There are several conclusions that can be drawn from the foregoing five country case studies and the cross-country analysis that have implications for Korea. First, as is the case for each of the other countries in the study, Korea is unique and must develop its NII with regard to its own circumstances and values. Although many features of Korea’s NII might be similar to the NII plans and experiences of other countries, they will almost always be different in their actual implementation and details in Korea. This is as it should be.
Second, it behooves Korea to continue to watch NII developments in other countries, but especially in the United States, Japan and Europe. These are the three largest markets in the world, and what they do will shape the future developments NII developments elsewhere. While Asia might become a unified market sometime in the future, such a prospect is longer term rather than intermediate and probably cannot be a serious consideration in NII plans at this time.
Third, the other countries in this study tend to see the United States as the leader in creating or innovating with technology, but they see Japan having the manufacturing capability to compete strongly once the direction of new technology becomes clear. Europe is seen less as a leader in technology than a possible leader in standard-setting. Thus, different developments bear watching in the different leading markets.
Fourth, Korea can benefit from looking more closely at Singapore’s strategy and plans. Singapore’s NII, like Korea’s KII, is influenced by external forces and hopes to have influence beyond its borders, within the Asian region specifically. How feasible is this goal for Korea? Does Korea now play roles within the region that the KII can reinforce? Or, is Korea trying to use KII to become a force in the region? The latter aim might be too ambitious if Korea does not already play a significant role.
Fifth, Korea probably needs to further clarify for itself the primary goals and strategy of its KII initiative, and to ensure that government and industry are moving in the same direction so that their efforts are synergistic and mutually reinforcing rather than in conflict and mutually defeating. The Korea case study suggests that industry is moving in close relationship with government, but it does not indicate whether industry and government are moving in the same direction. Now might be a good time for Korean government and industry leaders to consider this question because the KII initiative is very young and the country is about to embark on the next Five Year Plan for the NBIS Project.
Finally, as indicated in the cross-country comparison, it is going to take longer than expected for NII visions to be realized and the future reality will probably look different than anyone imagines today. This is indicated by the fact that the initial euphoria and rosy forecasts for NII are giving way to a host of realities which have the net effect of slowing down implementation. Korea needs to closely look at this issue when it updates the KII Plan.
1 The concept of grand projet refers to the development model of French high-tech industries (space industry, telecommunications, aeronautics, nuclear industry). It designates a type of government-industry relations through a complex of actors - public administrations, public research centers, public operators, private manufacturers - linked together by common values and goal (see Brenac 1991, Cohen 1992). French elites - les grands corps techniques - play an instrumental role in this linkage by providing communication channels and sharing the same technical culture. (see Vedel, 1984).
For reasons we will explain below, France Télécom did not lobby in favor of IS.
3 "I cannot stand the idea that, someday, I will have to read the catalogue of Le Louvre museum in English on an American data base", said a famous philosopher on a TV show. Anecdotally, this cultural resistance to Information Highways is also reflected by the attempt - so far in vain - of Jacques Toubon, the French minister of Culture from April 1993 to May 1995, to rename IS in infoducs - in reference to the Latin heritage of France (and probably in opposition to the industrial culture that IS information would suggest!).
4 In particular, the question raised by the Bangemann report "whether the information society will be a strategic creation for the whole Union, or a more fragmented almagam of individual initiatives by Member States" (p.5) was not addressed.
Closing statement at the Conference on IS organized by the Ministry of Industry, Paris, December 1994.
On the notion of interpretative flexibility, see: Pinch and Bijker (1984), Bijker and Law (1992).
7 Loi du 29 décembre 1990 relative à la réglementation des télécommunications. depending on their nature, services are either fully liberalized, or submitted to a licensing process.
8 In May 1995, 60% of the work force went on a one-day strike to oppose the planned privatization. The November and December 1995 social movement also showed a strong opposition of France Télécom's employees to privatization.
9 Lorente (1995) argues that, in an unexpected way, the Clinton-Gore 's Agenda for action puts a much more emphasis on the role of state than the Bangemann report.
