Rich Man, Poor Man: The Geopolitics of Internet Policy Making
Kenneth Neil CUKIER <firstname.lastname@example.org>
Policy making for the global Internet -- by both the Internet community and offline institutions -- is today a matter of geopolitics. At the same time, despite the Internet's terrific growth, the technological disparity between the developed and developing world grows ever wider. Thus as Internet policy structures are created, there is a risk that the traditional "West versus the rest" dilemma of diplomatic tension based on economic standing will migrate to the newly-formed institutions. This paper critically examines the international institutions likely to play a role in Internet policy making and analyzes to what degree they provide representation for developing economies. It concludes by recommending a policy-making perspective for the "Internet diplomat."
As the Internet matures, its governance structure is also becoming more formalized. This is currently happening on both governmental and Internet community fronts. Traditional offline international organizations, governments, and political unions seek to extend their mandate onto the Internet, either because they believe they have a legitimate role to play in representing their constituency or simply to remain relevant in future policy areas. Examples of this include the issue of electronic commerce, which has been championed by the Clinton Administration, the European Commission, and the Asia-Pacific Economic Council, among others, as well as the US government's stewardship for reforming the Internet domain name system.
The Internet community, for its part, has also contributed to this newly formed area of public policy. Internet institutions act not only for setting technical standards and operational procedures, but also to determine their own fate as an interest group for policy making. The institutions that currently coordinate the Internet, such as the Internet Assigned Numbers Authority, IANA, are undergoing change to reflect a powershift in Internet use by the private sector. Regional Internet Protocol number registries are becoming formalized with new legal structures. Additionally, entirely new bodies are forming, such as regional confederations of national top level domain name registries.
Finally, the private sector is also represented in the institution-building of Internet governance. Their interests are generally represented in user groups, trade associations, ad hoc committees, and Internet service provider forums. There have been calls for their interests to become a formal part of Internet governance in a new corporate structure for IANA in a U.S. Department of Commerce notice of proposed rulemaking.
The relationship between the Internet community and traditional governmental authority, of course, remains tenuous for many reasons. It is partly due to a classical reluctance to accept top-down authority -- be it governmental or corporate -- among Internet engineers. Like landowning farmers of a bygone era, they prefer to work their plot and live by their own rules, suspicious of central power beyond their field, or, in this case, their hard drive. Indeed, data networking engineers themselves designed a network that intentionally incorporated that spirit: The Internet is an open network based on non-proprietary standards and requires few centrally organized functions. Another reason for their reluctance to relinquish complete control to governmental authorities is a well-meaning recognition that many political concerns also have engineering consequences that policy makers do not have adequate technical expertise to address. Finally, in recent experiences, such as the formation of a new IP number registry for the Americas in 1997, called the American Registry for Internet Numbers (ARIN), the US government actually delayed its creation while it learned more about the issue, which damaged the stability of the institution's transfer from the public to the private sector.
Thus in tandem do the Internet community and national and transnational governing institutions develop Internet policy, while the latter increasingly exerts authority over the Internet. In 1998, this has catapulted the Internet -- and the creation of Internet institutions to set further policies -- to the level of international diplomacy. A host of multilateral fora have proven the battleground for negotiations on Internet issues. Slowly emerging from these efforts are different perspectives on how Internet policy should be forged, divided by regional concerns. These, such as in the case of online content regulation, seem to resemble the traditional distinctions that color the policy approaches in classical, non-Internet-related matters. Many countries, such as Germany or Arab nations, have sought to graft their offline laws onto the Internet. (Admittedly, in other, generally "new" areas, such as data protection and privacy, the customary divergences in policy approaches do not exist.)
Moreover, many of the fora where the matters are negotiated come from the offline policy-making world. Their legitimacy to act on Internet questions has not been established. Additionally, these fora in some cases are considered inherently prejudiced as serving to promote the interests of a particular nation or region above others.
These institutional concerns are coupled with technical ones: Today, over ninety percent of Internet hosts are located in the world's richest 29 nations, according to the Organization for Economic Cooperation and Development in Paris . But things are changing quickly. While today over half of all Internet users are based in the United States , President Clinton's senior Internet advisor Ira Magaziner pointed out in 1998 that in five years' time, "80 percent will be outside the United States." 
Yet it is likely that the representatives from the Internet community and government who play a role in setting Internet polices will come from regions where Internet use is most prevalent -- the industrialized world. Since developed economies today have a greater stake in the outcome of the current Internet evolution, it is understandable that this is the force driving events. However, there are few incentives for this community to develop policies today that will afford protection for the interests of other regions of the world and the unique issues they face. Also, there are no guarantees that today's institutions will contain mechanisms for less-developed regions to play a policy role in the future when they represent more Internet users than they do today. Indeed, the reverse is true: Today's Internet statesmen have an incentive to set rules that fall in their favor.
