Click for next page ( 41


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 40
4 Services and Systems Integration Once thought of as separate entities, telecommunications and computing equipment have merged to become '`information technology," the umbrella term for the growing array of digital devices and software that gather, process, dis- play, and exchange data and information. According to the Department of Commerce, revenues for computer professional services (e.g., contract pro- gramming, systems and network management, education and training on the use of computer systems) were expected to reach $37 billion in 1989.1 The still- evolving merger of technologies has added emphasis to the notion that informa- tion is a strategic asset that every competitor must have. This notion has spawned a large and rapidly growing market for information-related services: database and videotext services, contract programming and design, value-added networks, training and education, data processing, systems integration, and oth- ers. Annual rates of growth in nearly all segments of the services industry exceed 10 percent; in some, growth rates top 20 percent and are expected to remain at that level for the near future.2 One booming segment, and potentially the most lucrative, is systems integra- tion, which entails designing, implementing, and maintaining complex systems of computing and communication equipment and software, often in combina- tion with other types of equipment and systems. Systems-integration firms help businesses and other clients achieve the desired interoperability of their infor- mation technology. The systems mesh technology-often a combination of equipment, software, and network-based services supplied by many different vendors with an organization's activities, operations, structure, and planned 40

OCR for page 40
SERVICES AND SYSTEMS INTEGRATION 41 development. Some firms specialize in turnkey systems, but most provide cus- tomized systems that are developed in accordance with customers' needs. In 1989, system integrators were expected to earn approximately $13 billion, with established firms reporting annual revenue increases of more than 15 percent in their domestic business and more than 25 percent in their foreign operations.3 Hardware and software manufacturers, accounting firms, communications companies, and other types of companies have targeted the systems-integration market, and the number of businesses diversifying into the market is growing rapidly. In 1988, more than 1,450 U.S. firms earned all or the bulk of their rev- enues from systems integration.4 The U.S. market for systems integration is the world's largest, and it has attracted Western European and Japanese firms. Entry is achieved almost exclusively through acquisitions of U.S. firms rather than through start-up subsidiaries. SYSTEMS INTEGRATION: A U.S. BUSINESS CONCEPI The U.S. federal government launched systems integration in the 1960s, when federal agencies hired contracting firms to design large-scale systems to coordinate data processing or communication operations. Although still young and growing, the U.S. systems-integration industry is in a much more advanced stage of development than that of the emerging foreign competition. The differ- ence, according to Jeffrey M. Helter of Electronic Data Systems (EDS), can be attributed to U.S. fines' early recognition of the business opportunities in inte- grating computing and communications equipment that organizations were assembling from various vendors as they tried to keep step with technological advances and the apparent opportunities they created. Building on expertise acquired in work done for the federal government, pioneering firms have recently branched into commercial markets, which were slow to develop at first but have been gaining momentum with the growth of private networks and other large-scale private applications. "ET]he gradual move in the industry toward open systems architectures and networking standards," said Paul A. Turner, executive director of the Price Waterhouse Technology Center, "created an opportunity for brokers to interpose themselves between clients and the traditional proprietary hardware and soft- ware vendors." For prospective clients, Turner added, systems-integration firms offered an "alluring" service: "a reduced tie-in to the proprietary systems of any one ven- dor, a more cost-effective solution to the needs of the client, made possible by 'cherry-picking' the best of the offerings of several vendors, and all this coupled with a single source of management and accountability that gave the chief infor- mation officer some degree of career insurance." The attractiveness of the concept has been proven, compelling even vendors

