iMP Banner subscribe iMPcontents -- October 2000November 2000
What Counts on the Net? Frameworks and Issues in Measuring the Internet

"Communication policy is social policy, and effective policy makers require accurate information as to where we stand and where we are heading. One aspect of this is an understanding of the global diffusion of the Internet."

Larry Press

Larry Press is Professor of Computer Information Systems at California State University, Dominguez Hills and a Contributing Editor for the Internet Society publication OnTheInternet. Dr. Press studies the global diffusion of the Internet. He last wrote for iMP on electronic books in June 2000.

As I write this, laborers on land and sea are laying cables around the world. We are investing heavily in communication networks that will impact our societies and lives. Communication policy is social policy, and effective policy makers require accurate information as to where we stand and where we are heading. One aspect of this is an understanding of the global diffusion of the Internet. This article discusses indicators and broader frameworks for tracking the global state of the Internet or the state of the Internet in a nation or region. We begin with an examination of some of the indicators that have been used to characterize the Internet


Today’s Internet is a vast, decentralized maze of interconnected networks. Host computers with names like www. cisp. org connect to local networks which are tied together through regional, national and global backbone networks. The Internet grows from the edges and its topology is in a state of constant flux, making it difficult to measure its size and extent. In the early days, Internet measurement was relatively simple. The growth of traffic (a measure of usage) and nodes (a measure of the physical infrastructure) were regularly recorded on the ARPA Net, which preceded the Internet.[1] The National Science Foundation (NSF) ushered in the “modern” Internet by establishing a 13-city backbone in 1988 (Figure 1), but measurement of numbers of nodes and volume of traffic, remained straightforward. Since all U.S. Networks connected to this backbone, NSF traffic and connectivity data, e. g., Figure 2, gave us a picture of Internet. When the Internet spread to other nations, their networks also generally connected to and routed traffic through the NSF backbone (see Figure 3), so NSF statistics continued to reflect the state of the Internet.

Figure 1


Figure 2

Source: Merit Network,


Figure 3

Source: Merit Network,


NSF was not the sole source of Internet data during that era. In 1991, Larry Landwebber the University of Wisconsin began publishing biannual maps showing which nations had some connection (perhaps quite minimal) to the Internet. Mark Lottor of Network Wizards also began reporting the number of Internet hosts before the advent of the NSF Network in 1988. His host counts, Figure 4, are probably the most frequently cited, longest running data series showing the Internet growth.

Figure 4

Note: The counting method was adjusted in 1995. Prior to that time, a program "crawled" from network to network issuing a "show all names" command to each domain-name server. When system administrators began filtering out those commands for unrelated security purposes, ISC switched to a program that generates the set of possibly assigned addresses, and tests to see whether they have in fact been assigned names.
Source: Internet Software Consortium,


But the NSF Network began phasing out in 1995, and today we have hundreds of national and international backbone networks connecting hundreds of millions of users and hosts through hundreds of thousands of service providers and networks. Tracking the diffusion of such a complex, dynamic network is a daunting task, but many organizations have come forward to monitor Internet-related indicators. [2]For instance we see measures of numbers of users, hosts, Web sites, languages used, infrastructure availability, etc.

I surveyed a number of these indicators recently,[3] and found both rapid growth in nearly every aspect of the Internet as well as growing gaps between and within nations. For instance, we saw rapid growth illustrated in the host counts of Figure 4, but if we dig deeper, we see that those hosts are distributed very unevenly. Africa has only a tiny fraction of the world’s hosts, and, as shown Figure 5, the great majority of those are in South Africa. If we were to drill down further, we would find a very uneven distribution of hosts within South Africa.

Figure 5. Internet Hosts in Africa, 1999

Source: Internet Software Consortium, www. isc. org, figure prepared by Mike Jensen, htm


While Internet indicators are collected and disseminated by many organizations, we should not over estimate their precision or reliability. An indicator may be poorly defined or defined differently by different people measuring it. For example, we see estimates of the number of users in a nation, but what is the definition of a “user?” How often and for what purpose must one use the Internet to qualify? Is someone who reads e-mail messages once a month in a community center near a rural village in India to be counted as a user? Is someone who has browsed the Web a few times at a public library in California a user? Regardless of the definition, how do we count users? Many surveys merely estimate the number of ISP accounts, perhaps adjusting it to reflect shared accounts in a poor nation.

There are similar difficulties with many indicators including the host count that seems so solid. Do we wish to count hosts behind corporate firewalls that carry internal intranet traffic as well as external Internet traffic? Do we wish to count host names that have been assigned, but are not yet associated with operating computers? Is, a server "farm" comprised of roughly 3,000 computers, one site? A Web server (i.e., a computer, connected to the Internet that is running Web server software) at my university hosts Web sites for all faculty members -- is that one Web site or many?