A lack of adequate Internet governance representation from developing economies might have dangerous consequences. It could set back Internet growth in those regions, which would deny those people the benefits of the Internet for political and economic empowerment. It risks artificially cutting out a large future constituency of Net users from policy decisions and further entrenches the historical dependence of the developing world on the industrialized economies. For the newly created institutions themselves, a lack of legitimacy to speak for all Internet users might easily escalate into a potential fracture of the global Internet if other nations or regions decided not to respect those bodies.
That said, the arguments against immediately "internationalizing" Internet governance are compelling. Many developing nations today would be "naked usurpers" if they sought an Internet policy-making role. The world of politics and especially foreign affairs deals with the here-and-now, not tomorrow's utopia. Settling controversial questions among stakeholders in the immediacy has often proved to be a better approach than setting up institutions based on future aspirations. Indeed, the Internet itself has often operated in this regard: Roughly written code aimed at resolving an immediate crisis but actually lasting far longer than ever intended is so common as to be a proverb of systems design.
In analyzing Internet policy-making directions, this paper approaches the topic from an institutional perspective. The presumption is that traditional organizations that serve governmental interests will play an active role in setting Internet policy. This, prima facie, is a notion often anathema to Internet community members. The Internet, according to the popular mythology of Net culture, can "route around" so-called "meatspace" bureaucracy.
Yet more and more evidence suggests the remit of public policy will continue to carve inroads into Net governance, as the Internet takes its place as the global communications infrastructure of the next century. As a result, the institutions that aim to take on this role need to be scrutinized in a manner not done before, to understand how their constituencies and interests cross-apply to the Internet realm. One excellent early work in this area, by William Foster of the Commercial Internet Exchange , argues that Internet governance requires legitimacy and thus the United Nations or its affiliates should play a pivotal role. Foster's examination, however, is limited to domain names, though he intends it to be broader. And his analysis does not broach classic political matters, such as the North-South conflict, which has bedeviled international diplomacy in the latter half of this century.
As traditional politics intersects with Internet self-governance, the challenges are not necessarily new ones. The solutions, however, in some cases must be. By better understanding where potential conflicts exist, perhaps the Internet community and government officials may soothe differences and craft more reasonable and representative policies that create more solid, durable structures for Internet governance.
There are two approaches possible for examining Internet policy making as it relates to international representation. One is thematic and the other institutional. The first approach, dividing the issues by subject, enables a case-by-case comparison of whether international participation is happening or is likely to occur. Yet this approach is problematic, since many different organizations with varying degrees of international representation address similar issues. Secondly, a thematic perspective is limited, because once the issue is resolved, the analysis becomes a postmortem on whether international input occurred.
Instead, an institutional approach proves a better means to examine the likelihood of representation. By nature, organizations are intended to outlast particular immediate issues and make a contribution by evolving over time. Indeed, the ability of institutions to outlive their original mandate and annex other issues in a bid for survival makes them particularly important to study in light of the Internet. It is with this vice or virtue in mind that institutions will be scrutinized.
Today's international organizations, in very general terms, are of two sorts. One category is the institutions designed to address specific matters, with closed memberships. They are inorganic, having been intentionally created with a particular purpose in mind. The second type is the organic institutions that have simply emerged over time with shifting agendas and open memberships .
An example of the former is the North Atlantic Treaty Organization; it was designed to address specific security issues after World War II and in a defined geographic area. An example of the latter is the Group of Seven. It is a "talking forum" -- it has no specific mandate or membership. It was previously the Group of Five until a different national accounting mechanism adopted in the early 1990s placed Italy as the fifth largest world economy, displacing the United Kingdom, and so in a political bow, the group expanded its membership -- as it has more recently with the inclusion of Russia. (The boundaries are not always firm, either. One ironic example of crossover is, again, NATO: One member is Turkey, geographically remote from the North Atlantic. And today NATO's membership is being enlarged for what in the eyes of many critics reflects a dilution of its original purpose of mutual defense in favor of "out of area" conflicts -- signaling a shifting agenda.)
This distinction between organic and inorganic international fora will serve as a framework to compare today's Net policy institutions. However, the analysis does not support one organizational form being "better" than another for providing greater policy-making representation.
To be sure, organic institutions are in the spirit of the Internet itself. The standards-setting body, the Internet Engineering Task Force, for example, prides itself on its open-membership policy; mailing lists contributors and meeting attendees are not "members" but "participants." The group flexibility creates and kills working groups as the situation calls. The obvious drawback is that the IETF is considered by some to be governed by an elite cabal of leaders to the exclusion of others.
The inorganic institutions, however, often better serve the interests of less powerful players. For instance, the European Commission's processes, while heavily influenced by a French and German nucleus, actively seek to support the interests of less powerful members such as Luxembourg and Belgium. One area where this is built into its process is the EC presidency, which rotates every six months among member states (albeit critics can charge it has limited powers, generally in agenda-setting functions). Although less powerful members may be buttressed in such organizations, the institutions themselves are often criticized as overly rigid and bureaucratic -- due to their inherent nature of seeking consensus among parties. This approach, which relies heavily on "process," is distinctly "un-Internet-like."
With this framework in mind, a host of "offline" institutions can be examined regarding the areas in which they intend to play a role in Internet affairs, who they represent, and the effects it may have on international representation.
