated distribution of Internet skills in Eastern Asia in 2009, percentages).

----------- Insert TABLE 12 here -----------

----------- Insert TABLE 13 here -----------

These extrapolations, which look at the most likely developments six years from now, indicate that by 2009 the fundamental divide between Internet skills in people in Taiwan, Japan, Singapore, and Hong Kong on the one hand, and in the rest of the Eastern Asia, on the other, will have become more marked. Incidentally, this result also confirms the view that such a divide is a structural and long-term feature of the country-to-country relationships in cyberspace, and not a momentary blip in the statistics.

7. Magnitude and intensity of Internet skills in Eastern Asia

The above analyses of the distribution of Internet skills in various Eastern Asian populations are interesting, but still they do not tell the full story. Such a story, in order to be formulated and narrated, definitely needs an extra dimension. Such a story requires the juxtaposition of relative (percentage wise) values with their absolute numbers.

The underlying idea is, actually, very simple. Whereas the saturation of a population with people who seem to have more than basic Internet skill can be thought to reflect the "intensity" with which such skills are present in a country, their raw numbers can be viewed as the "magnitude" or volume of such skills. Both the intensity and magnitude can be measured and expressed in numerical form. This can be easily done by multiplying the actual numbers (and percentages) of people with a certain skill-level by a value which expresses the complexity and sophistication of that level of skill.

Table 2 (Three levels of Internet skills) guides us in choosing such multipliers. As we remember from Table 2, the entire range of skills is divided into three hierarchically arranged clusters, with each cluster of skills having its own score for its inherent complexity (i.e. "Basic" skills 1%-33%, "Moderate" skills 34%-66%, "Advanced" skills 67%-100%). On the basis of this information we can establish that the mid-range scores for each of the three skill- levels are: 17, 49 and 82 points, respectively.

Armed with this information, we can calculate "magnitudes" and "intensities" of the online skills for the countries in our sample. The results of those operations are expressed in Table 14 (Magnitude of Internet skills in Eastern Asia in 2003) and Table 15 (Intensity of Internet skills in Eastern Asia in 2003).

----------- Insert TABLE 14 here -----------

----------- Insert TABLE 15 here -----------

Finally, when for every country in our sample we multiply the "magnitude" score of its skills by the "intensity" of such skills we are able to arrive at a coefficient, an index which synthetically describes a country's (or group's) potential to tap and systematically exploit the extent and depth of its online expertise. Such a coefficient will be called the "Internet Power Index " or IPI, for short. The IPI can also be seen as a measure of country's dependence on online information.

----------- Insert TABLE 16 here -----------

Data from Table 16 can be graphed and are presented in visual form as Charts 16a and 16b. The first of the charts shows developments in a number of places across the globe. The second chart focuses on Eastern Asian region.

----------- Insert CHART 16a here -----------

----------- Insert CHART 16b here -----------

The first chart shows that in terms of ability to manage and control the shape of the world of online information, the most dominant country in the world is USA (a score over 14,000 IPI points). The equivalent Japanese score is worth about 2,000 points, and Taiwan's ranking is about 1,000 points strong. We can also see (in detail, from Chart 17b) that South Korea (497 points) is less powerful on the Net than Australia (645 points). Furthermore Hong Kong (143 points) and Singapore (90 points) are less powerful online entities than the Net players represented by Finland (163 points) and Norway (143 points). Finally we can see that in terms of our Internet Power Index, Malaysia (48 points) scores slightly higher than China (41 points), which in turn has more online clout than Iceland (24 points).

In other words, the model finds that among countries of East Asia there are very strong divisions with regard to the countries' extent and depth of skills required for effective operations in the realm of networked information.

When the 12 countries of Eastern Asia are compared in terms of their Internet Power Index, they form a clear-cut hierarchy, one which did not exist at all a few years ago. On that vertical ladder every country has approximately half of the points scored by its immediate higher ranking neighbour (and twice as many points as its immediate lower ranking neighbour). However, this delightful regularity in 2003 data seems to be purely accidental. With the passage of time it will get replaced by some other numeric pattern, or - more likely - by no pattern at all. In Eastern Asia the IPI hierarchy in 2003 is as follows:

There is an additional pattern: when the country data on the magnitude and intensity of Internet skills are put side by side (see Table 17) an interesting interaction between the two variables is apparent.

----------- Insert TABLE 17 here -----------

Such an interaction is uncovered through the labelling of every studied country either as a high performer (HI), or as a low performer (LO). This is done first in terms of the magnitude, and then again in terms of the intensity of the country's Internet skills. Such a HI or LO label is allocated depending on whether the relevant score attained by the individual country is higher or lower than the region's average.

