Click for next page ( 18


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 17
4 Views of Workshop Participants DISCUSSION OF THE PREMEETING SURVEY RESULTS The participants at the workshop, having been handpicked as being actively involved in standards issues, were viewed as a principal source of input to the workshop discussions. A premeeting questionnaire was designed to collect their views in advance of the meeting to allow study and summarization. The large response was both gratifying and informative. Several invitees who knew they would be unable to attend the workshop also submitted answers to the six questions that had been posed. In addition to questions about information technology standards, the participants were asked to characterize themselves in terms of their roles in the information technology industry. It was hypothesized that their views would depend on affiliations or level of management responsibility. This does not appear to have been the case, based on an informal analysis of the data. The complete text of the responses is available for a more formal analysis of response correlations; however, the strong consensus reached in the conference conclusions provides little motivation for further analysis. The analysis of response content described here was done in two stages. First, the National Research Council staff provided an abstract of the key points that had been made by the respondents, largely based on a clustering of direct quotes from those who presented their ideas most cogently. That abstract of information was provided to workshop participants along with the complete set of verbatim responses. 17

OCR for page 17
18 Fred Andrews, the workshop chairman, further distilled the clusters of statements that had been abstracted into a collection of short statements related to each of the six questions. The purpose of this second step, admittedly subjective, was to provide a baseline on which the workshop dis- cussions could focus and interact. The material that follows here discusses that baseline for each question and includes the further conclusions from the interactive presentation led by Mr. Andrews. Question 1: What is the place of a strong information technology industry? The written responses focused on the following points: (1) It is of critical importance to competitive advantage because: it is a significant portion of U.S. exports. it supports other U.S. industries. it can be a boon to productivity. (2) The success of the technology in productivity enhancement de- pends on matching user needs. (3) The U.S. public network infrastructure is lagging. This lag will be most detrimental to the productivity of small to mid-sized companies. The discussion tended to reinforce the first two main points in this set of statements, supporting both the importance of information technology and the dependency of success on matching user needs. There was general consensus that information technology is a major element in US. economic competitiveness. The views on the lagging U.S. public network infrastructure are widely held but still arguable. At this time the opinion is based more on the perceived greater vector of change in Western Europe and Japan than on the absolute level of public network services. The one notable exception in which another country is moving ahead is the French Minitel deployment. The statement of the dominant eject of a strong public network on small to mid-sized companies was further discussed. Such companies have depended more on public network offerings than on private networks for economic reasons. Hence, this void in the public network may leave them further behind large businesses in this vital area. However, since large organizations deal with small businesses as both suppliers and customers, all suffer when information exchange is not most cost effective. The consensus was that both large and small companies, as well as government and private citizens, require ease of access to information. Thus, it was concluded that a strong information technology infrastn~cture within the United States is absolutely essential to our continued success as an advanced industrialized nation.

OCR for page 17
19 Question 2: How do information technology standards enhance the effec- tiveness of U.S. enterprises? The written responses focused on the following points: (1) Open standards are a catalyst for growth and global connectivity. (2) Global connectivity will allow the United States to exploit superi- oriV in software applications. (3) Standards provide a mechanism for legal cooperation among U.S. companies to maintain U.S. market share. (4) It is advantageous for U.S. technology to prevail in standards. (5) A minimum set of standards is essential to ensure compatibility. (6) Room must be left for product differentiation. (7) Information technology standards on an interindustry basis will al- low the United States to be the world's most modern and efficient economy. This audience strongly supported the importance of open standards for growth internationally as well as nationally and was confident that the United States would do well in open and fair competition. Members of the information technologies industry could work together more effectively as teams within the framework of standards and still retain the ability to differentiate their products from those of their competitors. There was support for the sentiment expressed in the fourth itemthat it Is advantageous for standards to conform to U.S. technology. However, it was far more important to respondents that there be standards at the international level, regardless of whose technology forms the basis. The fifth and sixth items state the well-recognized need to strike a balance between standards that are sufficient to achieve compatibility but not so constraining as to unduly limit product differentiation. Subsequent discussions made it clear that in general the dominant problem has been too little compatibility rather than too little product differentiation. The lack of compatibility stands in the way of efficient interchange of information among service and manufacturing entities. In particular, electronic data interchange is only possible in those industry subsegments that are led by dominant players or by trade associations. The nation or group of nations that implements standards most widely will have significant cost advantages in world markets. In summary, the consensus was that standards are My to the development of a strong information technology infrastructure. Question 3: How do information technology standards impair the effec- tiveness of U.S. enterprises? The written responses focused on the following points:

OCR for page 17
20 (1) When standards are too inflexible or promulgated too soon, they inhibit the growth of innovative technology. (2) When standards development is too slow in coming, confusion and uncertainty are introduced into the market. (3) Leading-edge system providers and users risk having systems that are incompatibile with ultimate standards. (4) Standards and their interpretation are used by multinational re- gions to block market entry. (5) The rest of the world influences U.S. standards more than the U.S. influences the standards of the rest of the world. (6) Foreign competitors exploit U.S. standards faster and better than do U.S. enterprises. With regard to the first two points, the need for balance between standards that are too early and too slow emerged from the written survey responses. However, when the workshop group was challenged to provide examples of standards that have been arrived at too soon, little was offered. In the case of the typewriter keyboard, which was developed 100 years before it was associated with computers, the positions of the keys are not optimum. The keyboard configuration is as firmly embedded in our culture as is our system of English measurement units. However, it is ridiculous to argue that the standard should never have been set in the first place. The market confusion and uncertainty mentioned in item 2 are best illustrated by the proliferation of proprietary information networks that has taken place. It is extraordinarily difficult for a user to solve the puzzle of interoperability even for such basic needs as electronic mail. Another example of the failure of a market to develop because of a delay in promulgating standards is that of AM stereo. With regard to the third point, the question is, who will risk being the leading-edge system providers when the rest of the industry has the opportunity to second guess those features necessary for compatibility? The consensus at the workshop seemed to be that standards work must anticipate to a greater and greater extent the application of technology before the market is stifled by incompatible products. The participants concluded that the foremost challenge to the individual standards committees is to anticipate the application of technology, rather than to t7y to catch up. The last three items reflect the view that, for a variety of reasons, standards can work to our disadvantage with regard to international com- petition. However, the call of the group as a whole was not to eliminate standards but, rather, to eliminate any abuses of the process and any ineptness we have in dealing with the opportunities that exist.

OCR for page 17
21 Question 4: What issues in the U.S. standards developing process must be resolved? The written responses focused on the following points: (1) Interoperability requires that standards lead technology, but the adoption process is too slow. (2) The process lacks top-down leadership and commitment from industry and government. (3) The process of developing and controlling U.S. positions to the In- ternational Consultative Committee for Telephone and Telegraph (COMIC needs improvement. (4) U.S. antitrust law inhibits collaborative efforts to develop new standards and ensure effective implementation. (5) Checks and balances in the approval of standards avoid even the appearance of anticompetitive behavior at the price of slow progress. (6) It is difficult to integrate independently developed products and services through our profoundly democratic process. Of these six points, the strongest support of this group coalesced on the need for standards activities to lead the application of technology more effectively than they do at present. Slow progress is blamed more on the lack of leadership and motivation of standards participants than on defects in the process. There is no organization that sets goals for a standards- based infrastructure and then leads the various bodies to develop standards to meet those goals. Such leadership would motivate all participants to expedite the process. There was not strong support for legal or regulatory action to correct the problem identified in item 4. It is true that antitrust paranoia exists in some quarters, but it is not clear how much that has actually chilled the work on standards. The point was made that standards participants are too timid in working with others, in view of the lack of evidence of any unwarranted enforcement actions. Even though there was no such evidence, some of the participants from the service provider segment felt that a change in the law is in order. Viewed as more important than overcoming legal obstacles was the creation of a sense of urgency about standards from top to bottom through- out the information technologies industry. This means setting aside at least some competitive tactics for the common good of all. The industry leaders should show leadership and make decisions when nearly equal choices are presented. Many of those present appeared unwilling to drop the checks and balances of the democratic standards developing process in the interests of speed.