10 France Télécom experts strongly disagree with Théry's figures and provide costs estimates considerably higher: up to 30 000 FF to equip each household during the first years with fiber to the home..
Colloque "Les autoroutes et services de l'information". Paris, 7 décembre 1994.
12 We use the future since this debate has not taken place so far (if one excepts a one-day conference devoted to the presentation of the government policy.
13 Eloquent evidence of such a flexibility appears in that applicants to trials were asked to indicate in their proposals which regulatory, technical or financial provisions they would need from the part of public authorities.
14 José Rossi,, Minister of Industry, Post and Telecommunications, and Trade, at the Colloque "Les autoroutes et services de l'information". Paris, 7 décembre 1994.
On the French cable plan, see Vedel and Dutton (1990).
16 Not including the 14,9 billion FF "contribution" that France Télécom has to make to the state budget.
17 In the prospect of a possible privatization, such a management is indeed indispensable, given France Télécom's debts. In September 1995, France Télécom 's chairman Marcel Roulet was replaced by Michel Bon. Interestingly enough, and for the first time in the organization, the head of the public operator is not a telecommunications engineer, but comes from the food distribution sector.
Personal interviews, July 1994 and November 1995. See also Busson, 1995.
Cable operators have now turned to coaxial or hybrid systems.
20 With the notable exception of Canal Plus, a pay-TV station. Canal Plus's annual revenues amount to 10 billion FF and its net profits to 1 billion FF. Canal Plus is also an integrated group with subsidiaries in decoders manufacturing, TV and film production, satellite operation. Canal Plus a developed its pay-TV activities in Belgium, Germany, Spain, Poland and different African countries.
Bertelsmann and Canal+ are already partners in Premiere, a German pay-TV channel.
22 In 1984, foreign VCRs had to be cleared through customs in Poitiers, a medium city 400 km from Paris without international airport. This lengthened for a while the penetration of foreign equipment.
Loi du 3O septembre 1986 relative à la liberté de communication.
24 On neo-corporatism in general and in France, see respectively: Schmitter and Lehmbruch, 1979; Wilson, 1987.
25 Loi du 2 juillet 1990 relative à l'organisation des Postes et télécommunications; loi du 29 décembre 1990 relative à la réglementation des infrastructures et services de télécommunications.
26 The title and responsibilities of this Ministry may change depending on cabinets. It is only since 1988 that the Ministry of Industry has supervised telecommunications and postal activities. However, most of the time, the Minister of Industry is associated an adjunct Minister for Posts and Telecommunications (known as Ministre délégué or Secrétaire d'Etat). It should also be noted that, from May 1995 to September 1995, existed a Minister of Information Technologies and Posts (who had in fact the same responsibilities than the former and current Minister of Posts and Telecommunications).
Le Monde, 10 janvier 1996, p. 20.
28 Members include: Association of Commercial Television in Europe (which groups private TV stations), BT, Canal+, Compaq, the European Publishers Council, France Télécom, ICL, Mercury, nCube, Oracle, Philips, Silicon Graphics, Sony and Thomson
29 The ERT, which is chaired by Carlo De Benedetti, the head of Olivetti and a member of the Bangemann group, stressed that the current liberalization process within the Union is too slow. Unice's report, "Making Europe more competitive - Towards world class performance " published in June 1994, called for an overhaul of Europe's public sector, including the privatisation of telecoms, postal, energy, transport and social services.
30 This was the case of the DVB: four of the largest TV companies, TF1, the CLT, the BBC and ITV, did not endorse the agreement for a common encryption standard.
31 This section is based on author’s analysis of France Télécom data published in The Lettre de Télétel et Audiotel.
Survey commissioned by GESTE (a service providers association).
33 Unfortunately, only a few services are available through gateways open to PCs users. And the other high-speed gateways require the purchase of a specific modem.
In Germany, a license used to be required for modems.