Once the labelling is completed, a cross-tabulation of the results reveals that in Eastern Asia there are four distinct groups of societies. In fact, each of those groups is representative of a particular strategy or scenario regarding the uses of the Internet.

This four-fold division among the countries is almost precisely echoed by their placement on the IPI scale. The three top scoring countries (i.e. Japan, Taiwan, and South Korea) are those which comprise the HIHI group. Next, below the three leaders, there are the two LOHI countries, namely Hong Kong and Singapore. The third echelon is formed by China, the sole representative of the HILO scenario. Finally, the lowest rung of the IPI scale is occupied by the LOLO group, that is by Thailand, the Philippines, Indonesia, Vietnam, and Mongolia.

Malaysia is the only country which does not conform with the above pattern. In fact, Malaysia seems to represent a border-line case. According to Table 17 that country has a 5.50% score for the "intensity" of its Internet skills. Such score is lower than, but within a short distance from, the cut-off value of 8.4% which - if attained - would have allowed Malaysia to be elevated from its present place in the LOLO group to the LOHI (i.e. Hong Kong and Singapore) group. Detailed inspection of Table 13 (Projected distribution of Internet skills in Eastern Asia in 2009, percentages) indicates that by year 2009 the percentage of people in Malaysia with better than basic Internet skills is likely to increase from the current rate of about 9% of total population to about 31%. It is likely, therefore, that in a few years that country can be re-classified as the LOHI type of society.

Our results show also that in its encounter with information technology China's progress has clearly been held back by its enormous population. China is a HILO country with a large overall pool of good Internet skills. Nevertheless, it is going to face tremendous difficulties with educating many hundreds of millions of its citizens in advanced ways of dealing with the Internet. Such a daunting task is necessary before that country will able to join the ranks of the more advanced and more exclusive HIHI group.

Finally, we can only wonder with regard to the nature and mix of domestic factors that have placed Thailand, Philippines, Indonesia, Vietnam, and Mongolia within the vicious circle of the LOLO scenario. In the LOLO countries small volumes of the Internet hardware are not conducive to the widespread uses of the Network. Simultaneously, small volumes of online users are not conducive to greatly expanded acquisition of such hardware. How firm that trap will prove to be and which of these five countries will be able to escape it first, if at all, is not clear at this stage.

8. Conclusions

It is time to wind-down our frantic modelling activities, and look at the main lessons learned so far. Our conclusions form two groups. The first one deals with the model which we used to analyse the data. The second group deals with the findings of our analysis.

8.1. Evaluation of the model

The model which we have developed and worked with is chunky and very rough. It remains, to some extent, a Frankenstein-like creature, for it has been cobbled together from a number of not-too-well-fitting parts. Fortunately, the model, unlike Frankenstein's monster, has worked obediently. At the same time, it is obvious that several improvements might be required for any future modelling operations.

Firstly, we need better baseline data, especially that dealing with the current and future sizes of populations. The present model has used the CIA's population estimates for July 2002 and has frozen them at that awkward level not only for analyses applied to the year 2003, but also for analyses dealing with the future, i.e. with developments in 2005, 2007 and 2009. Certainly, this methodological short-cut needs to be corrected.

Secondly, the current model assumes that the user numbers in various countries increase annually by a fixed amount (see Table 6). This is not a realistic assumption. Instead, we need to determine the exact percentage values for the annual growth rates. Also, we need to ascertain whether those values tend to remain steady, grow or decline. Thirdly, the model needs to be buttressed by sociological field work. Such a study is needed to establish by observational means the actual proportions between the numbers of Internet hosts and "Advanced" users of the Net. Such observational data should be used to replace the currently deployed constant (i.e. the ratio assumed to be 1:1). Also one has to determine, again by means of field observations and experiments, the actual ratios of those who are "Moderately" skilled in the use of the Net, to those whose skills are only "Basic" ones. The constant which is currently employed in the model is set at the 3:7 level, and it needs to be verified and adjusted, if necessary. Also, it is very likely that such a ratio might not be constant at all but variable not only from one geographical area to another, but also from one period to another in the Internet's three-decade history.

8.2. The findings

We have spent a quite a while dissecting and comparing numbers. Eventually, we need to step back, and ask ourselves "what do all these figures mean?"

Well, there are at least three answers to such a question.

8.2.1. A cultural split
The rapid uptake of the Internet in Eastern Asia noted above seems to coincide with two long-term cultural processes unfolding in the countries of the region. The first process is about gradual evolution from the emphasis on recorded modes of publicly circulating information to networked ones. The other process deals with either retention or gradual loss of control over integrity and self-reliance in one's communicational arrangements.