OCR for page 17
22 Question 5: Are there fundamental public policy changes that could help? The written responses contained a wide selection of proposals for change. The following points appeared several times: (1) Move the balance between protection against anticompetitive be- havior and the acceleration of standards toward acceleration. (2) Broaden the National Research Collaboration Act to cover standards-related activities. (3) Modify copyright law to protect unique expressions within stan- dard interfaces for users and document exchange. (4) Shift complete standards responsibility to the Federal Communica- tions Commission (FCC), striking a balance between deregulation and imposed interoperability. (5) Establish a process for standards input for those who cannot afford to fully participate. (6) Create an Information Technology Standards Research Founda- tion to support research and a data repository. (7) Insist on foreign reciprocity in establishing open standards devel- op~ng processes (8) Make standards compliance a prerequisite to competing for public business, with local government and large business following suit. The group did not rally around any of the first six public policy changes that had been suggested by the participants in advance of the meeting, although there were supporters for each. The last two points, that the government take a tough trade stance to force reciprocity in open standards processes, and that agencies insist on standards compliance in their own procurement, were generally supported. There was much discussion on the subjects of items 2 and 3. Broad- ening the National Collaborative Research act to cover standards-related activities would remove the threat of treble damages in antitrust actions. As mentioned previously, this was not seen as a major factor in today's standards developing process. As with proposals to modify antitrust laws, there was no ground swell of support for modification of the copyright laws to protect unique expressions within standards. This and the other suggestions in the group of answers were not sufficiently specific to provide a basis for recommendations or conclusions. The most negative response to any of the above suggestions was to giving more power to the FCC, item 4. There was considerable support for the foundation idea (item 6), but there was no agreement on how the foundation would be funded.

OCR for page 17
23 Question 6: What are the key issues relative to the integration of multi- vendor products? The written responses focused on the following points: (1) The development of complete and unambiguous specifications. (2) The timely resolution of inevitable oversights and differences of interpretations. (3) Government and industry agreements in a single, worldwide, con- sistent set of test requirements and test systems. (4) Liability for the failure of "certified" products to work together and meet user needs. (5) Rapid resolution of issues that favors end users rather than special interests. In the discussion, the attendees agreed that these are all nagging problems to those immersed in trying to make information systems function. Ken together, they suggest that an approach is needed to manage the life cycle of standards from concept to implementation. In an imperfect world, the best hope for solving many issues is for rapid prototyping of standards to uncover the inevitable oversights and imperfections. In an era in which the quality of products and services is equated with the satisfaction of customers, resolution of issues in a way that favors end users is generally accepted as making the most sense. MAJOR CHALLENGES Following the general discussion of the written responses to the six questions, the participants separated into small groups based on the type of industry they represented. Even though the interests might seem to be different, the challenges to the industry that the different groups saw were very similar. The following were the main points of agreement. . In a large majority of involved U.S. organizations, top management does not appreciate the benefits and costs of adequate participa- tion in the standardization processes. The United States should be a leader in the standardization pro- cess through the international standards bodies, which are the International Organization for Standardization (ISO), the Interna- tional Electrotechnical Commission (IEC), and the International Telecommunication Union (ITU). We must develop a better U.S. (North American) leadership struc- ture if the United States and Canada are to be leaders in interna- tional standards.

OCR for page 17
24 The United States must maintain a state-of-the-art information infrasoucture. The infrastructure must be based on standards that are accepted internationally so that U.S. companies can sell their products and services worldwide. There is a need for an organized approach to manage expeditious prototyping of proposed standards on an end-to-end basis to ensure ~nteroperability of new standard products with the existing standardized infrastructure. Of the above issues, the most critical one is the lack of concern by top management in moving essential standards forward rapidly to implementa- tion. The greatest need for an increased contribution to the standardization process is with the user community. In the information technology arena, standards can be developed before the availability of products. Applica- tion services are a major force for current standardization. Users must be involved if these standards are to meet their needs. Users have knowledge to contribute that complements vendor skills. No many in the user com- munity are still in the habit of waiting for the vendors to offer products before quantifying what the real needs are. In the community of product and sentence developers, management often puts standards at too low a priority. All too often the people who work on standards do not have the authority to compromise and come to closure on issues. The process could be speeded up considerably if top management understood the importance and the value of reaching consensus on good standards expeditiously. The issue of leadership is the second most critical, although it may be that if there were leadership, then top management would under- stand the problems. The American National Standards Institute (ANSI) is perceived as a reactive body and its present structure does not provide leadership on information technology standards. It accredits standards- writing organizations and waits for those organizations to develop draft standards through established procedures. Then, ANSI coordinates the draft standards through final review and promulgates the standards. It is not uncommon, in the final coordinating process, to find that another accredited group is processing a standard that competes at least in part with the draft being coordinated. A process (organization) is needed that involves all interested groups and that identifies needs, brings together ideas from interested parties, resolves conflicting views, and produces standards in a reasonable period time. The participants felt that, although the government could help, a government standards writing agency was inappropriate. Many felt that ANSI could provide this leadership if it redefined and expanded its role