35 The following aspects are being examined: access under fair conditions by other on-line services to the publications controlled by Europe Online's founders; possibility for other groups to offer content under conditions similar to those enjoyed by Europe Online's founders; existence of anti-competitive agreements with third companies.
37 This Christmas toy effect probably plays for the Internet. However, it does not appear globally as long as additional users get on the network.
38 Beaussant Antoine, "Les éditeurs télématiques français sont prêts pour l'après-Minitel", Le Monde, 2 novembre 1994, p.13. Beaussant, the head of Geste, an information providers association, also criticizes the hype about IS and the Internet. "The Minitel has been providing for years a variety of services which are just being discovered on Internet (...). We, in France, have known the cyberpizza for 10 years", he ironically writes.
Le Monde, 10 janvier 1996, p. 20.
40 LASFARGUE Yves, representative of the CFDT, a French trade union, at the Conference Les autoroutes et services de l'information, organized by the Ministry of industry, Paris, 7 décembre 1994.
41 In English, the term “Information Superhighway” has been most popular, as used by Gore (1991), but often shortened “Info Highway;” its equivalent in Japanese (joho haiue) has also been used. The term “National Information Infrastructure” seems to have first been used in Singapore, as in (NCB, 1992); the Japanese equivalent is joho infura, which is now preferred. Other variants on the NII theme include “Asian Information Infrastructure,” “Global Information Infrastructure” (GII), etc.
Ito (1991), referring to Umesao (1963), Hayashi (1969).
43 This is consistent with Calder’s conception of Japan’s postwar economic policy as that of a “Reactive State”: see (Calder, 1988). For a more detailed discussion of catch up rhetoric in Japan’s NII plans, see West (forthcoming).
44 It should be noted that construction bonds do not count against the national government’s requirement to balance the budget, i.e., they can be used to pay for deficit spending (Lincoln, 1988, p. 74), thus overcoming one potential objection from the powerful Ministry of Finance.
45 It is well beyond the scope of this paper to summarize Japan’s postwar industrial policy and the central role played by government ministries, even if limited to just high technology industries. The standard discussion of the role of MITI is given by Johnson (1982), while the interpretation of Okimoto (1989) emphasizes the role of private firm; Johnson, et al., (1989) offers views of Japan’s developmental policies in several industries. The best account of the incubation of Japan’s mainframe computer industry is given by Anchordoguy (1989), while it, along with Fransman (1990) and Flamm (1987, pp. 125-153), outline Japan’s computer industrial policy. Other relevant discussions would include Japanese incubation of the semi-conductor industry (Anchordoguy 1989, pp. 138-147; Mason, 1992, pp. 174-187; and Okimoto, et al., 1984, pp. 95-115).
46 This includes DOS-Windows (Microsoft), Macintosh (Apple), UNIX (various versions from IBM, Hewlett-Packard, Sun and DEC), Netware (Novell) and IBM's mainframe operating system.
47 For more detailed analysis of the troubles of the Japanese software industry, see Cottrell (1993), Baba, et al. (1995), Nakahara (1993), and Dedrick and Kraemer (1996).
48 The report was printed in both Japanese and English, as well as a widely-distributed ten page summary. The rapid availability of the English summary contributed to its heavy use outside Japan, as did its publication (in both languages) on the World Wide Web once MPT established a web site later in 1994.
49 The net effect of such right-of-way policies would assist the development of rivals to NTT’s network, since NTT already has a national right-of-way network.
For a generalized typology of bureaucratic rivalry in the Japanese government, see Campbell (1984).
51 So, for example, the MPT vision makes a priority of wiring “teletopia cities” by 2000, but no mention is made of MITI's “new media communities”. (Telecommunications Council, 1994a, p. 49).
52 Unlike Teleway and Japan Telecom, DDI lacked a development partner to provide a ready-made right-of-way for fiber optic cables, so DDI is more dependent than NTT and the other NCC’s on microwave relay transmissions—which proved to be an advantage in the Jan. 17 Hanshin earthquake (Kageki, 1995a).