The ITU, based in Geneva, is the oldest international organization that exists today, founded in 1887. It was designed to house international agreements, which today one might call standards, concerning the telegraph. In the same way as the body, originally called the International Telegraph Union, slowly extended its mission to the telecoms sector, the ITU today has made a concerted effort to play a role in the Internet.
More than any other institution, the ITU attempted to carve for itself a crucial and singular Internet policy-making position. It was one of the original backers of an Internet-community-led initiative to reform the domain name system in late 1996. In May 1997, the ITU served as the repository of the International Ad Hoc Committee's plan on domain names, called the gTLD-MoU. The plan, if fully enacted, would have established a number of new institutions to oversee the Internet's DNS and given entrenched positions of power to representatives of the ITU and other organizations. While the ITU was able to win the support of its membership council, some countries, notably the United States, were wary of the ITU's role and diplomatically indicated their displeasure to the organization. The plan has since gone through numerous revisions and today is in limbo pending the creation of a reformed IANA to treat wider DNS issues like IP numbers and root servers organization. A decision whether to add additional top level domain names, and if so, whether it will be in the framework of the gTLD-MoU, will rest with this new body, being established under temporary US government oversight.
The ITU has a short history for Internet activities. Symbolically, for Telecom95 the ITU quickly assembled an Internet forum and gave a lifetime achievement award to Vint Cerf, an MCI Communications Corp. vice president who cowrote the initial specification for Internet Protocol in the 1970s. In terms of the body's traditional areas, standards and reports, the ITU sends a representative to IETF meetings and in September 1997 issued the report "Challenges to the Network" about the Internet. The report was published alongside a special ITU conference-exhibition called Telecom Interactive. However, very few companies and individuals attended, signifying that the Internet community is wary of -- or at least oblivious to -- the ITU and its functions.
A vestige of the ITU's attempt to exert power over the Internet is its titular role as registry of the ".int" gTLD designated for international institutions, such as UN bodies. However, the current authority lines over DNS are unclear, and with a reform under way, it is quite possible that the ITU neither has such power today nor will be granted such responsibility in the future.
The ITU's membership is not predominantly focused on the Internet -- though this is changing. Politically, the ITU represents the will of its member states, which send representatives from national telecoms ministries, regulators, and diplomatic personnel. Its chief "constituents" are national PTTs and former monopoly telecoms operators. Commercially, the ITU's membership also includes equipment, software, and infrastructure vendors, as well as new carriers and operators, who all pay a fee to participate. Noticeably, there is a significant bifurcation among these groups, between those with a vested interest in maintaining the status quo -- be it circuit-switching or accounting rates -- and those that are free to forge new rules and relationships.
Where the ITU has a healthy record is in upholding the interests of the developing world. As an organization founded to promote communications, it naturally focuses much of its energy and resources today on regions where teledensity is most lacking. As a result of numerous projects in developing regions, it has gained the respect of nations in those areas. Less generously, as a UN-affiliated body, the ITU offers its representatives from less-developed regions a taste of the diplomatic lifestyle, and based in Europe, it serves as a popular and comfortable destination for meetings.
That the ITU has consistently toiled to promote the concerns of the developing world regarding telecommunications matters makes it a particularly pertinent forum to represent Internet development issues, with regional representation in mind. However, its bureaucratic nature and the telephony world's historical antagonism to the Internet make the ITU unsuitable for any concrete governance role.
The Geneva-based WTO is relatively new to the family of international institutions, having been founded in 1995 as the successor body to the General Agreement on Tariffs and Trade, or GATT, which emerged after World War II to coordinate international trade policy among the industrialized powers. The WTO is playing a central role regarding the Internet in matters of electronic commerce and telecoms liberalization.
Regarding electronic commerce, the WTO issued a report in March 1998 specifying areas where international cooperation is required, such as taxation, privacy, and security. The report laid the groundwork for a US drive to declare the Internet a duty-free zone. However, during meetings in March, WTO delegates from developing economies such as Egypt, India, and Pakistan expressed their reservations for an accord. They believe that since the vast majority of Internet businesses are in the West, the policy primarily benefits the industrialized world by letting it avoid paying duties to less-developed nations.
Despite those concerns, the 137 member states of WTO agreed in May to refrain from levying customs duties for one year on goods and services delivered digitally. The agreement was not formally ratified and does not interfere with national taxation. The body also decided to begin work on other electronic commerce issues.
Yet developing countries remain skeptical of the accord. At the WTO May meeting, Cuban leader Fidel Castro complained: "The United States dominates 70 percent of the Internet. So many Africans or other poor people do not even go to school or cannot read. What good does the Internet do them? Where is this world Clinton was talking about?" A delegate from Pakistan stated: "We continue to have doubts regarding the fiscal and revenue implications of the standstill on customs duties." 
A similar schism took place in February 1997, as the WTO's General Agreement on Trade in Services was struck. Of the WTO membership, only 72 governments agreed to open their markets to basic telecoms service competition, which includes creating independent regulatory agencies. However, the WTO notes these countries "account for nearly 93 percent of the total domestic and international revenue of $600 billion generated in this sector annually." 