OCR for page 40
42 KEEPING THE U.S.COMPUTERINDUSI2YCOMP~WE of hardware to set up systems-integration units that, in principle, are free to use the products of other manufacturers. Given the bright forecast for the industry, the stream of newcomers may increase. "Systems integration," said Irving Wladawsky-Berger, vice president of IBM's data systems division and general manager of their Kingston facility, "is an area we all expect to continue to grow for a very simple reason: The same forces that are driving the technology and producing the tens of millions . . . of workstations with the networks and huge data bases, those same forces are driv- ing complexity. This business would dry up if complexity disappeared, but I think there is about as good a chance of that happening as finding a parking space in Manhattan." ELEMENTS OF SUCCESS Being the first to market with a commercially valuable service is one obvious and important factor contributing to U.S. firms' dominance of the systems-inte- gration market. Another is the large domestic market that served as a spring- board for expansion into foreign markets, which are now growing more rapidly than the one in the United States. Wladawsky-Berger and MCC's Belady suggested that an American flair for mastering complexity may also be fundamental to the strong showing of U.S. systems-integration firms. "Management of complexity is a most American endeavor; it really plays well to the things we are good at," Wladawsky-Berger said, summarizing the position thus: "It is skills intensive; it is highly technical and, so, our R&D effort will serve us well." Turner of Price Waterhouse elaborated on the implications for skills. "As often happens," he said, "the major advances come from effort in those areas that fall between the cracks of the conventional disciplines. The skills . . . nec- essary to perform a radical reevaluation of system functionality include the con- ventional technical-system skills, but also vital are the skills needed to under- stand the business, [its] management, and the organizational issues. And also required, critically, is a problem-solving process that gets the people involved and that is all of them, both users and systems designers, thinking outside the box and beyond the obvious solutions." Inherent in the process of integrating multivendor information technology is the need to form flexible partnerships, not unlike the horizontal and vertical cooperative relationships that many have advocated for other industries in the computer sector. In the European market, for example, EDS competes "rather heavily at times with IBM," Heller explained. However, he noted, "we team together sometimes and [EDS is] their customer and they are ours. So, we have complex relationships." "Seldom does a system integrator have all the resources needed to do a par- ticular job," Turner added, "and these partnerships increasingly cross national

OCR for page 40
SERVICES AND SYSTEMS INTEGRATION 43 boundaries and lead to a very complex set of interdependencies, but ones that are required in order to cope with the diversity and complexity of technology ...." Other factors that underlie the success of U.S. systems-integration firms are neither unique to this market nor cultural in origin. Rather they are good busi- ness practices, sometimes applied in ways that may go beyond conventional expectations. Foremost among these is the demonstrated importance of devel- oping close working relationships with customers, a point made by all industry representatives at the colloquium. Customers want, according to one speaker, a "total applications solution," requiring intricate understanding of a client's oper- ations. Price Waterhouse's Turner noted that businesses have invested heavily in information technology, but many, particularly those in (other) service indus- tries (e.g., financial services), have not realized expected increases in productiv- ity. In fact, the service sector has experienced a"clear slowdown in productivi- ty," he said, after an "unprecedented spending binge" on information technology. Systems integration can solve this dilemma, but it requires a "back- to-basics questioning" of the functions served by information systems that have often evolved in piecemeal fashion. Needed applications of information tech- nology may be quite different from those considered to be common practice. Speed is another requisite attribute. Systems-integration firms must meet changing needs and opportunities, placing a"premium on being able to move fast," Turner said. "Solutions to problems that can be solved with yesterday's answers can be provided by the client's own resources." Given the intricacy of the working relationships between systems-integration firms and their clients and the often substantial investment that integration requires, industry representatives stressed the importance of a local presence in foreign markets. Closeness and longevity of contact can also overcome the sus- picion that new entries often encounter in other nations. DIVERSITY IN THE WORLD MARKET "Networked computing is a galloping customer requirement" throughout the world, according to Sam R. Willcoxon of AT&T. Ongoing deregulation of telecommunications services in many countries and the confluence of communi- cations and computer technology could fuel "exponential" growth in overseas markets, he said. In the European Community, the competition facing U.S.-based firms comes from two primary sources. One is in-house competition. Many organizations draw on internal resources and technical staff to design and develop their own systems, according to Helter. 'what has always been our [EDS's] competition as we tried to get involved in this business back in the '60s," he said. "So, we feel like from a marketing standpoint, we know how to address in-house compe