53 Kokusai Denshin Denwa, International Digital Communications and International Telecom Japan, but universally referred to by their initials.
54 NTT's decision is probably influenced by the fact that its existing narrow-band ISDN network is underused and is a big money loser.
The district of Tokyo where all Japanese ministries can be found.
The Meeting took after the Export Promotion Meeting practiced in order to mobilize the nation’s capacity to its maximum to attain the goal set by the Economic Development Plan. However, the Meeting was held only once at the very beginning of the NBIS project.
57 As an example of unsuccessful system due to the limited consideration on demand side, employment information system, a subsystem in the Administrations project(NAIS), has been frequently cited. One of the objectives of this subsystem was to apply IT in managing the labor-employers’ relations in the Ministry of Labor. The system had been poor in serving this purpose, while it turned out to be useful in having labor market operated in the more efficient way by providing on-line information on demanders for labor and job seekers.
58 Identifying impacts of IT investments has been an important issue in Korea in the sense that it determines the priority of government budget allocation schedule. This issue has now become actively discussed under the name of, so-called, productivity paradox. Popular ideas explaining the paradox are introduced through the academic journals as well as media in favor of IT investment.
59 Data cited in this part describing the current capability of producing and using information technology are, if not otherwise mentioned, from the “White Paper for National Informatization, 1994 and 1995”. The White Paper is annually published by the NCA in Korea.
60 In order to compare the data with world production, we need to refer to another source. According to ETRI cited by a Korean newspaper report (The Electronics Daily, Sept. 25, 1995), the total amount of industry production in 1993 was $20.4 billion. While this number differs from the figure in table 1 probably due to different items contained in the production data, it is mentioned here depending on its availability. The production of $20.4 billion in 1993 was 3.8% of world production which was $539 billion. In the meantime, the other countries’ share in the world production were as follows: the U.S., 31.0%; Japan, 26.1%; Singapore, 3.4%; Taiwan, 3.1%, and Hong Kong, 1.0%.
61 This policy goal was based on ‘one terminal per household’ for the purpose of universal information service. It was similar to ‘one telephone per household’, a goal set at the beginning of 80’s when the ambitious project had been launched to eliminate the notorious excess demand for telephone lines. This goal was accomplished in 1987, when the total number of telephone lines exceeded 10 million.
62 It is pointed out that the accumulated number of PC sold is not the same as the number of PC actually used, since some of the PC’s newly purchased are simply replacing the obsolete ones.
63 Among 503 TICOM computers sold as of 1994, only 96 (less than 20 %) had been purchased by the private sectors. (Dedrick, Kraemer, and Choi ,1994)
64 There are some public corporations which have communications networks for their own business operations, for example, electric power company, highway maintaining company. While they are prohibited from providing commercial communication services to the general public, there has been a hot debate concerning whether they are allowed to do so or not. The debate is still under way in Korea.
65 Privatization of KT was initiated by EPB which announced a privatization plan for public corporations in 1987. According to the plan, 15 % of KT stocks was to be sold to the public in 1989, and additional shares sold annually until 1992, when the government would hold 51% of KT ownership. However the process of selling shares in the stock market had been very slow for such reasons as weak stock market, opposition from the KT, etc. No shares had been sold by 1992, while only a very small fraction was sold to the public based on competing bid during 1993 and 1994.
66 The Association of Cable Broadcasting Firms announced in December 1995 that the number of homes subscribing to CATV had reached to 500 thousand by the end of 1995.
67 The number of host computers instead of users is taken as a variable representing the extent of the Internet use, since it is very difficult and sometimes meaningless to count the number of people on the Internet. Users do not register in a central location before getting on the Internet. Furthermore, Internet providers do not register in one central location because they may be part of someone else’s network. This means that counting the number of Internet users is probably impossible. (P. Hoffman, 1994)
68 The Plan for the KII has been revised by a series of KIITF’s updating efforts since its first appearance in April 1994. The presentation of the Plan in this paper is based on its English version prepared by the KIITF to present in the ICCC (International Conference on Computer and Communication) 1995.