As an institution, the WTO is untested. Its dispute resolution process, for example, has not treated enough cases or handed down sanctions to reveal insights into its tendencies. Ostensibly, it is a neutral forum for trade issues. However, there is a danger that the WTO, like the GATT of an earlier era, can become a tool of the industrialized world to extract important trade concessions from the developing world. In terms of the Internet, the WTO is an obvious place to address uniquely offline issues, such as taxation and market access. But as a paragon of a global trading order that implicitly serves the interests of its most wealthy members, it fails to provide any assurances for international representation on Internet policy issues.
Indeed, the only means by which the developing world can protect its interests at the WTO is by exploiting its multilateral nature. Developing nations can bring before the body contentious issues where they are weakly positioned bilaterally in the hopes of winning the support of others or using the WTO's neutral arbitration mechanism. This view, admittedly, does not bode well for the future of the WTO as an Internet policy-making forum. The debate over Internet duties may have set the tone for other Internet issues that go before the WTO.
The G7 is an ad hoc association of the world's seven most powerful nations, measured in economic terms. It does not have a secretariat, staff, headquarters, or budget. Nor does it have a specific role or designated powers. G7 meetings and locations are decided informally; heads of state meet once a year, and special theme summits among members are called for on a case-by-case basis. One recent head of state summit, in Lyon, France in 1997, addressed unemployment. A specially-called summit in June 1996 in Paris treated international terrorism. Since 1996, the G7 has put itself forth as the "Political Eight," or P8, to include Russia, and the May 1998 summit in Birmingham, UK used the formal title Group of Eight. Other members are from North America, Western Europe, and Japan.
The G7 does not provide any means for direct representation from outside nations, due to its nature as a closed association. Despite this, however, in terms of the Internet the G7 has addressed information society issues generally and the concerns of the developing world specifically. The G7 hosted a 1995 meeting in Brussels, Belgium on global information infrastructure. At the time, South African Deputy President Thabo Mbeki, the only representative of Africa invited to attend the meeting, called on the leaders to not exclude developing nations in their information society plans. In that regard, he proposed that South Africa host a G7 meeting where the issue, as it pertains to the developing world, could be discussed.
A year later, the G7 sponsored the Information Society and Development (ISAD) Conference in Midrand, South Africa in May 1996. The body vowed to help mobilize investment, promote access to technical training, and further universal service goals. In the closing statement, the group recognized "the necessity of worldwide cooperation with particular attention to less developed countries" regarding the information society. 
Since the conference, the G7 has not specifically addressed Internet issues as they relate to developing economies. However, it has done much work on matters that are indirectly related to the Internet among themselves, and the jointly forged initiatives may serve as a template for policy coordination with other nations. At the Terrorism Summit in June 1996 it singled out the Internet and especially cryptography as a vehicle by which terrorists are able to communicate to plot their acts. The ministers called for controls placed on cryptography for law enforcement purposes.
In December 1997, at a day-long meeting of law enforcement officials of the G7 and Russia at FBI headquarters outside Washington, DC, the group agreed on a 10-point plan to coordinate national policies to combat international high-tech and computer-related crime. Member countries said they would ensure that cross-border legal mechanisms are in place to digitally track and prosecute suspects and would work to eliminate safe-havens for computer criminals. In May 1998, heads of state of the G8 were briefed on the issue of computer hacking and the need for coordinated international action.
The group clearly intends to tackle issues that elide with Internet governance, as the Internet becomes a major driver of economies. For G7/G8 members, the forum provides a valuable closed-door, high-level setting to discuss Internet-related policy issues without distracting input from nations that are not yet technologically developed, as the example of cryptography demonstrates. Yet what the group works on among itself will almost surely be exported to larger, multilateral organizations in a bid for global support, since many of these issues can only be adequately treated with worldwide cooperation.
As an elite club of rich Western nations and Japan, the G7/G8 is especially inappropriate as a forum for regional representation in Internet policy making. In fact, developing nations would be well to remain suspicious of the body's deliberations over policies that affect the Internet, since it will certainly be designed to serve the specific interests of industrialized nations. Developing regions should not treat all policy initiatives emanating from the G7/G8 as diabolic, but they should heavily scrutinize them in terms of their own strategic interests.
The Paris-based OECD is an economic think tank funded by its 29 member states, which include most of Europe, North America, and some of the Asian economies. It also has direct links with the international business community via a specially organized committee of representatives from the private sector. The OECD harmonizes national statistics so that figures can be considered internationally and also recommends policies for its members. Its work is generally used as the basis for policy making in other fora. The OECD has experience treating telecommunication matters specifically, with bi-annual reports on national and international infrastructure and regulatory issues. In the past, it has addressed issues that today are directly related to the Internet, such as privacy, data protection, and cryptography.