Of course, such problems are not unique to Internet indicators. Gross Domestic Product (GDP) is a widely reported economic indicator, but is it defined the same and measured with equal care in all nations? More fundamentally, what does GDP measure? A high GDP is generally thought to indicate economic development, but a rising GDP might be accompanied by environmental damage, anger over growing disparity in income distribution, disappointment when expectations rise faster than they are fulfilled, displacement of traditional values and customs, crowded cities and so forth. GDP also counts positively many painful transactions like performing bypass surgery, cleaning up hazardous spills, buying a second home after a divorce, or the paycheck of a spouse who is forced into the labor market to make ends meet.[4] The fact that defining and measuring indicators is daunting does not mean we should not do it, but we should keep the limitations -- including the subjective ones -- in mind.

More Comprehensive Frameworks

We have been focusing on specific indicators, but one can also construct indices based on several indicators and other qualitative factors. An index may be more robust than an indicator in trying to measure a qualitative concept. For example, the concept of "development" used by the United Nations Development Program (UNDP) goes beyond GDP. The UNDP wishes to measure human autonomy and breadth of choice, equity, sustainability and empowerment as well as economic productivity. In an attempt to capture this complex concept of development, they compute a comprehensive Human Development Index (HDI) as a function of productivity, health, and education.[5]

Several groups of researchers have constructed frameworks of indices like the HDI in an attempt to characterize the Internet more comprehensively than a group of indicators might. The Harvard Center for International Development,[6] has devised a framework for assessing the Internet readiness of a nation. Their framework includes indices of:

Network Access: What are the availability, cost and quality of information and communications technology (ICT) networks, services and equipment?

Networked Learning: Does the educational system integrate ICTs into its processes to improve learning? Are there technical training programs in the community that can train and prepare an ICT workforce?

Networked Society: To what extent are individuals using information and communication technologies at work and in their personal lives? Are there significant opportunities available for those with ICT skills?

Networked Economy: How are businesses and governments using information and communication technologies to interact with the public and with each other?

Network Policy: To what extent does the policy environment promote or hinder the growth of ICT adoption and use?

My colleagues at the Mosaic Group and I have also defined a framework, which we have used in survey research and 25 national case studies,[7][8] since 1997. We characterize the Internet in a nation in terms of six dimensions. Each dimension has five ordinal values ranging from zero (non-existent) to four (highly developed). Figure 6 illustrates values for Finland over time.

Figure 6. The Finnish Internet, 1997-2000

Source: The Mosaic Group


The first dimension in our framework is pervasiveness, and Table 1 shows the definitions of its levels. Note that the explanations of the values are somewhat subjective and imprecise. For example, we do not try to pin down the exact number of users per capita, which would be impractical, but are satisfied with a rough, order-of-magnitude estimate. As such, we would expect to find consensus among those rating a given nation.

Table 1

Level 0

The Internet does not exist in a viable form in this country. No computers with international IP connections are located within the country. There may be some Internet users in the country; however, they obtain a connection via an international telephone call to a foreign ISP.

Level 1

The ratio of users per capita is on the order of magnitude of less than one in a thousand. There is limited availability, and use of the Internet is embryonic. Only one or a few networks are connected to the international IP network.

Level 2

The ratio of Internet users per capita is on the order of magnitude of at least one in a thousand. The user community has been expanded beyond networking technicians.

Level 3

The ratio of Internet users per capita is on the order of magnitude of at least one in a hundred.

Level 4

The Internet is pervasive. The ratio of Internet users per capita is on the order of magnitude of at least one in ten. Internet access is available as a commodity service.










Nearly all nations have some Internet connectivity today, but access may only be available in large cities. As such, we selected geographic dispersion as our second variable. This dimension measures the concentration of the Internet within a nation, from none or a single city to nationwide availability with points-of-presence or toll free access in all first-tier political subdivisions and common rural access.

While widespread access is desirable, the payoff is in who in a nation uses the Internet. This is accounted for in our sectoral absorption dimension, a measure of the degree of Internet utilization in the education, commercial, health care and public sectors. These sectors are seen as key to development, and were suggested by the measures used by the HDI.

Connectivity infrastructure is our fourth variable. It is based on international and intranational backbone bandwidth, exchange points, and last-mile access methods. A highly rated nation will have high-speed intranational and international backbone connectivity, public and bilateral exchange points, and a high proportion of homes with broadband connections.

Organizational infrastructure is a measure based on state of the ISP industry and market conditions. A highly rated nation would have many ISPs and a high degree of openness and competition in both the ISP and telecommunication industries. It would also have collaborative organizations and arrangements like public exchanges, ISP industry associations, and emergency response teams.

Our final variable is sophistication of use, a measure characterizing usage from conventional to highly sophisticated and driving innovation. A relatively conventional nation would be using the Internet as a straightforward substitute for other communication media like telephone and FAX, whereas in a more advanced nation, applications may result in significant changes in existing processes and practices and may even drive the invention of new technology.


We also consider determinants of Internet success within our framework, as illustrated in Table 2, showing the effects of determining factors in Cuba on the state of the Internet there. We can predict likely Internet success or constraints by considering the levels of variables in several areas.