69 There is also a counter-evidence. For example, NSF(1995) predicts that, in analyzing high-tech areas in Asian countries, Taiwan and Korea are the most likely to make the greatest impact in technology related fields and high-tech product markets. In addition, the report also indicates that both economies have significant technological infrastructures in place that should serve to support further growth in high-tech industries.
70 The action plan does not intend to exhaustively list up services and required technologies. It is not possible in advance at the time of preparing the KII Plan. This is particularly so for the rapidly sophisticating information technologies. It is difficult to predict the outcome of something that is changing so rapidly and seems to generate so many misconceptions as it evolves. Since the Plan is supposed to be revised year by year taking the speed of diffusion of network services and technology development into account, the schedule for the available services will be accordingly rearranged.
71 Nobody could even imagine such a widespread use of computer technologies as these days when they began to be used commercially in the early 50’s. As technologies advanced, they have found their own application areas with surprise, while these applications were initially unforeseen, finally to become enormously popular services.
72 KIITF(1995) places emphasis on standards by stating in the action plan for technology development that “From the early stage of R&D, concerted efforts will be given to standardization, including active participation in the international standardization activities.” (p.8)
73 One example for wasting resources has been pointed out in NRC(1994b) regarding the GOSIP. “Indeed, past attempts to influence the process directly have not been effective. The attempt to force the use of OSI protocols by the promulgation of a federal government version, GOSIP, must be seen as a misguided attempt to exercise a governmental mandate. In the commercial marketplace, the contest between the OSI and TCP/IP protocol suite is over: the OSI market has largely disappeared, and vendors who invested enormous sums in trying to develop this market are understandably upset.”
74 In discussing the economic role of standards, J. Farrell notes on the government’s role in the standards setting process not as enforcing compulsory regulation but as providing indicative planning. He writes in NRC(1995): “Government involvement may be less risky in standards setting than in some other forms of regulation, because even standards set with government involvement are often voluntary and can be ignored if unsuitable enough: if there is clearly superior alternative it may replace a poorly chosen standard. For example, the U.S. government tried to use its purchasing power to encourage the use of OSI computer networking protocols, but attractive products were not readily forthcoming, and most other buyers continued to prefer the more established TCP/IP protocol suite. As a result, the government recently withdrew its procurement specification of OSI. Of course, this is at best an imperfect check on errors--the government choices are powerful and can impose bad outcomes. But government involvement in standards, if properly managed, may be more like indicative planning and less like compulsory regulation. This can potentially help private parties coordinate without coercion.” (p.13)
75 One example of research project for standards development regarding the information infrastructures can be found in the B-ISDN project carried out by the KT (Korea Telecom).
76 Organizations concerning the standards in developing information infrastructures in Korea include TTA(Telecommunications Technology Association), PEC(Protocol Engineering Center), OSIA(Open Systems Interconnection Association), IIS(Institute for Industry Standards), KSA(Korea Standards Association), etc.
77 The reasoning behind the conflict is clearly provided in M. Shurmer and G. Lea(1995). “In the broadest sense, standardization and IPRs share the same economic objective--namely to ensure that society benefits to the full from the innovation. However the approach adopted to achieve this objective are very different. IPRs are oriented toward producers and reflect the trade-off between the need to create sufficient incentives for innovation and the public good nature of an innovation once it has been discovered. Standardization, on the other hand, is much more consumer oriented and seeks to encourage a common platform whereby users benefit from enhanced competition and trade(pp. 384 ~ 5).
78 Network construction schedule given in the section on the KII Plan is apparently based on the assumption that the future demand for network services will be successively created enough to require the planned capacity of networks in each stage.
79 The test-bed for the KII provides a tool to evaluate possible problems arising from the context of network evolution, even though this is not the only objective of establishing the test-bed.