Since 1996, the OECD has turned its attention specifically to the Internet in a number of wide-ranging areas, from communications infrastructure to electronic commerce. It served as a forum in 1997 to forge industry self-regulation for consumer protection in online commerce. Its March 1997 cryptography guidelines recommend a private sector-led approach to balance the needs of law enforcement with privacy and security. In August 1997 it cosponsored a conference on cryptography in Australia with members of the Asia Pacific Economic Council to educate non-OECD members on the issue. It is currently revisiting its privacy guidelines, drafted over a decade ago, to keep them updated with current technology matters. In the area of content regulation, for "illegal" and "harmful" material, the OECD is drafting a report that disdains calls by France and other nations for a "code of conduct" for the Internet; instead, the OECD favors industry-wide self-regulation such as content-filtering software for users.
The OECD is also putting considerable effort into investigating the economic consequences of taxation on the Internet. This is expected, since the only area where the OECD has any real power is the area of taxation -- the body's international taxation agreements are generally accorded the status of treaties. The OECD also called for globally uniform policies regarding the convergence of media, telecoms, information technology, and entertainment, in a fashion similar to the European Commission's 1997 convergence green paper. However, member nations have indicated they prefer that the OECD treat e-commerce issues piecemeal instead of seeking to build an overarching international framework that tries to encompass all aspects of the online world.
Finally, the OECD has specifically addressed Internet governance. It issued a report on domain name allocation policies in 1997 that explained to international policy makers how the Internet is governed and urged that private, competitive registries be instituted. It treated the issue of Internet traffic flows in a 1998 study on Webcasting. Finally, the body noted the complaints of Asia and Europe that the Internet is US-centric in a May 1998 report on Internet infrastructure and traffic exchange.
The OECD could play an important role in Internet governance by approaching controversial issues from a neutral perspective -- one that seeks, above all, statistical evidence to back up its recommendations. Additionally, its recommendations themselves are nonbinding, and countries are able to manipulate the process as it moves forward, making the OECD ideal in that it is relatively impotent and nations are free to defer to the body when it is useful.
However, while the OECD has made a strong effort to host dialogues with nonmembers generally, it is still regarded as the "rich man's club." True international representation would require the OECD to change its remit and open its process to other economic constituencies. This would destroy the OECD's specific raison d'être as a forum for industrialized nations and would be unacceptable. It would also be unnecessary: As the OECD notes, today the vast majority of Internet users come from member countries, and so the body is appropriate to serve as the first forum for these issues.
For developing economies, it is in their interests that the OECD work on Internet policy issues, even though they are not part of the formal process. The OECD is good at balancing the hegemony of any single member or group to achieve a broad consensus among all its members, which in cases other than North-versus-South may prove beneficial in terms of broad, international influence. Additionally, since the OECD's powers are extremely limited, developing regions do not need to fear that if their interests are ignored there will be no recourse. Finally, the OECD secretariat is made up of economists rather than policy makers, and so the emphasis is on objective study and not politics.
However, the inherent bias of the OECD is in favor of its members. For this reason, it is also in the interests of the developing world that the OECD not be used to craft broad Internet policy. Where it most likely will draft a specific policy, in the area of an international tax regime to treat the sale and delivery of digital goods among OECD members, developing nations will probably have little to no voice in the discussions. However, in this one area, developing nations should defer to the OECD (as the West likely will). This is because the issue of taxation is by nature highly charged politically, yet the OECD's representatives often come from the long-standing bureaucracy rather than fashionable political administrations. This gives the OECD an advantage over other multilateral organizations to forge a better, more stable policy than would be made if politicians handled the issues in a treaty negotiation-like setting, with the associated theatrics and suboptimal compromises.
Two organizations affiliated with the UN stand out for seeking a role in Internet policy making in the narrow areas where they have special competence, and they should be briefly noted in terms of international representation.
WIPO, like the ITU, is affiliated with the United Nations and based in Geneva. It has embarked on a process to re-examine its long-standing trademark and intellectual property rights law in terms of the Internet and digital economy. In December 1996, it held a month-long conference that came close to reversing traditional law regarding information use and reproduction. The original proposals, heavily influenced by the US position, strongly favored media owners and trademark holders to the disadvantage of ordinary people who access information. For example, it is unclear whether today's "fair use" doctrine would have been maintained. Regarding the Internet specifically, the original accord forbade many forms of storage and reproduction of information and was technically unsound, since it ignored current caching technology common among ISP networks. The wording, later modified, threatened to place a debilitating burden on today's Internet industry. However, WIPO's most significant Internet-related initiative involves its work in fostering a policy to treat domain names internationally. In a bid to protect trademark holders, it began work in 1996 to draft a framework for IP rights and DNS. A WIPO representative participated in the IAHC group to reform the domain name system. Separately, it held a series of meetings throughout 1997 to formulate a dispute resolution procedure for contested domain names. Critics charged that WIPO came close to overturning traditional copyright and trademark law by assuming infringement in cases where two parties claim the same name, rather than requiring infringement be proven. The policy, not yet fully developed, appeared to place the interests of corporate domain name holders above the interests of individual users. While developing countries send representatives to WIPO, its constituency is mainly current intellectual property holders, the majority of which are from the developed economies. The developing world should be vigilant that WIPO's polices do not hinder access to information or place untenable burdens regarding domain names for their regions.