OCR for page 40
44 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE titian," he added. Large hardware- and software-manufacturing firms- so- called full-service providers-constitute the other category of competition. In addition, European firms that produce software are only beginning to expand into systems integration, according to Heller. Several companies have emerged as "formidable" competition in niche markets for specialized systems, such as for computer-aided design and computer-aided manufacturing. Japan, South Korea, and other Far Eastem nations are more difficult markets for systems-integration firms. Heller described the use of outside firms for internal tasks as "countercultural." 'whey like doing everything in an integrated sense and for themselves," Heller said. "It is a sign of weakness" to hire an out- side firm and entrust a contractor with the responsibility for enhancing function- al and strategic capabilities. Nevertheless, EDS has managed to make some inroads in the Japanese market. Turner added that Price Waterhouse's Japanese office encountered stiff resis- tance upon its entry into the systems-integration business. Over time, however, the firm's marketing efforts have become progressively more successful. At this point in the industry's development, as Heller observed, systems inte- gration as an exportable service is largely a U.S. invention that the world market is gradually emulating. AT&T's Willcoxon noted that the concept of systems integration is still evolving, but at a differential pace faster in the United States than in most other parts of the world. Interestingly, early advances in overseas markets to date have been facilitated by business relationships launched in the United States. For example, the requirement for a local pres- ence overseas by accounting firms like Price Waterhouse that serve multination- al clients has provided a springboard for serving foreign clients, a development noted by Turner, while the merger of EDS with General Motors opened doors to new relationships with foreign firms for EDS, according to Heller. DEPENDENCE ON OTHER COMPUTER-RELATED INDUSTRIES That the global systems-integration market is poised to take off-offering an "exponential" increase in business opportunities, in Willcoxon's view is not disputed. Nor is the characterization of U.S.-based enterprises as world leaders in this market. But will the apparent U.S. advantage deteriorate if the underly- ing industries in the U.S. computer sector- hardware, telecommunications equipment, and software manufacturing falter in the world marketplace? Major players in this market acknowledge its dependence on other computer- related industries. Strategically, Heller pointed out, EDS views the health of the other industries in the U.S. computer sector as directly affecting the health of domestic systems-integration firms. "lilt behooves us to have a very strong force in both software and hardware domestically," Heller said. This is so even though, in the short term, the lower cost or higher quality of some foreign-made goods may influence his firm's buying decisions, or local laws may require use

OCR for page 40
SERVICES AND SYSTEMS INTEGRATION 45 of components manufactured in the home country. Electronic Data Systems carries "a fair amount of capability in the software arena," Heller noted, "but not to the extent that we want to shift a major portion of our investment into carrying it as fully as the industry itself. In the hardware arena, we have no interest in that kind of shift in capital investment. So, long term; we are willing, able, capable, and have the intention of supporting the domestic capability, whether that means in terms of our dollars, our energy, or our political influence.... [Wle see it as a team effort." STANDARDIZATION On scales small and large, systems integration is made easier by standards: The greater the level of standardization, the greater the level of interoperability between computers and other digital devices as well. Today, as noted in earlier chapters, idiosyncrasy rather than standardization is the norm, and systems-inte- gration firms have devoted some of their resources to overcoming barriers to compatibility. "We have had to formulate solutions where standards didn't exist," Heller said. From an industrywide perspective, the process may be inef- ficient, a duplication of investment and effort. Standards can motivate innova- tion by serving, explained Heller, "as the act of declaring innovation significant . . ., [establishing] a benchmark that the next round of innovation must signifi- cantly surpass-or leapfrog-or sidestep." But without adequate safeguards for protecting intellectual property, according to Heller, standardization will quicken the transfer of innovations from inventor to imitator. Therefore, if stan- dardization is to increase, there must be stronger safeguards for preserving the returns to innovation and greater international compliance with those laws, in Heller's view. AT&T's Willcoxon contended that without greater standardization, the "mag- nificent opportunity for productivity improvement" that computers afford would go largely unrealized. "Users are beginning to reject closed proprietary systems that isolate their data in computers that can't talk to each other," he said, adding that clients want to "mix, match, and interconnect products from whatever ven- dors meet their needs." Standards are critical to meeting customers' varied needs, enabling the infor- mation industry to "really begin to deliver on its promises," according to Willcoxon. Willcoxon called for an "enlightened combination of cooperation and competition." He said that firms must cooperate in developing and select- ing appropriate standards and then compete in "devising solutions that are best tailored to user needs." Turner of Price Waterhouse offered a complementary view. While acknowl- edging fears that standards may hurt innovative software and hardware compa- nies, he urged firms to view the issue from the perspective of users who have invested heavily in information technology but who have not reaped the antici