80 As mentioned in Section III, the objective of the report on NGNBIS was to identify service applications likely to be available which, at the same time, are able to create their own demand. This was based on the understanding that availability of application services would be a key factor to revive the faltering NBIS project.
81 The remote college education system has been applied in the University of the Air, which was established in order to, through broadcasting, deliver life-time educational services for those who stay in the distant rural areas. By the nature of remote broadcasting education, the University of the Air was selected as the most appropriate place to apply network technologies.
82 For example, according to the Act, the Head of the MIC is supposed to control the management of the fund established to support the promotion of informatization while he does not have sufficient control on some of its sources. In this case, the ministry in control of the source may not cooperate in transferring the fund.
83 One source that had been initially discussed, but finally excluded due to the objections raised by other ministries is the proceeds from sale of the government ownership of Korea Telecom, which is scheduled in the process of privatization. The sale’s proceeds would make the biggest contribution to the Fund.
84 For example, in Jan. 1994, the Japanese MPT released research report on economic impacts of proposed optical fiber network deployment, estimating the size of new applications market equivalent to 56 trillion yen (about $560 billion), the number of created jobs equal to 2.4 million by the year 2010. In the U.S., the economic impacts of proposed legislation regarding the NII had been documented, for example, “Economic Benefits of the Administration’s Legislative Proposals for Telecommunications”, in June 1994, which estimated the economic benefits to the nation that could be achieved through new legislative and administrative reform of telecommunications policy proposed by Vice President Al Gore. Furthermore, a report released in 1994 by the NIITF cites: The Computer Systems Policy project estimates that the NII will create as much as $300 billion annually in new sales across a range of industries; The economic Strategy Institute concluded that accelerated deployment of the NII would increase GDP by $194-$321 billion by the year 2007, and increase productivity by 20 to 40 percent; Industry experts believe that the Personal Communications Services industry, a newly family of wireless services, could create as many as 300,000 jobs in the next 10-15 years.
85 The input-output analysis was adopted not because it was the best tool to estimate economic consequences of IT investments but because it was the only one available based on the availability of data. In fact, the input-output framework could be the worst choice in the context of IT application than any other case, since this framework is based on the clear division of various industries while it is highly probable that new industries emerge and existing ones disappear as the applications of advanced technologies evolve.
86 By global, we emphasize the value of technologies enabling the NII from the point of network connectivity. The word “global” does not deny the fact that each country’s network activity reflects the unique characteristics of its local environment. For example, while the Internet was developed largely in the U.S., the Internet activity in Korea is taking place under its own economic and regulatory framework.
87 Indeed, this is a stated reason by many nations as to why they are developing their information infrastructure (BPT, 1990; MPT, 1989; TCOJ, 1994; MITI, 1994; NCB, 1987, 1992; Bangemann Group, 1994).
88 Harris makes this point in his arguments for deregulation of telecommunications, but it is more widely felt as well. For example, in response to a question about whether Microsoft should be broken up, Newt Gingrich said “All we see right now are theoretical worries by a lawyer and the Department of Justice about a system that doesn’t exist, which could screw up the American capacity to dominate the world market” (Dyson, 1995:160).
The federal government has subsidized technology investments in general through R&E tax credits since 1981. Most software development does not currently qualify as eligible for R&R tax credits, but extending the range of software eligible for R&E tax credits would be a stimulus to infrastructure investments because software is an increasingly important part of computer and communications equipment. Investment tax credits were eliminated in the 1986 Tax Reform Act and legislation has been proposed to reenact them, but has not been passed by Congress.
90 Although the Communications Act of 1934 created a new institution—the Federal Communications Commission— to regulate interstate telecommunications, most of the telecommunications sections of the 1934 act were transferred wholesale from the Interstate Commerce Act of 1887 which was designed to regulate the interstate traffic of railroads. And this was despite many efforts in Congress to alter the provisions of the act to match the differences in the technologies and market structures between railroads and telephones.