At a February 1997 meeting of the UNCITRAL electronic commerce working group in New York, the body drafted a report on digital signatures. It specified areas that require uniform policies for digital signatures to be legally valid for cross-border transactions. Some lawyers and policy makers oppose UNCITRAL's recommendations, contending that the regulations are technology-specific because they back only today's known encryption technology and are not flexible for new forms of authentication, such as biometrics. However, the issue of representation of developing economies in this area is essentially moot. Developing countries, because they neither manufacture, sell, or constitute a significant user population of encryption techniques like digital signatures, currently have no inherent interest in the outcome. While this will surely change over time as developing nations go online, the policy making is best handled by nations that bring empirical experience to the debate, which excludes the developing world. Although legal recognition for digital certificates hinges on cross-certification, and standards for authenticating identity in the developing world differ from the West, the developing world must accept it is a follower rather than a participant in the discussions, and developing nations will likely need to modify or construct their national laws based on the outcome. Finally, it is ultimately unclear whether UNCITRAL's work will become a dominant force in creating international policy, since industry-backed solutions regarding digital signatures, such as by the Paris-based International Chamber of Commerce, seem to be taking a lead in the matter.
As the network of networks evolved, so sprung up institutions to govern its operations. While this paper is primarily concerned with international representation in offline institutions, a consideration of Internet self-governing institutions is pertinent since they will probably continue to set policies in their limited, technical domains.
First among them is IANA. It is currently in the process of a sweeping reform to formalize its functions and make the body more representative to user interests. The US government, which is spearheading the reform due to its historical role of funding the development of the Internet, also seeks to make the body more international.
The current IANA's powers are varied and vast, centered on a single individual, Jon Postel, a research director at the Information Sciences Institute at the University of Southern California based in Marina del Rey. IANA oversees three broad -- but critical -- aspects of the Internet's infrastructure. It coordinates the allocation of Internet Protocol address space, a finite resource that denotes Internet hosts. It sets technical parameters, such as port numbers, for protocol specifications. IANA maintains the zone file of all top level domain names, including choosing who may be a registry and whether new TLDs are created. (Beyond the handful of US-based gTLDs that were created at the network's infancy, IANA deferred to the ITU's two-letter country code as the basis of national TLD names.) Finally, IANA has some degree of authority over who operates and the location of the 13 root name servers.
Dr. Postel, in a personal capacity, also acts as editor of the IETF's standards documents, called Request for Comments. While exercising a seemingly perfunctory role, the RFC editor actually wields tremendous power. The editor serves as a point of checks and balances to see that no new Internet specifications are created until they are scrutinized by a final judge representing the interests of the original architects of the Internet, of which he is one. IANA's powers evolved over time, and the body, partially funded by the US contract, receives its authority based on the consent of protocol developers and ISPs, in a model of trust and mutual benefit.
IANA has delegated responsibility to regional bodies for the day-to-day IP address allocations. The American Registry for Internet Numbers, or ARIN, based in Virginia, handles allocation for the Americas, Caribbean, and parts of Africa; the Reseaux IP Europeenes, RIPE, based in Amsterdam, covers Europe and parts of the Middle East and Africa; and the Asia Pacific Network Information Center, or APNIC, currently based in Tokyo pending a move to Brisbane, Australia, covers that region. The regional registries also partially fund IANA as of May 1997, based on the revenue they receive from ISPs who requested IP address space. In this limited regard, Internet policy making can be said to be international and bottom-up since IANA's authority can be traced down to specific geographic regions and Internet users there. 
It is envisioned that as other regions of the world develop the Internet, specific registries will be formed there also. New registries are being planned for Africa, Latin America, as well as possibly Russia and the Commonwealth of Independent States.
Formal international representation in IANA is lacking. This is comprehensible -- though perhaps not justified -- since the Internet was created in the US and the majority of users are based there. As an institution that evolved organically, and with its legacy in the academic and US military communities, IANA has been slow to incorporate the interests of new users from the commercial sector and from around the world. The new corporate structure of IANA will likely redress this situation by formally cementing the regional representation of the IP number registries in its policy-making powers.
The majority of domain names today are registered under gTLDs, which are managed by a private firm, Network Solutions Inc., in Herndon, Virginia. In a bid to boost the interests and policy participation of TLD registries outside the US, European registries created in March 1998 a federation called RIPE CENTR, the Coordination of European National TLD Registries. The European Commission sends an official to the meetings in an "observer" status. Like IP address registries, it is also envisioned among Internet leaders that nation TLD federations will form in other regions as well.
Regarding standards, the IETF's participants are worldwide, and nearly half of those who attend the thrice-yearly meetings come from outside the US. The power structure of the governing bodies of the IETF, the Internet Engineering Steering Group and the Internet Architecture Board, is complex, but basically founded on technical prowess and ability to manage the work of others. Around one-third of the members on the IESG or IAB come from outside the US.
