Washington, D.C.
October 18-20, 1995
Bruce W. McConnell, Chief, Information Policy and Technology
Office of Information and Regulatory Affairs
U.S. Office of Management and Budget
Electronic information transforms relationships among the traditional producers and distributors of information, creating new opportunities and new challenges, especially for the long-term sustainability of information. The most prevalent popular metaphor, the information highway, is somewhat misleading. A new model — information ecology — can bring new thinking to bear on these challenges. Exploring this model suggests that current government information dissemination institutions and practices must jointly transform themselves if access to publicly valuable information is to be maintained and expanded.
Electronic information presents new opportunities, but also new challenges, with respect to keeping the public informed about government activities and providing and maintaining economically important government information.
Information dissemination requires a combination of activities, including creation, editing, production, marketing, and distribution. Electronically-networked information is transforming the relationship among these activities. In the paper world, the economics of manufacturing technology binds the individual creator to a publisher, who is responsible for editing, production, marketing, and distribution. Information technology breaks that bond, creates different roles, and broadens participation.
Networked information technology today supports electronic mail, file transfer, and the services of the World Wide Web. The transforming effect of these services on organizations and work processes, widely documented in anecdote, remains poorly understood from a systemic perspective. In the sphere of information dissemination, the immediate impacts of the technology are to open up the creative process, permit the disaggregation of post-creation value-added, and flatten the distribution structure.
Creation is opened up by the ability to share preliminary work easily, to draw on a broad range of resources, both human and electronic, and to iterate the product quickly. As a work proceeds from creation through the activities presently associated with publishing, creators can choose to do more themselves, or to engage a variety of new collaborators. Posting one’s work to a moderated newsgroup or maintaining one’s own website are viable alternatives to traditional publication. The more specialized of these activities (e.g., representing the author in intellectual property negotiations) may continue to benefit from economies of scale or scope in continued aggregation.
But the effect on distribution is the great power of the technology, and the one we least understand. Final production and distribution of “copies” become extremely inexpensive. The technology’s support for interactivity breaks the mold of one-to-many publishing. As one industry leader envisions, “If there is a market for 500 channels, imagine the market for five million, 50 million, 500 million!” But this proliferation of sources and outlets raises issues about the present and continued quality of information.
Once created and distributed, information products must be sustained if they are to retain their value. Sustaining includes maintaining the information’s currency, availability, location, and integrity. Currency means keeping it up to date. Currency is complemented by availability — the retention and preservation of information for future users. Both require the ability to locate the information and to validate its integrity.
The present system works well for availability and integrity, fairly well for location, and poorly for currency. Libraries serve as public archives to retain and preserve published works, though increasingly challenged by the costs of cataloguing and shelving growing collections and by the problem of deteriorating acid-based paper in works produced since 1850. Regulations govern the retention and preservation of business and governmental records. Paper is intrinsically hardened against tampering. It can easily be imbued with evidence of its authenticity, attested to by bindings, seals, or signatures. But, rapidly changing information is soon out of date, and updating is resource-intensive. Finally, locating paper information is a mixed bag. Only if it has been catalogued can it be found, and which catalog to search remains an issue. Here the technology is assisting — for example, the Government Information Locator Service (GILS) provides an on-line catalog of government paper and electronic information products.
While information technology can improve location and currency, present automated tools do not support continuing availability and integrity very well. Search tools make information easier to find. The ease of replacing out of date information with the latest version is a double-edged sword. It is not always obvious where one goes to find the previous version. As for availability, technological obsolescence in this rapidly changing industry makes acid-based paper seem like a simple problem. Technology can increase integrity — for example, by making peer review easier and more effective. Ultimately, the integrity of electronic works will be best assured by getting them directly from reputable sources. For the most important documents, the so-called digital signature will provide ironclad assurance. But the institutional infrastructure to support digital signature services, like many websites, is still “under construction.”
Meeting the challenges created by networked information will require new ideas. Often when looking ahead into the unknown it is useful to employ metaphor as an analytical tool, permitting us to reason from the familiar.