The new dimensions of the digital divide appear to indicate that a major cultural schism might be underway. That schism seems to take place between various organisations and groups, different subsections of the same population, as well as between neighbouring countries.

In the case of some human groups the bulk of public information tends to be created and exchanged largely by means of oral and recorded traditions. If online information is accessed there at all, then it is done mainly for the purposes of consultation. Such a culture can be called a culture of online spectatorship.

In other cases, however, the operation of oral and recorded communicational modalities is strongly augmented by networked channels of communication. In those situations if online information is accessed, the chances are that it will be not only read but also written; not only used, but also contributed to; and finally, not only downloaded but also modified. Therefore, such a culture can be called a culture of online performers.

Naturally, whenever these two sets of people enter a recurrent communicational arrangement, that is, whenever they start sharing the same portion of cyberspace, the "performers" cannot fail but exert intellectual and cultural dominance over the "spectators."

The consequences of that cultural split are huge. Mediaeval European and Asian villagers had to rely on technicians - village or temple scribes - in order to communicate effectively with the world. Increasingly often we, who are often highly trained and thoughtful people, are in a situation not much different from that of a mediaeval peasant. And that after a century or two of enforced literacy and the resultant freedoms in our daily communicational arrangements.

This is because with the introduction of networked technologies the act of communication ceased to be a straightforward, domestic-scale affair. Surely, we keep gaining on speed and versatility in our private and public exchanges, but this occurs at a great cost to our very existence as self-reliant participants in our societies and our cultures. The more we become immersed in the world of networked information, the more fully we trade our communicational competence (one which was first developed, tested and demonstrated in the realms of oral and recorded information) for global reach, miraculous speed and pervasive convenience. We keep gaining the capacity to interact with the world, yet we do so at the expense of our ability to control the production of our electronic utterances, and our control over its networked distribution.

This last point is forcefully made by Elvin's (2003) observation that while many of his Chinese friends, who are historians, can fluently work with French and German books and other paper publications, they - nevertheless - find themselves unable to read French and German online documents simply because they do not know how to switch their computers from a Chinese font set, to the one which displays diacritics in Western languages.

Obviously, accelerated courses in computer and Internet literacy organised for as many people as possible and in the shortest possible time are not the answer. Over the last 30 years Internet-based information and communication tools have grown technically more complex far faster than their users' skills to trace, unravel and master them. The so-called leading edge of networked information technologies resembles more closely the surface of a bursting supernova, and than the edge of a kitchen knife. The growing chasm between what the Internet is capable of, what it really does, and what people who use it can actually do with it, is the fundamental and structural aspect of our relations with the information in the networked format. The basis for the emergence of the two types of cultures outlined above is firmly set.

8.2.2. Digital divide revisited
Our intellectual experiments with the revised and extended notions of Internet skills and online users, suggest that there is another research concept which needs to be investigated and revised: the notion of a digital divide.

The current colloquial meaning of that term refers to the presence of a serious split within a population of a country, or geographical area, or socio-economic or ethnic groups. That split, which is discussed at some length by Anonymous (2003), creates the digital haves and have-nots, that is those who have access to the Net and who benefit from its online resources, as well as those who do not have such access, and are - therefore - strongly disadvantaged in their attempts to function as fully fledged citizens of a modern and equitable society. However, the notion of a divide, while technically correct, is - nevertheless - incomplete. It fails to take into account the possibility that even among those people who nominally seem to have identically easy and frequent access to the Net, there are bound to be individuals and groups who, due to the lack of adequate online skills, are unable to make any meaningful contribution to the realm of online information, and therefore - as citizens of a society - are relegated to progressive irrelevance and oblivion.

In other words, the Internet, which only a few years ago was touted as a great global levelling force (see, for instance, Rutkowski 1994), is itself divisively and restrictively distributed and proves to be, paradoxically, a divisive and restrictive way of circulating information.

The emerging picture is ironic in the extreme. Over the last ten years or so the existence of great masses of poor and uneducated people who are unable to access or express information in the written format started being echoed by the emergence of equally numerous masses of well-situated and well-schooled people who, nevertheless, day after day are incapacitated and de-skilled by their regular exposure to the Internet.

They are incapacitated and de-skilled by electronic publishing technology which over few years has grown so ridiculously complex that now only full-time web-masters can manage it efficiently. They are hampered by bouts of computer viruses, by incompatible file formats, by the vagaries of their ISPs, not to mention by an imposition on their networked operations of officially and unofficially sponsored filters (see for instance, Zittrain and Edelman 2002), censorship and electronic eavesdropping (Reporters Sans Frontieres 2003). Ultimately, they are incapacitated and disenfranchised by their own inability to take control of their digital lives.