Meanwhile, standards for the Internet's most popular application, the World Wide Web, are the purview of the World Wide Web Consortium, or W3C, which has two headquarters: At the Massachusetts Institute of Technology in Cambridge and at the French computer and networking laboratory INRIA in Grenoble. One reason for the concrete inclusion of international representation in the W3C is that the Web, unlike most Internet applications, was not created in the US but by a Britishman based in Switzerland and therefore was inherently sensitive to the need for global representation. The W3C also has affiliates in other nations, such as Japan. It is made up of high-tech firms, mostly based in the US, which reflects the geographic distribution of the information technology sector. Yet non-US companies are members. Because developing countries do not have strong technology sectors, they are not represented at the W3C. However, the W3C has rarely strayed from engineering questions into the public policy realm (it has done so most notably in the area of privacy protection), so this has not set back the developing world regarding Internet policy.
Representing the user community is the Internet Society, or ISOC. It was created in the early 1990s originally to serve as a legal umbrella for both the IANA and the IETF. For numerous reasons, this did not fully happen. By 1996, when the question of new domain names threatened to fracture the Internet due to renegade registries that began selling self-minted top level domains only routable on a few networks, ISOC assembled the IAHC to reform the DNS system. It proposed a new Internet governing structure to oversee the DNS, and new gTLDs. Today the IAHC proposal, known as the gTLD-MoU, remains in limbo at least until the new IANA structure is in place. It is unsure whether it will ever be adopted.
Although ISOC is the most widely internationally representative Internet organization other than the IETF, with around 6,000 members from chapters in 25 different countries, it is an irony that ISOC's IAHC itself had few members from outside the US: One Japanese, an Australian, an American living in Israel, two Americans living in Geneva, and six Americans based in the US comprised its 11 members. The gTLD-MoU received backing from companies around the world, including France Telecom, Deutsche Telekom, and others. However, very few companies based in developing countries signed on to the process, or it seems, were aware of it.
Finally, a number of Internet interest groups have formed around the world, representing the voice of specific regions. The first ISP trade association was the Commercial Internet Exchange based in Herndon, Virginia, formed in 1992. It treats international issues, yet remains the premier voice of the ISP community among policy makers in the US. In 1997, the Asia and Pacific Internet Association was formed in Singapore, and the EuroISPA, the European Internet Service Providers Association was formed in Brussels. Both groups have undertaken to lobby national and International officials to support Internet-friendly policies; initial work has been on DNS regional traffic-exchange costs and content regulation. These groups represent the interests of the Internet sector in their regions and do not intend to play a formal policy role themselves.
Few of today's institutions that seek to play an Internet governance role provide a means for input from developing regions. This is an expectable outcome, since the Internet is mainly based in the West and so it is Western-dominated organizations that have been first to seize the initiative. Moreover, this factor has further divided the Internet into two branches, like the traditional offline world, of industrial and developing regions. The danger is therefore very real that this will be a source of conflict among regions in the future.
In terms of organic and inorganic international organizations, neither type of forum on its own guarantees the developing world influence in Internet governance, nor do traditional offline bodies as opposed to groups that evolved from the Internet.
The inorganic institutional structure where all members are equal and locked in, with a specific remit, is unlikely to be effective for the fast-paced Internet. Many of these types of groups come from the traditional offline world. They can serve developing interests as a multilateral forum that lays out certain procedural assurances for representation but in itself does not ensure representation.
At the same time, organic organizations that are constructed on much looser terms risk becoming dominated by their most powerful members, since developing economies have no built-in protections to preserve their status as policy actors. These types of institutions better fit the needs of the Internet due to their flexibility and more closely resemble today's Internet institutions. Yet they might actually prove worse for the developing world since there are no concrete assurances that regional representation will happen.
The best chance for regional Internet representation exists in a combination of both the ad hoc and the formal institutional approaches. The IANA itself, for example, offers zero representation to the developing world, while the IETF's representation can wax and wane depending on circumstance. However, the regional IP registries, if granted earmarked representation on the new IANA board that is forming, will provide stable and consistent international representation, albeit not from the developing world. Yet this can be addressed by rotating representatives from the regional registries -- including those still to be formed -- to serve on a board of trustees for IANA, for instance. Additionally, a committee comprised of IP registries from all geographic regions could be formed to ensure that international representation happens. This might be implemented as a council of registries that reports to IANA's board of trustees or serves as ex officio members of the board. While its role might be predominantly symbolic, it would ensure that the voice of the developing world has a formal outlet.
The fundamental problem with this idea, however, is one that plagues the broader issue of Internet institution-building: Current bodies that serve the interests of a constituency in one area do not necessarily have the competence to do so in another. This applies both to government officials working on DNS, as well as -- in the case of the above example -- IP number registries involved in politics. The offline institution threatens to jeopardize the smooth functioning of the Internet. The Internet body risks being distracted and inefficient at handling its intrinsic purpose.