In the present context, the dominant metaphor is overripe. “Under construction,” “the rules of the road,” “the highway patrol,” and other derivative images have served their purpose fully, at least for serious thinking. The image, beneficially concrete, compelling to a nation of motorists, will remain, perhaps forever, in the popular mind. Its grossest liability is the impression it gives that government will fund the highway’s development. It intimidates by evoking speed, dirt, noise, and danger. But it misleads in more subtle ways.
The highway metaphor emphasizes the auto-mobility of users. It evokes no common journey. We each travel to our own destination. Television may be the analog of mass transit, but a parking lot is no community.
Networks are concurrent. Collaboration among individual creators and senders does not flow from the metaphor. Information is local. Creators have always been local, but, today, even faxes do not send the original. On-line, physical location is no longer meaningful. And off-line resources aren’t on the highway, yet they remain the most important information sources for many of us. A more organic image is needed to assist in thinking about information management.
As much as those of us in the business think highly of information, its value pales in comparison with the object of ecology, life itself. Approached carefully, however, “information ecology” instructs. It is not a new idea. For example, John Perry Barlow argues, “Information is a life form.” Information forms patterns that “evolve to fill the empty niches of their local environments . . . the surrounding belief systems and cultures of their hosts, namely, us.”
The thesis of this article is that information is most usefully created and sustained in its own niche, connected and interdependent with other information. Cybernetics considers natural systems to be "wholes whose specific structures arise from the interactions and interdependence of their parts . . . and the nature of the whole is always different from the mere sum of its parts." Following this thought, “deep ecology” asserts that the world is not a “collection of isolated objects but rather . . . a network of phenomena that are fundamentally interconnected.” Of course, natural ecology is far better understood and more mature as a discipline than information ecology.
Nevertheless, thinking about information as an element of an ecosystem immediately enables us to draw upon the importance of diversity in living systems. There is a redundancy in healthy natural ecosystems that permits survival under stress. Likewise, healthy information systems exhibit redundant pathways and resources, so that there is no single point of failure. Resources reside in multiple locations and are sustained there.
A second implication of information ecology is complementary information. Like symbiotic organisms, many kinds of information are lifeless without other, complementary kinds. Raw government economic statistics, for example, may be nourishing tidbits, but without the analytical engines of academe and business to adduce trends and provide context, they make for slim pickings. Conversely, the derivative works of business and academe depend on government information as raw material. Moreover, the analysis and conclusions contained in those works inform government as to the most appropriate directions to take its products (and, in the best case, policies) in the future.
Finally, information ecology permits us at least to consider the “problem of the commons” that plagues our information space. Spamming and piracy pollute the system. The challenges we face in managing and sustaining “public good” information cry for attention. Though a light hand is needed given the complexity and culture of the Net, we are today reminded of Odysseus; though the Sirens sound, truly No-Man manages the infosphere.
How then should the information commons be managed? More accurately, given our insight into its eco-nature, what is the best way of guiding it, together? Three tenets may serve: distribute responsibility, cultivate intermediaries, and use the technology.
Responsibility for managing the infosphere must begin at the information. No single mind or machine can comprehend the universe of knowledge. Instead, centers of excellence and expertise — subject-oriented, client-oriented, regional — should reside in an environment that lets them build on their strengths and fill in their places in the network. For this method to work, standards may be needed — standards of diligence, quality, and presentation. Community standards have long existed for what newspapers look like — no ads on the front page, for instance. But as the editor of the Washington Post’s on-line service “Digital Ink” observed recently, no such standards exist for electronic newspapers, and the Post’s marketing strategy involves ready links to advertisers relevant to particular news stories. The government faces the same issue — proliferating home pages (several hundred at last count) and no standards of appearance or content. (OMB is working on this.)
Distributed responsibility can sustain information. In a thorough and ground-breaking analysis, a broad-based task force of information professionals has examined the concept of a national system of digital archives. The proposal envisions a distributed storage environment based in centers of excellence. Each archive would be supported by contractual arrangements with suppliers, so that even these distributed centers would not physically (or even electronically) store all the works for which they were responsible. The model compares favorably from a cost standpoint with depository libraries. Each archive would meet or exceed standards and criteria set by an independently-administered program for archival certification.