Due to all these factors, the Internet is far from being a benign and gently accommodating medium. From time to time the very exposure to the Internet can slow people down, disable their creative operations, or even lock or erase their personal collections of electronic documents. It is far amore mercurial medium than pen and paper, a typewriter, a printing press, or a photocopier. In fact, the Internet has a Janus-face, with which it can easily overwhelm an unwary user.

On such occasions masses of people are relegated to the wrong side of the digital divide simply because they are unable to sort out by themselves the logistics of their networked operations and, instead, must rely on the good will and good faith of others, that is the network technicians and experts.

In a sense, it is an inevitable process. Our pursuit of information, and our contacts with fellow human beings, as soon as they enter online, necessarily become a strongly mediated pursuit and contact. It is mediated by the complicated and fragile technology which continues to grow complex and fragile far beyond our comprehension. Moreover, our communicational activities are additionally mediated by human operators who can impact on our electronic well-being, and with whom and their intentions we are not familiar at all.

8.2.3. The political dimensions
The division of populations of Eastern Asia, as well as the rest of the world, into two differently skilled classes of people alerts us to the rise of yet another set of phenomena.

If, as I have noted earlier, the power which stems from the ability to affect the flow of networked communications can be accrued by individuals, groups, organisations, and states, and if such power can be exercised in relation to the less-skilled online users, then some institutions and societies may undergo a major transformation.

In addition to all other existing forms of stratification, a fresh one which is based on people's online competence will arise. Sometimes that power, which is a by-product of people's addiction to online information and their inability to take care of their own online affairs, will grow slowly and involuntarily. In such situations the holders of such power might be gradually co-opted by and diffused amongst other groups and subdivisions whether in an organisation or within a society. However, there could be also times when the accrual of such power might be rapid and deliberate.

Therefore, a question arises regarding the political options open to those who realise that there are now a large number of them, and that they posses a rare and valuable skill, and that the have a capacity to exert close control over the informational activities of other, less skilled people.

Will such experts try to capitalise on that opportunity? Will they silently form an equivalent of the praetorian guard in the employ and service of businesses and governments? Or alternatively, will they try to establish a guild, or a professional association which will champion and safeguard their economical well-being, their privileges and high social status? Or, perhaps, will they be tempted to form a separate social class which will be conscious of its uniqueness and its strength? Such a new class would be that of new managers and mandarins, i.e. ones who will try to accrue even more power, and translate it into the political control of the rest of the society.

Judging by the way the magnitude and intensity of Internet skills are distributed in Eastern Asia, the above political scenarios may unfold very differently in each of the four HI/LO societies. We might guess that in the HIHI and LOHI countries where good Internet skills are well spread across the bulk of the population those who are Internet literate will be commonplace and as such will not enjoy any special treatment.

However, in the LOLO countries it is very likely that people with Internet skills might form a small elite, and that such an elite will be enrolled to buttress the pre-existing power structures. Finally, in the case of the HILO countries, at present exemplified by China, but also likely to be represented by other states in Asia and elsewhere, the chances are that people with Internet expertise might also be the very ones who in the near future will gain a substantial share of political power.

Such political rearrangements, however, are not likely to eventuate through any kind of a take-over of the HILO countries by those knowledgeable in the ways of the Net. Most likely it might take place through the current rulers' (say, an army, a political party, a religious establishment, or a cabal of large businesses and media) determination that only the trustworthy people whose allegiances are known and tested shall be given the necessary training together with employment in the critical areas of computers and networking.

Furthermore, this rearrangement might also be reinforced through rulers' determined efforts that the monopoly on strategic intellectual skills is not breached by those who are deemed by the digital gate-keepers to be unworthy of possessing it.

9. Tables with statistics

10. References

11. Acknowledgements

I am grateful to Monika Ciolek, Irena Goltz and Greg Young for their illuminating comments on the first draft of this paper.

12. Version and change history

13. About the Author

Dr T. Matthew Ciolek, an archaeologist and anthropologist, heads the Internet Publications Bureau, Research School of Pacific and Asian Studies (RSPAS), The National Institute for Asia and the Pacific, The Australian National University, Canberra, Australia. Since December 1991 he has been responsible for making the RSPAS' research materials and expertise available to the Internet community via a range of online technologies. He is a pioneer in networked research and communication related to the Asia-Pacific region, and creator and editor of an electronic journal "Asian Studies WWW Monitor" [Est. Apr 1994] ( and numerous subject guides to the Internet, including the influential Asian Studies WWW Virtual Library [Est. Mar 1994] (

Site Meter
visitors to since 08 May 1997.

Maintainer: Dr T.Matthew Ciolek (

Copyright (c) 2003 by T.Matthew Ciolek. All rights reserved.