OCR for page 40
46 KEEPING TlIE U.S. COMPUTER INDUSTRY COMPETITIVE paled benefits. If standards can deliver "better productivity enhancement," then firms must accede or risk "increasing dissatisfaction from users." But would high levels of compatibility between the computer and telecom- munications equipment of different vendors eliminate the need for systems-inte- gration firms? Absolutely not, according to industry representatives. "where are a lot more things involved than just whether boxes and software hook together compatibly," Helter said. The "real value-added" part of systems inte- gration is the "understanding of the problem you are trying to solve." Given the number of hardware and software options facing customers and the many potential approaches to solving a problem, systems-integration firms are not likely to see complexity or demand for their services evaporate with increasing standardization, according to Helter and other industry representa- tives. Raj Reddy of Carnegie Mellon University shared the following analogy: "Just because you have a vocabulary and a language and a syntax and a dictio- nary, it doesn't mean you can write a best-selling novel. A best-selling novel is the composition of these things in ways that make an attractive plot and a solu- tion and various other things. I think no matter how many standards you have, you still have that problem of creating the best-selling plot and novel. That is the systems-integration business." Perhaps, as several colloquium participants suggested, the debate over stan- dards should shift from whether they are needed to how standards are developed and how they influence innovation and technological progress. Robert E. Kahn, president of the Corporation for National Research Initiatives, suggested that rather than viewing them as "unchanging over time," standards should be con- sidered dynamic, evolving in step with technology. With a flexible standard- setting system, the "critical issue" becomes how to manage standards so they accommodate change, in Kahn's view. INFRASTRUCTURE FOR THE ~FORMA~ON AGE Firms that specialize in computer and communications services are, in effect, building an Information Age equivalent of the highway system, an infrastruc- ture that accommodates the flow of information in textual, graphic, and voice form. Systems-integration firms, for example, link once-independent islands of hardware, applications, and their specialized domains of software into a func- tioning whole, just as roads connect previously isolated communities into zones of commerce. Like the businesses that now line many stretches of the nation's highway system, a growing array of electronic databases and other specialized services have sprouted up to serve an increasing number of users whose com- puters are connected to networks. Colloquium participants argued that this movement toward network-based enhancement and integration of computer and communication capabilities is proceeding much more haphazardly than did the development of the national

OCR for page 40
SERVICES AND SYSTEMS INTEGRAT10~1 47 highway system; the full promise of networked computing in which one com- puter can communicate with any other, regardless of its location or its His far from being fulfilled. Many obstacles stand in the way of making computer- to-computer communication as easy as making a telephone call. The potential economic advantages of a nationwide information network appear to be moti- vating Japan and Western European countries to begin developing the necessary infrastructure. In the United States, where the geographic scope and scale of the task and its cost are significantly greater, a cohesive approach is lacking. In recent years several groups, including the Computer Science and Technology Board, have called for the development of a national information infrastructure or a partial implementation of this concept devoted to the support of research and development s 6 At the colloquium, Charles Ferguson of MIT explicitly advocated a government-coordinated program to build a nationwide advanced-technology computer communications infrastructure over the next one or two decades. The network would be the conduit for communications required by an ever-growing number and variety of digital devices, from com- puters and telephones to high-definition television sets. According to Ferguson, a national information infrastructure will not only become an economic necessity, but its development also will be a direct stimu- lus to a "very wide spectrum of information- and technology-intensive indus- tries." Manufacturers of base technologies for semiconductor packaging, automated assembly, optoelectronics, and production of printed circuit boards and fiberoptic cable would benefit from infrastructure-oriented research pro- jects and, ultimately, from the demand generated by the actual construction and implementation of the network and the associated user equipment. Moreover, the proposed network could be the means for U.S. reentry into markets for con- sumer electronic equipment, which increasingly incorporates the same digital technologies used in computers (see Chapter 2~. Some colloquium participants equated a government-led project to build a national information infrastructure with the U.S. decision in the early 1960s to send astronauts to the moon by the end of the decade. The "symbolic value" alone of such a project could "galvanize the technical community" and the rest of the country, suggested Wladawsky-Berger of IBM. But more important, Wladawsky-Berger and others stressed, is the great underlying utility of having a networked economy. "If U.S. industry generally is served by inferior net- works and digital systems," Ferguson has written, "the economy will suffer. Conversely, if the United States were to possess the world's most advanced [information] infrastructure, the economy would benefit, and so probably would U.S. vendors of digital information systems."7 One significant benefit would be to open up new markets for information and information-processing products markets that the United States might under- stand and exploit before foreign competition does. A possibility is to computer- ize commerce, using digital communications among computers acting as agents