Intermediaries continue to matter. San Francisco’s librarian Carl Fleishauer once remarked to the effect that libraries need to shift from building collections to making connections. Of course, librarians have always been the connection between the public and the vast stores of information in their care. This role will only be heightened as information continues to explode into the Net. The same model applies in the for-profit world. Esther Dyson argues that successful information suppliers will focus not on content, but on relationships — “selecting, classifying, rating, interpreting, and customizing content for specific customer needs.”
Finally, the technology contains within it the seeds of its own management. The scientific research community, the earliest adopter of the Internet, is now attacking large information management problems collectively. As part of the massive Human Genome project, GenBank, a public, electronic repository, lets scientists compare new genetic sequence data against its Web-resident database of known genes. Receiving 12,500 hits a day, it is faster and easier than e-mail, and more up to date than any traditional source. Academic journals are following the trend — by economic necessity. Several purely electronic journals already exist in their own right.
A diffuse, distributed, locally-organized crop of information, pushed by the technology and supported by the metaphor, is sprouting up amidst the tended rows of traditional information products. It is producing an environment of challenges and opportunities for government information dissemination.
The metaphor and, more fundamentally, the underlying technological changes that have created the new wine of electronic information, have significant implications for the institutions, practices, and future plans of those associated with the creation, maintenance, and dissemination of government information. The time is critical.
The information explosion afflicts government information managers and users. The challenges of sustaining information in a networked environment — maintaining its availability, currency, location, and integrity — must be addressed thoughtfully and forcefully. Yet the centralized models are broken. Many firms once successful at manufacturing vacuum tubes remain viable today, but none is in the semiconductor business.
Under a new model, the government’s centralized publishers — notably the Government Printing Office and the National Technical Information Service — would adapt to the new opportunities technology provides creators. They would shift their focus from access and production to building customized relationships with information suppliers and users. Similarly, Depository Libraries would adopt a new role, adding value by providing knowledgeable intermediation as regional or subject matter centers of excellence. Federal agencies would take responsibility for sustaining their own information. And the central management agencies would promote the development of quality standards to support these changes.
Operationally, of course, resources are scarce. Agencies are under direction to focus government resources where there is comparative advantage, that is, to core mission activities. In today’s budgetary environment, anything that appears to be “overhead” is suspect. Many of the costs of information maintenance appear to be overhead at first glance, because of their indirect relationship to the core mission. In any ecosystem, as species die out, the food chain remains viable for a while. Eventually, however, the whole system becomes vulnerable to stress, and the richness and sustainability of the landscape suffer.
The decentralized, distributed approach suggested here will work best on a cooperative basis — building symbiotic relationships to recognize and deal with the real challenges. (Other scenarios — e.g., competition among institutions vying to maintain their traditional “turf” — can get to the same place, but more slowly and painfully. Two elements are needed: partnership and leadership.
Partnership among the stakeholders — government publishers and creators from all branches and levels, industry, the universities, the libraries, and users generally — means sharing responsibility, costs and risks. To work, it must be true sharing, not merely cost-shifting. A not-so trivial example is the shift to local printing for on-line documents. A better understanding of the costs of doing business in this new way is needed, for electronic storage can also reduce the cost of shelf space. A basis for beginning the partnership may be discussions that acknowledge the importance of complementary information — focusing on each party’s comparative strength in making more information more available and more affordable to a wider public.
While operational costs are an issue, capital costs are less so. Experience with information systems shows clearly that the effective capital costs are overestimated when compared with the ongoing costs of operations. The fundamental infrastructure build-out will happen — fiber will go to schools and libraries, computers will find their way into the classroom — through complementary public-private efforts. The real obstacle will be training users and intermediaries. In this area, one model is TechCorp, a national, nonprofit organization of technology volunteers, funded by the business community, helping to improve K-12 education at the grass roots through the effective integration of technology into the learning environment. Volunteers provide planning, technical support and advice, staff training, and mentoring inside and outside the classroom.
Leadership is needed in two areas — modeling cooperation and policy. Discussions about the future of the Depository Library program, which were initiated by action of the Senate Appropriations Committee, can be a first step in establishing a more cooperative relationship among the various stakeholders in Federal information. A more formal arrangement may be ultimately needed.