OCR for page 17
25 and mission with respect to information technology. This would require an amended charter. As the above problems are overcome, the workshop participants felt that the United States should have and maintain a state-of-the-art inforrna- tion infrastructure available to all. The large U.S. corporations, such as the Fortune 100 and other companies that have large distributed information needs, will always be able to maintain their own private networks and in- formation processing capabilities. Even these large organizations often find the cost of maintaining this capability too high without standards. The U.S. economy and culture have always championed the smaller, entrepreneurial company. In order to be competitive, virtually every organization must have access to information through modern information technology. Only a public network with publicly available advanced services can provide the needed access. At present, Japan and France have made electronic access to infor- mation technology a national priority. In the United States there is no organization providing a national vision toward universal access to infor- mation technology and information resources. The feeling of the workshop participants was that the concept of universal access to telephone service should now be expanded to include a much broader concept of universal access to information technology. If the standards for this infrastructure are international standards, then U.S. companies can compete domestically and in the larger markets. If we attempt to develop only U.S. standards, U.S. producers will have less incentive to produce for the larger world markets, while foreign competitors will still find the U.S. market large and worth fighting for. The consensus of the workshop participants was that the voluntary standards process is the best way to develop the framework on which the universal information infrastructure can be built. The present legal structure does not need to be greatly modified. The checks and balances in the standards developing process need not be the problem that makes the process inordinately slow; it is the lack of leadership from the top in government, in industry, and in the standards developing organizations. In the past it was possible to follow a less hectic pattern, in that the technology leader implemented a new capability, worked out the bugs, and then reached an agreement on an industry standard. ~day, there is no clear technology leader, and early standards are essential to avoid a chaotic market with products that are not interoperable. Some means must be found to develop a prototype of the results of standards deliberations to correct the inevitable hews and oversights. Information technology applications distributed geographically cannot always be checked out in the laboratory of a single developer. The objective is to develop functional

OCR for page 17
26 standards that will be longer lived than any specific hardware and software that implements the standard. A number of consortia have been organized in specific areas to de- velop and test standards before drafts are made final for forwarding to ANSI through an accredited body. Many interested groups would lee to participate in these prestandards developments, but they are unable (or unwilling) to bear the cost of ioinino and Participating in the consortium. ~ 0 r r - -A ~ RESPONSE TO THE CHALLENGE Based on the preceding identification of the challenges, a straw-man proposal for improving work on information technology standards was presented to five cross-industry discussion groups which were established on the second day of the workshop. The topic was presented as follows: A proposal has emerged to create an organization at the national level to address the following information technology issues: Provide a vision of the future supported by the executive levels of industry and government Establish an architecture and systems plan consistent with that vision Coordinate the work of already established standards working groups Put in place a management plan for initiating action early and achieving agreed upon goals Promote rapid prototyping and implementation of newly approved standards The groups were specifically asked: 1. Do you support this proposal in whole or in part? If so, why do you feel it is important? Should this organization be a new government agency, part of an existing agency, or a nonprofit corporation controlled by industry? Should funding come from government, from industry, or from both, perhaps by tax incentives? In the five group reports on this topic, some, but not all, of the el- ements of the proposal were supported or even elaborated upon. Most fundamental was the consensus view that the task of providing a view of the future should not be combined with that of managing the supporting standards activities. The need for a vision encompassing all of the in- formation technology issues, those related to computers, as well as those related to communications was considered essential to the future health of U.S. industry. This vision should provide "moon-shot-like" inspiration