91 Some predict that the Baby Bells may merge with one another. Bell Atlantic and Nynex reportedly are discussing merger, and it is argued that the others will need to combine or merge to compete with AT&T, MCI and Sprint and the media giants (James Flanigan, LA Times, Sunday, December 24, 1995.
92 For example, Microsoft says that someone watching the U.S. budget debate could call up Senate and House versions of the budget, plug in their income and other relevant numbers and get estimates of how each bill would affect their taxes. Or, someone watching the news on Bosnia could get access to maps of the area, video clips of fighting in the region in the nineties and during World War II and historical information about why the region has been so prone to conflict (Helm and Hall, 1995).
93 The ARPANET was designed to test the feasibility of building a packet-switched communication network that could survive nuclear attack and remain functional. It was also hoped that this communication network would help defense-sponsored scientists to share expensive computer resources. The ARPANET took shape in university research laboratories in the western United States in the late 1960's, and eventually grew into a hugely successful communications utility. From two dozen sites in 1971, the ARPANET grew to more than 200 sites by 1981. The initial military purpose was served, and the capacity for sharing computer resources was proven. But more important, the ARPANET had created a new form of communication infrastructure not previously foreseen, and by the 1980's only hinted at (Hiltz and Turoff, 1978). Geographically separated university researchers were united by technology through applications such as electronic mail, file transfers, news groups, and bulletin boards (Sproull and Kiesler, 1991). Sites in other countries were connected, initially via military installations, but the international dimension of the ARPANET grew rapidly and by 1993, 160 countries were being served (Abbate, 1994; Newsweek, August 8, 1994).
The ARPANET was not the only important networking venture. Several large companies also built their own private networks. For example, IBM built VNET which, for part of the 1980's, had more nodes than the ARPANET. There was also growing pressure during the 1980's to expand access to computer networks beyond the narrow academic domain of DARPA-sponsored computer scientists, and into non-DARPA computer science and engineering programs. The National Science Foundation's CSNET program was instituted to provide such services, and signaled NSF's entry into the academic networking business. Another network called BITNET was formed to provide access to networks in the sciences, social sciences, and professional schools (Gurbaxani, 1990). The research and development departments of many commercial companies also linked themselves to the emerging network world. This growth of heterogeneous networks began causing problems when users on one network wished to communicate with users on another network. To make such communication possible, "gateways" were established to link the networks together. The gateways made obvious the problems that arose from lack of standardization around such fundamental policies as address form. In an effort to straighten out the mess, a set of conventions was drafted and promulgated throughout the networking community to standardize key aspects of network operation. This was the genesis of the Internet, which is not a network itself, but an overarching network of networks. Also, as the Internet evolved, the role of defense declined and the governmental support apparatus shifted to the National Science Foundation. The NSFNet built to link NSF's national supercomputer centers became the backbone of the Internet, and the primary laboratory for the testing of new technologies such as the one-billion bit per second "Gigabit" testbed.
94 In fact, all messages sent to the President and Vice-President are loaded onto tape and stored. At this time there is not even provision for an automatic answer to acknowledge receipt of a message by the host machine containing those addresses.
95 Even Microsoft Chairman Bill Gates expects tough sledding. According to an article in the LA Times (Shiver, August 23, 1995), Gates recently said “Its the first time in our history we’ve gone into a business where there will be substantial losses early on.” But, he was quick to say that going on-line will eventually “become the preferred way to learn and collaborate.”
96 Commentators on NII often talk of information delivery through conduit, a notion that was explored in detail in precursors to the NII vision such as the "wired city" (King, 1987).
97 This is also happening among providers of information services. For example, considerable partnering has evolved following the success of Electronic Data Interchange (EDI) systems such as Singapore's TradeNet (King and Konsynski, 1991; Neo, King and Applegate, 1993). DHL, the integrated forwarder, has joined with IBM for the development of Tradelink in Hong Kong to learn what EDI means for its current business, to position itself for a role in EDI, and to find new business opportunities that exist because of EDI (Kraemer, Dedrick and Jarman, 1994).