On the policy side, history teaches that change will occur faster than we expect. The frustration with policy is that it usually addresses today’s problem. Fortunately, the Paperwork Reduction Act of 1995 (P.L.104-13, 44 U.S.C. Chapter 35) provides a strong framework for policy development. That Act recognizes the synergy that comes from a diversity of public and private information sources, and the importance of making underlying data available on an unrestricted basis in addition to providing information “by the drink.” Undergirding the framework is the Act’s pricing policy — that user fees for government information may not exceed the cost of dissemination.
Similarly, the Federal Records Act (44 U.S.C. Chapters 21, 29, and 33) is fundamentally sound. It creates a framework that permits agencies to manage and sustain information of record. It recognizes that not all information is created equal in terms of its preservation value. It generally empowers the agencies which are most familiar with the information to invoke their understanding of the need to preserve their unique holdings, such as a scientific paper or data set of dubious apparent value today but potentially priceless in the future.
Yet, in both areas, implementing policy is needed to handle emerging issues. Only recently did the Federal Government issue final rules governing the disposition of electronic mail. Dissemination policy must continue to play catch up with the technology.
The rest of Title 44 needs a comprehensive look. Its printing provisions date back to a time when economies of scale in production dictated a large centralized printing plant to serve all of the government. They reflect a time when government information existed solely in paper form, necessitating a centralized distribution system where printed publications would be assembled and shipped to libraries nationwide. Yet reform should avoid the tendency to piecemeal solutions, and commit instead to comprehensive review and restructuring.
We stand at a crossroads in information policy. As we cultivate the metaphors and institutions of the future, let us be mindful to sustain the truly important.
MR. KOBULNICKY (University of Connecticut): I just wanted to ask Bruce what he had in mind when he was talking about partnerships with libraries and universities. We, too, are interested in partnerships, but to go back to your own quote, it is “ultimately a question of resources.” One of the things that we are very aware of is that government likes to build partnerships with universities in order to cost-shift, and we cannot do that. There have to be real partnerships.
MR. McCONNELL: That is a fair point, because in many of the ways that we use the technology we are cost-shifting. We are shifting our printing costs, for example, on a basic level. We may be shifting some of our searching costs and analytical costs, too, so you have a valid point.
My response, then, is that a partnership discussion needs to focus on comparative advantages, but I agree that it is unfair for government to try to use partnerships solely to shift costs, and, instead, the partnerships need to be a way of sharing costs in a mutually supportive manner.
MS. MERRILL-OLDHAM: Jan Merrill-Oldham, University of Connecticut and Harvard, and consultant to ARL. I have a question for Martha Riche. Can you talk a little bit about the process of identifying things that you no longer need to do? How does that take place in a positive environment?
DR. RICHE: Where is that positive environment? It goes back to the question your colleague asked of Governor Carlin on how to set priorities. We are setting priorities on the basis of what only we can do versus what others can do. That is our first cut: looking at what we don’t do anymore, and what we can stop doing.
That goes back to what I spoke of earlier, the analytical printed reports that we very much enjoyed putting out, but that others can do. I’ll give you an example of what I mean by “others.” I myself did a report when I was at the Population Reference Bureau in the 1990 census. Resources had been cut for printed reports, and the Census Bureau did not do the tabulation it had planned to do with a related report of data. The Census tabulated according to women, so I went to the Ford Foundation, got a grant, did that tabulation privately, and compiled a report with data that otherwise would have been only available for researchers on computer tapes and not to anybody else.
We also have a state data center network, 1,800 organizations throughout the country and in every state. These organizations are finding new roles for themselves, too, and one of them is going to be precisely packaging data for local communities, something that we simply couldn’t do ourselves.
The Constitution grants us a special role in collecting data; that is to say, Congress can mandate that we collect data from you and that you respond, and it also mandates that we keep the data confidential, which means that we don’t even share our address lists with other statistical agencies. So there are many things that only we can do that no one else can do. That is where we are focusing our efforts. In other words, only we can collect and process the data; but once we get it out the door in a way that guards privacy and confidentiality, others can use that information.
The next step is deciding how to prioritize the issues. That is a process we are now undertaking. I have asked my staff to go through and reorganize our budget in terms of deliverables. What are we doing with the money that the American public, the taxpayer, gives us? What are our specific programs, what kind of a universe do they cover, and what kind of products come out of them? We need to organize that whole array into what overall knowledge we will give to the country.
