, The Evergreen State College, Olympia, Washington
, University at Albany, SUNY
Pages: pp. 26-32
Abstract—As they focus on the challenges that those who implement digital government face, computer science researchers practice nearly the entire spectrum of their discipline and work in collaboration with scientists from other disciplines in pursuit of answers to questions about information management, policy, and technology in government.
Digital government is the use of information technology to support government operations, engage citizens, and provide government services. As a result, digital government projects can, and do, involve a broad range of computer and information technology.
Digital government research, on the other hand, applies computer, information, and social science methods in the investigation of the information-related needs, management, and policies of government or the information-related characteristics of a democratic society. 1 For computer science researchers, this might mean redefining others' research to fit digital government challenges or conducting original research to solve a problem that is not as critical to other, mainstream, domains.
What distinguishes digital government research from research in its primary disciplines of computer and social science is that those who practice it engage in active partnerships with government agencies. Further, they often collaborate closely with scientists from other disciplines to examine interactions between technical and social processes in the context of governmental organizations, practices, and processes. In other words, computer science research in this domain differs from other computer science research in that it always involves a stakeholder who will want to implement a system, and it often involves formal studies of technology assessment and adoption.
Digital government research is thus, in all cases, multidisciplinary. At its best, it draws together information science and technology, computer science and engineering research and development, and the social sciences. These disciplines in various combinations and permutations provide the foundation for explorations in the digital government realm.
In the nearly five years since a special issue of Computer first highlighted digital government projects, the field has come into its own, drawing research funding from the National Science Foundation and elsewhere and yielding successful, mainstream information technology and social science research results ripe for integration into real government systems.
Further, a now well-established annual international digital government conference draws hundreds of university researchers and government collaborators to present research and development, discuss challenges of working in the area, and highlight new focus areas. Recent efforts to launch a digital government society provide further evidence of the growing interest and investment in this area and the continued interest in bringing together a new digital government community of interest.
Two of the accompanying sidebars describe efforts to improve collaboration by integrating data across government agencies. In " An IT View of Emergency Management," José H. Canós, Marcos R.S. Borges, and Gustavo Alonso provide an example of how modern technology can help improve the management of emergency situations by dispatching and tracking responders. " Public Safety and Cross-Boundary Data Sharing: Lessons from the CapWIN Project," by Christine B. Williams, Janis L. Gogan, and Jane Fedorowicz, describes a multistate transportation and public safety integrated wireless first-responder network designed to "enable data interoperability for first responders wherever they are."
That making a law is only the first—and perhaps the easiest—step toward bringing about compliance under the spirit of that law presents another significant challenge for information technology. The " In the Real World of Digital Government: Successes and Challenges of E-Rulemaking" sidebar by Neil Eisner describes how the US Department of Transportation uses modern information technology to develop new regulations and monitor compliance with them. He also outlines a vision for the future and discusses the challenges inherent in bringing about that vision.
Finally, in the " Research Issues in Healthcare Informatics" sidebar, Sylvia J. Spengler presents a dream for future healthcare informatics. Significant progress has been made in medical research and hospital information systems, but many factors, including computing and health technology, national security and disease threats, and healthcare regulations, continue to create new challenges for hospitals, patient care, and public health—challenges that new research into policy, management, and technology could shed light on.
The National Science Foundation (NSF) has identified emerging areas such as healthcare and natural resource management and is working proactively to alert computer scientists to new research and development opportunities in digital government, such as those for natural resource management ( www.evergreen.edu/bdei).
The five articles in this issue exemplify the range of perspectives, theories, and approaches that draw from the many disciplines of digital government research and are applied to the wide range of issues that digital government practitioners face.
These articles, selected from the best of the presentations at the 2005 National Conference on Digital Government Research ( dgrc.org/dgo2005/), provide a frame of reference for readers interested in learning more about the interdisciplinary research aimed at producing successful applications of information technology within government. The articles exemplify the focus of digital government research, that is, to identify the unique problems of government organizations, identify or develop relevant technologies, and explore interactions among those technologies and the social processes of government.