So what should policy makers do to ensure that Internet governance is based on international representation? Part of the answer lies in why policy making should include representation from the developing world. The overriding reason is one of fairness. The Internet's ability as an open system to empower users would be wasted on the very people it can best serve unless they are a part of the Internet governing process, which might actually help them speed up the deployment of the Internet in their regions. But another reason is that unless developing regions are granted representation now, they will surely seek it later. They might do so by forcing issues onto the agenda of international bodies in which they believe they will have greater influence yet which are ill-equipped to treat those matters. Such a development would set back the resolution of the original concern as well as confound other international organizations and their respective roles. Nations and regions who feel their interests are under-represented might one day "fracture" the Net by refusing to recognize the legitimacy of Internet governing institutions or particular policies. The Internet, as a weave of inter-relationships based on mutual benefit, needs the support of all its members to remain effective, and this fragile allegiance is most stable when all users have a voice in its operations and management.
To promote a healthy and internationally viable Internet and prevent possible tensions, a new way of thinking is required among policy makers currently constructing Internet policy and forging the new institutions. This may require a new conception of what constitutes national self-interests, or, at the least, a new way of thinking about international diplomacy.
There are no statesmen of the Internet, nor should there be. Internet policy must transcend national borders in a way similar to the way the network's topology itself ignores geography. That is not to say that altruism is the sole criterion of policy; that would be a foolish and unrealistic goal. Instead, it means that Internet policy will only be viable and lasting in areas where national governments and regional political unions can identify common objectives with the broader international community. The successful Internet diplomat will have an acute understanding of where the interests of singular nations and the larger user community elide -- indeed, these interests must be discovered, and that may require nations to forego a certain amount of traditional power for the larger good of the Internet community.
This harkens back to the historical model of Internet self-governance. Private networks did not always agree with IANA decisions, just as some protocol developers were occasionally set back by IETF standards, in the larger drive for rough consensus. However, all members of the Internet community continue to defer to these organizations because it is in their broader interest.
It will not be easy to achieve such selflessness, and nations, like individuals, are loath to relinquish control. One means to facilitate this is to insulate Internet policy making from politics. As noted with the OECD, civil servants are at times better equipped to serve broad national and international long-term interests than politicians who, at least in many countries, rely on the temperament of voters for their legitimacy to rule and at times exploit the baser elements of human nature to win elections. On the other hand, these civil servants, Internet diplomats, will still be accountable to their political rulers -- yet the actual policy-making institutions should have a degree of separation from crass politics.
A few principles and potential obstacles stand out for Internet policy making:
1. OECD. Statistics based on Internet host counts by Network Wizards.
2. International Data Corp.'s September 1997 statistics show Web users (to broadly mean Internet users) at 29.2 million in the US, 9.9 million in Western Europe, 8.4 million in Asia-Pacific, and 2.7 million in the rest of the world. Importantly, IDG projects the US will still account for over half of all users by 2001 (US: 94.2 million; WE: 32.1; AP: 36.8; rest of world: 11.4 million).
3. Ira Magaziner. Public remarks made 27 January 1998 at Re-Engineering the Internet conference in London, England. How he arrives at the figure is uncertain.
5. I am indebted to William Pfaff, foreign affairs writer and columnist at the International Herald Tribune, for first distinguishing some of these categories in 1994, which he uses to argue that international policy is built as much from informal liaisons among leaders as from intentional, institutional design. Cross-applying that to the question of regional representation within institutions is my own analysis.
Kenneth Cukier, "Funding plan bolsters key Net body," Communications Week International, 2 June 1997. London, UK: EMAP Plc.
William A. Foster, "Registering the Domain Name System: An Exercise in Global Decision-Making." in Brian Kahin and James H. Keller (eds.), Coordinating the Internet. Cambridge, Mass.: MIT Press. 1997.
Information Society and Development (ISAD) Conference, "Chair's Conclusions." Midrand, South Africa. 13-15 May 1996.
International Data Corp. Internet Commerce Market Model Report. Farmingham, Mass. July 1997.
Organization for Economic Cooperation and Development (OECD). Information Infrastructure Convergence and Pricing: The Internet. Paris, 1997.
Pierre Musso, Télécommunications et Philosophie des Réseaux, Presse Universitaire de France, Paris. 1997.
Wolfgang H. Reinicke, "Global Public Policy." Foreign Affairs, Nov./Dec. Council on Foreign Relations. New York. 1997.
Reuters. "WTO: No Net taxes for a year." May 20, 1998.
World Trade Organization. "Study from WTO Secretariat Highlights Potential Trade Gains From Electronic Commerce." Press Communique, Geneva. 13 March 1998.
Since this paper is based on priceless conversations with Internet leaders, policy makers, and members of the Internet community, the author, a reporter, deeply thanks his sources, his friends.
Kenneth Neil Cukier is a senior editor and Paris correspondent for CommunicationsWeek International, covering the architecture, economics, and public policy of the Internet. Prior to joining CWI in November 1996, he edited the Web site Sarajevo Online. From 1992 to 1996 he worked at the International Herald Tribune.
Kenneth Neil Cukier (email@example.com)