Some knowledge tells us where we have been and where we are now. Some of it tells us how we got there; some of it tells us where we are going to go. We need to look at the balance of things we have in a very specific way, such as by looking at our economic data. I indicated to you that half of our work is on economic data. Are we measuring industries of the past too much? Are we measuring industries of the future enough? We are calibrating around those kinds of questions.
It is very difficult, because, as you can imagine, there are tradeoffs all over the place. The particularly difficult tradeoff is the one between the past and the future. People who are accustomed to getting data about something that is very important to them get very upset if you say you’re not going to do it anymore, or if you are going to do it less frequently. But there is no one who really speaks up for the future. We haven’t measured growing industries very well because they don’t have a constituency, which, however, may be because we haven’t measured them.
So that is our task, and we are taking advice from every place we can get it. We have help from four advisory committees from the professional associations that represent our areas of study: economics, marketing, statistics, and demography, and from a broad array of other advisory committees, as well. So that is how we are looking at things.
MR. WILLIAMS (University of Colorado): Martha, would you explain again what your relationship is, if any, with the Government Printing Office (GPO) and NARA?
DR. RICHE: I went down to the Archives early on in my tenure, and they showed me some wonderful documents and gave me copies of them, and told me that census data is definitely their most popular product.
They showed me people sitting at terminals consulting census records, and they told me about people who line up the night before we are about to release a new census. We provide the Archives with records almost every day that they will then archive.
GPO also prints and sells many of our publications. We are not in a situation where we do our own printing, and some of their best are our publications, particularly The Statistical Abstract.
On the other hand, we are now moving our publication electronically, and not just on the Internet, but also as CD-ROMs, probably our most successful product. We do those ourselves, and what it enables us to do is to give everybody access to data that people before only had access to if they had a mainframe computer and someone capable of programming it for them, because, naturally, our printed reports were just the tip of the iceberg. We are now able to release the data to just about everybody in the form of CD-ROMs. It’s very exciting.
MR. WILLIAMS: Are you archiving the data on the CD-ROMs?
DR. RICHE: Well, the problem of the archiving is an issue I referred to earlier. It is still very much up in the air as to who’s doing what, and that is what Bruce was speaking of when he said that we are narrowing and tightening our missions. As I said, we are looking for leadership in that area.
MS. BAKER (Washington University at St. Louis): This is a question for either or both of the speakers. You are describing what sounds like massive change in your organizations, and those of us who work in relatively loosely structured private institutions find change hard, but compared to what you are doing, our changes are probably a piece of cake. Can you talk a little bit about making change in organizations that are structured and perhaps rigid?
DR. RICHE: It is very difficult. Unfortunately, coming from the private sector but having started off my career in the public sector, I knew what change looks like to employees, so I started off by having a series of meetings with my employees to tell them how different this would be, to describe why it’s different, and to point out the positive things about change for them.
For example, I pointed out that this is the information age, and we are the nation’s first information organization. We have been in the information business since 1790. So this is our time, and although we do have serious competition from the private sector for survey work, I pointed out to my employees that this is good; it means that they have other places to go to get jobs. That isn’t the way they are used to thinking.
Management information systems are another useful way to make change happen, as a type of open-book management that allows them to see for themselves our expenses, how many resources they are taking up, and what we need to get back from them, so they can monitor themselves and manage it. It is very refreshing for many of them to see that. It’s a whole new way of being, and some of them are enjoying more than others. I’m sure that is what you have found, too. The bottom line, though, is that reality therapy works. Thank you.
MR. McCONNELL: I can comment on the question from the perspective of a career bureaucrat, because, in fact, this administration is making a real difference as far as how things are happening in the organizations. It’s done in just the same way Martha suggested, by first involving the workers, the people who are going to be changed in the process. That’s what the National Performance Review has been all about.
It is very difficult to make changes in large organizations, and in the government, of course, it is very difficult to make changes stick because the leadership turns over, but, particularly with respect to the kinds of issues we have been talking about this afternoon, and regarding the general question of making change in the bureaucracy, this administration has gotten off to an incredible start. It just remains to be seen whether they can sustain and follow through on it, but it is at least worth considering.
MR. NEAL: I must bring this program to a close by thanking our speaker, and also by giving special thanks to Governor Carlin and his staff for hosting us in this outstanding new facility.