Digital government research is still in an embryonic stage; researchers from the many involved disciplines seek definitions and frameworks to consistently and effectively frame and guide research. Questions abound regarding the relationships among the various disciplines. In "Digital Government Research in Academia," dg.o 2005 conference chairs Lois Delcambre and Genevieve Giuliano shed light on these questions by discussing the future of digital government in academia, providing a means for readers to gain an understanding of the place of digital government research in universities.
In the seven years since the NSF began funding digital government research, every funded project has been expected to "enable the generation and use of a continuous stream of advanced information technologies for early adoption and integration in the federal information systems community."
The question is, how will US university researchers, who have long prided themselves on theoretical and conceptual findings of depth within a discipline, respond to this vision of digital government research? The authors grapple with this question and present a vision of future digital government research. Asking if digital government research is indeed a new discipline and if it is achieving recognition within the academy, they cite the challenges of pursuing interdisciplinary research in academia and identify the specific challenges and potential rewards for digital government researchers.
While a major thrust of e-rulemaking applications, as envisioned by Neil Eisner, lies in collating data from citizen commentary over the Internet, two articles in this issue address how researchers could more effectively use massive, already existing, government data collections.
In "Data Alignment and Integration," Patrick Pantel, Andrew Philpot, and Eduard Hovy highlight a partnership among information scientists and government agencies that was created to help make data collected by government agencies more valuable both within the agencies that provide the data and to the citizens who paid for it. In this work, information theory is applied to readily available data to create "new" knowledge of use to government.
In contrast, in "Accessing Government Statistical Information," Gary Marchionini, Stephanie W. Haas, Junliang Zhang, and Jonathan Elsas describe how ordinary citizens might in the future master "finding and understanding government statistical information"—data that hitherto could only be readily processed by those with sophisticated statistical and data processing expertise. Applying both new human-computer interaction research and a better understanding of organizing metadata facilitates these potential innovations.
In "Building Community Information Systems: The Connected Kids Case," Teresa M. Harrison, James P. Zappen, and Sibel Adali describe how government agencies could effectively use new technology to help communities build cohesion and collaborate with government developers to build systems that help citizens on both sides of the digital divide, thus enhancing the democratic process.
"Analyzing Government Regulations Using Structural and Domain Information" by Gloria T. Lau, Kincho H. Law, and Gio Wiederhold provides an example of how university researchers are addressing e-rulemaking challenges. The authors' Regnet project aims to solve problems that arise when agencies or citizens try to determine the "correct" action in the face of multiple government regulatory documents. Regnet also considers e-rulemaking scenarios using public commentary on drafted regulations.
The articles in this issue paint a landscape of current digital government research, answering questions about data, interfaces, communities, and legislation. These articles, together with the sidebars highlighting current initiatives and future visions, provide an understanding both of key questions facing digital government researchers and of how digital government research applies to recognized problems and intriguing questions that new technology enables us to consider. The " Digital Government Research Resources" sidebar provides additional information about key digital government research and researchers.
Several recent incidents involving loss of life and property have increased public awareness of safety and security issues. This heightened awareness has made efficient emergency management a priority for many governments and organizations. Most such efforts revolve around the emergency response, 1 the most critical phase of the emergency management cycle: prevention, response, investigation (PRI).
An emergency plan (EP) provides guidelines for making decisions promptly when a critical emergency occurs. The consistency and completeness of the information the EP contains, as well as the speed of access to it, are key factors in the emergency response.The MetroValencia's Hypermedia Emergency Plan
In 1998, the Security Office of MetroValencia and a group at the Technical University of Valencia started a pioneer project to improve the response to emergencies in the subway transportation system in Valencia, Spain. 2 The project's main goal was to reduce the response time to all types of emergencies, with special emphasis on fires in tunnels or underground stations.