OCR for page 40
48 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE for companies to find suppliers, determine price and availability of parts, nego- tiate deliveries, convey shipping information, and deal with problems. This array of services goes far beyond what electronic data interchange achieves today. It is an example of the kind of service that needs careful fostering by government and other institutions; it is not simply a university research project. Although no one at the colloquium opposed the idea of a national informa- tion infrastructure, levels of enthusiasm for the concept varied. Heller of EDS advised proponents to define clearly the national purpose that would be served by such a network. He also cautioned that private firms would not embrace the proposal without government leadership and government-sponsored incentives for companies that contribute to developing a public good. Without incentives, firms will act in their own self-interest and avoid projects that do not benefit them directly. Proponents of a national information network also injected cautionary notes into their endorsements of the concept. For example, Stardent Computer's Gordon Bell, who helped craft a proposal for a national research and education network while he was an assistant director of the National Science Foundation, warned that continuing erosion of the nation's manufacturing base may mean that many infrastructural components, such as switches and optical fiber, may have to be purchased from foreign suppliers. He also advised focusing first on networking for researchers, to avoid making the task so large and complex as to be unmanageable. Ferguson of MIT has cautioned that if proprietary controls of network-orient- ed cooperative research projects are lax, foreign firms may benefit more than U.S. participants. There is also the danger, he maintained, that policies intended to preserve the U.S. market for domestic Foss could result in "crude protection- ism." If policies are designed to benefit only a few selected industries, the com- petitiveness of others that produce goods and services based on digital technolo- gies could suffer, Ferguson suggested. A potential outcome is greater foreign penetration of U.S. computer and networking markets, in his view. "I think it is the government's business to provide infrastructure," said Robert W. Lucky, executive director of the communication sciences research division at AT&T Bell Laboratories. "I don't think there is money in this right now, but I think the nation needs it.... tW]hat it requires, I think, is govern- ment leadership, not government regulation." NOTES 1. U.S. Department of Commerce. "Information Services," 1989 U.S. Industrial Outlook, (Washington, D.C., 1989), p. 45-2. 2. U.S. Department of Commerce, 1989 U. S. Industrial Outlook, p. 454. 3. U.S. Department of Commerce, 1989 US. Industrial Outlook, p. 454. 4. U.S. Department of Commerce, 1989 U S. Industrial Outlook, p. 45-2.

OCR for page 40
SERVICES AND SYSTEMS INTEGRATION 49 Computer Science and Technology Board, National Research Council. The National Challenge in Computer Science and Technology (Washington, D.C.: National Academy Press, 1988~. 6. Computer Science and Technology Board, National Research Council. Toward a National Research Network Washington, D.C.: National Academy Press, 1988~. Ferguson, Charles H. "HDTV, Digital Communications, and Competitiveness: Implications for U.S. High Technology Policy," VLSI Memo No. 89-506, Massachusetts Institute of Technology, VLSI Publications, February, 1989, p. 6.