OCR for page 17
27 to achieve the great information technology network essential to achieving other national objectives, such as better trade and educational performance. The impact of the vision should be a "jump start" in raising awareness of the importance of information technology to economic growth. There was agreement that the United States lacks the vision of a fixture information technology u~frastructure that is needed to drive standards activities. It was not at all clear how the evolution of a national vision for infor- mation technologies could be best accomplished. However, the workshop participants supported some form of government-industry partnership, with government providing an environment within which industry and govern- ment leaders could create and execute a consensus plan. Ad hoc advisory groups have been chartered by the federal government to attack similar problems in the past. In this particular case, the vision and overall architec- ture of a national information network could be the product of a blue-ribbon commission with both the vision and influence to get results. This might well indude the senior executive officers of major information technology vendors and users. The Office of Science and Technology Policy within the Executive Office of the President is considered the most appropriate government agency to convene the advisory group. The Director of the Office of Science and Technology Policy should arrange for an ad hoc blue-ribbon study by top-level people from in- dustry and government to establish a vision and high-level architecture for an infonnafion technology infrastructure. The problems of reaching and implementing agreements on a national information technology infrastructure vision are formidable. The companies coming to the table have competing goals, and it is counter to American culture for competition to cooperate. Narrow parochial goals must be set aside to achieve a common future in which all players have a better chance of prospering. In some ways, cooperation on information technology standards also runs counter to the deregulatory philosophy that the market should decide what is best. That can be an inefficient way of selecting a standard and can result in very marginal technical advantages. It is far better to find a win-win situation among contenders, if that is at all possible. The exalting public standards fomms are experienced in dealing with these issues. While their performance in moving standards forward has been less than that which is desired, there was no call by the workshop participants for establishing a new agency for dealing with standards man- agement issues. In particular, it was felt that ANSI could be equal to the task if it were to have a broadened charter and better representation by the information technology industries (both users and suppliers) at their Board of Directors level. The workshop recommended that:

OCR for page 17
28 . The American National Standards Inside should strengthen its role in info~ion networking by the addition of appro- priate to~level industry representatives to its management. It should extend its influence through the life cycle of stan- dards by proactively planning earlier action, streamlining the process, and supporting implemeintadon agreements and prototyping. It should strengthen its role in representing US. needs for inte~na~nal standards. The new or revised organization would have to stimulate groups to develop standards where they are needed for the overall infrastructure of U.S. information technology. When standards lead products, this or some other organization would need to develop prototype systems to verify the standards before commercial organizations would risk the development of usable products. The discussion groups were asked to prioritize the possible approaches to accelerating the implementation of standards at the national and inter- national levels. The order was as follows: Top level management support and pressure for agreement Starting the right track early Life-cycle planning with phased introduction Streamlined approval process Rapid prototyping and problem resolutions Reduced fear of antitrust suits All discussion groups gave the highest ranking to top level management support of the standards process. The subject matter experts who labor conscientiously in standards developing committees must understand that in many instances it is far more important to reach an ear) agreement than for a particular technical view to prevail. Clear communication of strategic priorities from the top to the bottom of industry hierarchies is required. A national vision and a revitalized ANSI can certainly help, but in the end it is incumbent upon each industry player to accept this responsibility. Close behind top level management support in importance were the life-cycle planning of phased standards introduction and a streamlining of the approval process. Perhaps these two improvements can be accomplished hand in hand. It should be easier to reach quick agreement on a standard that is viewed as a stop along the way, rather than as a product never to be changed. The prototyping issue was very important to users. Since the user group was less than a quarter of the participants, this topic did not rank as high as it might in other forums. Reducing the fear of antitrust suits was either unranked or ranked near

OCR for page 17
29 the bottom by each group. The workshop participants, while recognizing that fear of antitrust prosecution may exist, were not persuaded that this fear was justified. No one was able to cite instances of unwarranted prosecution. Of course, that could be the result of everyone being overl5r cautious and, hence, less effective. It is clear that a more convincing case must be made if any change in government policy is expected.

OCR for page 17