As a result of this project, a hypermedia-based system replaced the existing EP—a document with more than 400 pages, including text, maps, and pictures. As Figure A shows, using multimedia let planners include 3D models and video and audio clips. Hyperlinks gave the system the flexibility to provide information to the user in the appropriate order according to how an emergency develops. A touch screen facilitated interaction with the system so that users could find relevant information faster and more accurately.
Figure A Screenshot taken during the evacuation of passengers being moved along a tunnel toward a subway station. The screen shows textual descriptions for actions to be performed, a surface map of the system, and a video recording of the subway tunnel.
The Hypermedia Emergency Plan was a significant improvement but still had only limited functionality in terms of other important aspects of the emergency response and, more significantly, emergency management as a whole. Building emergency management systems has been our main research focus since completing the project.From Emergency Plans to Emergency Management Systems
Focusing on emergency response hides aspects that are important from a global emergency management perspective. In our view, systems supporting the PRI emergency management cycle must cope with collaboration, awareness, coordination, and efficient access to information.
An emergency management system should support team decision making in such a way that each team member is aware of what actions to take—leaving room for improvisation due to the incidents' unpredictable nature—and also of the actions taken by other team members. All the information needed to execute an action must be available through personalized hypermedia interfaces. A digital library, audio and video streams, or sensor networks can provide the interface's digital objects. Every decision, as well as the interactions between team members that generate it, must be recorded for both logging and analysis.
During our joint research, we have identified several dimensions that characterize emergency management, including coordination, information management and retrieval, presentation, communication, collaboration, and intelligence. 2 Some of these aspects can be dealt with using current technologies, while others require further improvement of current tools. For example, in most cases, using a workflow management system allows dispatching responders and tracking their actions as well as measuring response times, which can help to analyze the response's overall quality. The analysis of these factors can lead to new preventive measures that improve the response to future emergencies. 3ReferencesM.Turoff"Past and Future Emergency Response Information Systems,"Comm. ACM,Apr.2002,pp. 29-32.J.H.CanósG.AlonsoandJ.Jaén"A Multimedia Approach to the Efficient Implementation and Utilization of Emergency Plans,"IEEE MultiMedia,July-Sept.2004,pp. 106-110.V.B.Dinizet al.,"Knowledge Management Support for Collaborative Emergency Response,"Proc. 9th Int'l Conf. Computer-Supported Cooperative Work in Design,Coventry University,2005,pp. 1188-1193.
The events of September 11, 2001, and Hurricane Katrina painfully demonstrated the need for a coordinated public safety response effort. One such effort is the Capitol Wireless Integrated Network ( www.CapWIN.org), a partnership between Maryland, Virginia, the District of Columbia, and several federal government agencies. CapWIN represents one of the first multistate transportation and public safety integrated wireless first-responder networks in the US. The project's vision is to "enable data interoperability for first responders wherever they are." Current CapWIN functionality includes
Planning for deployment of a pilot project for mobile in-vehicle systems began in June 2000. Beta-test user training began in fall 2003, and live tests followed.
As of July 2005, 32 agencies and 723 users had enrolled, with active participation during that month by 160 users across 22 agencies. CapWIN currently supports events in the US Capital region, such as September 11 memorial activities.
A cooperative agreement between CapWIN and the Automated Regional Justice Information System (ARJIS), a San Diego-based criminal justice enterprise network, will facilitate collaborative development of voice and data-sharing systems.
E-government collaborations such as CapWIN are affected by technical, financial, political, and organizational factors within the environment, participating organizations, and the interorganizational system itself. Our study shows that the environment was both a critical catalyst for and a shaper of the CapWIN initiative.
Formation of the CapWIN network was precipitated by a 1998 incident in which a man threatened to jump from the Woodrow Wilson Bridge across the Potomac River. This incident caused an enormous traffic tie-up because responders from multiple jurisdictions could not communicate with each other to share information and coordinate their activities. Subsequently, the 9/11 attacks reordered economic priorities, regulations, and funding around the US president's homeland security agenda, which affected public safety mandates for federal agencies.