Table 2. Relationships between Determinants and Dimensions in Cuba, 1997

Determining Factor

Dimensions most directly affected

Poor telephone infrastructure

Pervasiveness, geographic dispersion, connectivity infrastructure, and sophistication of use are all inhibited due to the difficulty in connecting end users and networks.

Difficulty attracting capital

Connectivity infrastructure cannot be improved without capital.

Cultural values stressing health, education and equality

Health and educational sectoral absorption is emphasized as is geographic dispersion outside the capital.

Centralized planning

Organizational infrastructure and pervasiveness are reduced due to monopoly power and planning delays.

Concern for national security given U.S. hostility

Pervasiveness is reduced by access restriction.

Protection of embargoed business activity

Pervasiveness is reduced by content restriction.

Propaganda to and from U.S.

Pervasiveness is reduced by content and access restriction.

Threat of use by subversive organizations

Pervasiveness is reduced by access restriction.

Non-commercial economy

Commercial sectoral absorption is inhibited as resources are shifted elsewhere.

Populist history

In seeking to serve rural areas and small towns, geographic dispersion is increased and connectivity infrastructure extended outside the capital.

Emphasis on human capital

Education sector absorption is increased.


Perhaps the most important determining factor is human capital. To thrive, the Internet needs both trained, demanding users and networking technicians and managers. Users who are aware of the Internet and its applications are particularly important because they drive demand. User education and awareness are stimulated by general news and education as well as policies focused on information technology. Nations as varied as Singapore, Cuba, Costa Rica and Chile have established comprehensive programs for general education on and using information technology. The presence of such programs, along with indicators like school attendance and completion rates, and literacy rate, predicts Internet success. Success in developing networking industry personnel would be predicted by the availability of current university programs in computer science, business and other relevant careers as well as junior colleges and trade schools focused more squarely on skill training and certification.

Complementary infrastructure is also needed, and a comprehensive picture of the state (and future) of the Internet in a nation must include the availability and cost of items like electricity, roads and accessible geography, the telephone system, access to international cables, etc. These factors can easily constrain the Internet in developing nations where capital is scarce and the existing infrastructure limited or unevenly diffused. To attract investment capital from financial markets and international corporations, developing nations will have to adopt policies favoring open markets, competition, free press, intellectual property laws, market regulation, judicial consistency and power, international arbitration, democracy, and so on. In short, they will have to slip into what author Thomas Friedman calls the “golden straightjacket. ”

Note that government policy is involved in developing both human resources and complementary infrastructure. While the information conveyed over the Internet may be eroding government power and sovereignty, successful adoption of the Internet requires government support. Governments must assess the possible risks to their culture, security and autonomy in deciding Internet policy. Some will opt for full-speed ahead; others may choose a more controlled approach.

In either case, it is the responsibility of government to make such decisions explicitly rather than allowing policy to drift. We have examined Internet indicators and multi-index frameworks for assessing the state of the Internet in a nation as well as factors that determine networking success. While Internet measurement is imprecise, it is still of value to policy makers, researchers, business people and others. We need to work toward standard definitions of indicators and indices and gather data globally at regular time intervals using automated procedures, sampling studies, questionnaires and case studies. Such an effort would be costly, but worth the price because the stakes are large.

Scholars like Ernest Wilson of the University of Maryland argue that the Internet has widened the multidimensional gap separating industrial and developing nations, exacerbating an already significant moral and practical problem. A more optimistic hypothesis is that the Internet may lead to improved economic productivity, education, health care, entertainment, awareness of the world, and quality of life in developing nations (even if the "digital divide" increases). The Internet is a decentralized, two-way medium, therefore, conducive to freedom, which, as Nobel-Prize economist Amartya Sen points out, is both a facilitator and constituent of development. Ongoing tracking of Internet diffusion will allow policy makers to plan its growth and scholars to understand its impact.


[1] Heart, F. , McKenzie, A., McQuillian, J. , and Walden, D. , ARPANET Completion Report, Bolt, Beranek and Newman, January 4, 1978. Return to text

[2] See Press, L. , "Tracking the Global Diffusion of the Internet," Communications of the ACM, pp 11-17, Vol. 40, No. 11, November, 1997, Return to text

[3] Return to text

[4] For a discussion of difficulties with GDP, see Cobb, Clifford, Halstead, Ted, Rowe, Jonathan, If the GDP Is Up, Why Is America Down? The Atlantic Monthly, Vol. 276, No. 4, October 1995, pp. 59-78. Return to text

[5] Return to text

[6] Return to text

[7] Return to text

[8] The site contains a more detailed description of the background, framework and surveys. Return to text

Released: November 22, 2000
iMP Magazine:

(c) 2000. Larry Press. All rights reserved.

  Next Focus Story next focus story

Disclaimer: The views and opinions expressed on unofficial pages of California State University, Dominguez Hills faculty, staff or students are strictly those of the page authors. The content of these pages has not been reviewed or approved by California State University, Dominguez Hills.