At the organization level, CapWIN participants and sponsors represent multiple jurisdictions and functional areas, with varied strategic objectives, budgets, technology, and operations. Major integration challenges arose because of wide variation in their IT readiness factors.
Specific interorganizational system design and implementation decisions also had an impact. Logistical and physical attributes, such as terrain, led to the decision to use mobile wireless technology and an Internet-hosted communications bridge.
Built on an open, standards-based architecture, CapWIN uses COTS software where possible. Like many e-government collaborations, the participants had substantial prior investments in legacy systems that required a workaround or design accommodation. In the CapWIN case, some applications run side by side in a "sidecar," with data and interface integration accomplished via conversion to and from XML.
Another critical challenge for e-government collaborations is developing effective governance over the sharing and use of key data and resources. CapWIN had skilled project leaders who were successful in brokering formal relationships among participating agencies and jurisdictions.
CapWIN is one of several multiorganizational case studies of public- and private-sector interorganizational systems in the Bentley College Invision Project ( www.bentleyinvision.org). The project received funding from the IBM Center for the Business of Government and is also a part of a larger study of public safety response infrastructures funded by the National Science Foundation Digital Government Program.
Comparison of CapWIN with other Invision Project cases continues to uncover common elements that highlight the special challenges and advantages inherent in e-government collaborations.
Rulemaking is the process of transforming legislation into specific, legally binding rules, and e-rulemaking uses electronic technology to facilitate this process ( http://diggov.org/news/stories/2005/0505/0505_eisner_heymanreed.jsp). The government currently uses electronic technology in all aspects of the process, from researching the need for a rule, to managing the process, to providing opportunities for public participation, to letting those who violate rules pay penalties on the Internet.
Within the past 10 years, the US Department of Transportation (DOT) has developed intricate applications for managing the e-rulemaking process. Among other things, these applications allow electronic submission of public comment; list-serve notification of agency action; and electronic submission, coordination, and tracking of rulemaking documents. The department has even experimented with using chat rooms to solicit comments and loading chat transcripts into the department's Internet-accessible public docket.
As advanced and powerful as current systems are, however, the government needs more and better tools for assessing and responding to the millions of public comments it receives. Government employees still spend significant time reading and organizing comments, and they need better tracking systems, an interactive comment system that prompts commenters for more specific information, and an engine for summarizing the comments and the huge documents used to support changes in the rules. Additional needs include comment-gathering systems with front ends that respond better to commenter needs.
Particularly important is how a digital government can facilitate interactive communication. DOT, for example, encourages electronically submitted comments. People who fear that electronic formats are looked upon as form-letter junk should be reassured that attention is being paid. A mistake citizens frequently make, however, is not including enough detail in their comments. It isn't enough to say, "Don't do it!" A writer must give financial, environmental, or other significant reasons why the government should or should not undertake a particular project.
The key for the public is to provide good data that will help the agency make its decisions, and the key for future e-rulemaking is to help the public and the government do a better job of getting that data. What is needed is a real-time, interactive form on which a software program can scan a digital comment on a government Web site, picking up key phrases and prompting the writer for details. Thus, if someone were to write, "The new project will waste money," the software would respond with a phrase like, "Please tell us how it will waste money."
Research is also needed to develop quantitative assessments of digital interactivity. Some successes as a result of converting to Web-based systems have obvious metrics, such as reduced staff time and space allocation. But communication successes are often anecdotal, and it's not easy to tell whether the people who comment electronically would have commented as well by traditional means. For example, the DOT can't yet answer why some people send both electronic and hard copies of their comments.
Other needs, undoubtedly shared by many inside and outside the federal government, would be for software that automatically scans proposed rules for duplications or contradictions with other proposals or existing requirements, or a system that helps a citizen understand regulations before acting. An example of where such a system might have helped occurred when a dry cleaner shipped cleaning fluid in a Coke bottle. The corrosive chemical ate through the bottle and the box and, but for the vigilance of shipping agents, would have eaten through the hull of the aircraft on which it was to be shipped. When a DOT agency fined the dry cleaner for shipping the fluid improperly, he complained that he didn't know anything was wrong with how he shipped the fluid. Because the agency prefers to achieve compliance rather than collect fines, it must make it easier for people to find rules online that cover their activities.
The challenges facing the e-government research and development community at large can be summed up as the need for technical innovation and better user-oriented services within the constraints restricting all e-government projects. These constraints are smaller budgets, resistance to change within agencies, legal issues, difficulties identifying needs and desires for system improvements, and concerns about increased workload or loss of control due to increased access. One approach to meeting these challenges would partner agencies such as the DOT with academic researchers.
Practical suggestions for how digital government academics and government practitioners can best work together can be summed up as "Ask first." Researchers should check with agencies about their needs, and agencies in turn should check with citizens about their needs. When researchers work with people in government, they should not make assumptions without asking questions or seeking suggestions. For example, if government agencies are involved early enough in software development, making certain text computer-readable might only require a few symbols that agency personnel could easily add to documents as they create the text. On the other hand, if a researcher presents a fully finished program, the agency might have to hire dedicated data entry personnel. Using an "Ask first" approach might render a potentially useful prototype a first iteration of an evolving and useful application.
All healthcare is local, focused on an individual, but the information technology requirements of a system that addresses healthcare are at a minimum national, and they also have global implications.
The vision for healthcare information is to integrate data management and analysis to provide both machine-readable and human-readable records tailored to the type of user. The level of personalization for users will require interfaces that can learn preferences and adapt to the specialized vocabularies used by all who are involved in providing healthcare, from the student to the policymaker level.
Assuring privacy and confidentiality is a critical aspect of healthcare data management. This is often a major public concern that impacts decisions about the distribution of data, its integration, and perhaps its centralization, versus on-the-fly integration with permissions obtained only for specific instances.
Healthcare informatics must address the specific requirements of good data management, each with a generalized set of tools. Finding the data implies having the ability to search within and across distributed data sets. Accession requires the mediation of disparate media such as written records, test results, and image scans and their associated file formats. Interpreting the data assumes that both the metadata generated when the data is created and metadata that might be generated with later reinterpretation and updating of the data are appropriate.
In many instances, the data must be manipulated, for example, tracing the changes in blood sugar level, correlating with patient weight and blood pressure, and displaying the results. This is often accomplished using generic tools, but making these tools user-ready can be as much work as the original implementation.
Finally, the data must be stored for extended periods, raising issues of media fragility, federated versus centralized systems, storage-management tools, and a host of other concerns.
An integrated analysis environment for healthcare requires not only access to data but also to several additional components. Instrument and sensor input, particularly for patient monitoring for telemedicine; analysis tools that can help to meet the basic data requirements of searching, merging data, mining, and automatic annotation; and metadata acquisition are supplemented by access to collaborators and the literature. These factors expand both the vision of the system and the required research efforts.
Although these are standard issues for information integration and informatics across numerous domains, a true vision of healthcare informatics requires the kind of multifaceted approach that has been a hallmark of the NSF digital government program. First and foremost is the issue of human-computer interaction, addressing both citizen needs and professional interests. This will be critical for gaining acceptance of the technology by physicians, patients, and policymakers.
We are only beginning to understand the sociology and to develop the network analysis tools to make the most efficient use of the limited time for new technology adoption. None of this will happen without adequate training opportunities and research.
Digital government supports many of these research areas and brings the added benefit of formulating a prototype of the research or even putting it into practice. There is also a significant opportunity for international collaboration. It will be an exciting time.
Additional information about key digital government research and researchers, as well as conferences and publications that either focus on or include digital government research, can be found at the Web sites for three US centers that are recognized leaders in the field:
Center for Technology in Government
University at Albany, State University of New York
Digital Government Research Center
Information Sciences Institute at the University of Southern California and the Department of Computer Science at Columbia University
National Center for Digital Government
University of Massachusetts Amherst