Francesca Barrientos, Elizabeth Foughty
In 1450 the Ottoman Empire rejected Gutenberg's press. To this day, scholars argue over whether or not this event was the beginning of the end of an empire that had until then been at the technological forefront of civilization. Whether or not it actually caused the eventual decline, it almost certainly led to a 50-year stagnation. Gutenberg's press, meanwhile, went on to change society's relationship with the printed word, helping to fuel the Reformation and further the Renaissance. Web 2.0 and social media are today's printing presschanging society's relationship with computers and institutions, accelerating the spread of ideas, and influencing a collective determination of what's relevant and what's not. For that reason alone, government can ill afford to ignore this movement.
To be sure, government agencies have made forays into the Web 2.0 and the social media world. The EPA's Facebook network has more than 750 members, and the Library of Congress shares historical photos on Flickr. During the recent space shuttle mission to the Hubble telescope, astronaut Mike Massimino became the first person to Twitter from orbit . More than 33 agencies have blogs . But efforts such as these are largely meant to cultivate good will with the public. These are conservative stepsa use of technology in which the government still controls the message.
The new administration has called on government agencies to create a more transparent and participatory government , and Web 2.0 is a part of that change. Yet government, in general, doesn't yet know how to implement collaborative Web applications where the public becomes an equal contributor. Uniform guidelines for how to proceed do not currently exist and there are no best practices to follow. This has left agencies to feel their way forward unassisted, despite the legal risks and implications of using these technologies and opening their websites to public participation.
So far, progress across federal agencies is uneven with some sectors of government forging ahead with Web 2.0 projects while others choose to wait. The potential for security breaches and the accidental release of sensitive information are very real and legitimate reasons for hesitancy. The utility of these technologies, and more pointedly social media, is also in question. It is not surprising that some decision makers may view these technologies as entertainment unsuitable for the workplace and not worthy of the investment of taxpayer money. Even when an agency desires to make itself more transparent and participatory, current policies and lawsalong with a riskaverse cultureimpede technological transformation.
How do we bring government more fully into this world of new media, user-generated content, and the rest of Web 2.0? Part of the answer is that it is inevitable. Like the printing press and the published book, Web 2.0 and social media are too much a part of everyday life to ignore. These technologies are effective communication tools we increasingly need in our social and business lives to be considered functioning, connected members of society. Government can hardly expect their employees, who use social networking sites and multiple communication modalities in their personal lives, to give them up when they step into the office. Moreover, they often do not. Look at computer screens in government offices today, and often you will see instant messaging applications and social networking sites alongside spreadsheets and email.
The greater question is, how can government employ these tools in meaningful, useful, responsible, and secure ways? How does an institution that has legitimate concerns about releasing information become more transparent? Change will have to happen on several different fronts. At the higher levels, government leadership has a lot of work ahead changing laws, updating policies, and promoting best practices. At the lowest levels of actual deployment, anyone who undertakes a government Web 2.0 project will need to find creative solutions to meeting government's unique constraints. To devise effective solutions, it helps to understand more about what makes building IT systems for the government different from building them anywhere else.
If the Internet in general, and sites such as USA.gov in particular, have made government more accessible to its citizens, then open technologies and practices can make government more interactive. President Obama's now famous website, Change. gov, for instance, gave citizens access to the President's transition team and timely information on the their progress. Tools that allow citizens to better reach their government could have benefits ranging from a more informed legislative process (tweets from your senator?) to new scientific discoveries (academic and citizen scientists collaborating with government researchers). Government can benefit from open innovation networks, leveraging knowledge from outside experts or the massed wisdom of constituents.
Although open government is popularly associated with gaining access to White House documents or keeping an eye on congressional influence, science and technology also stand to benefit. Government research scientists monitor climate change, track emerging health dangers, develop air traffic control systems, and lead the way in space exploration. Currently they work with private-sector scientists through grants and formal contracts. With open collaborative systems, outside experts without formal government partnerships can help keep government on the cutting edge of scientific knowledge and technical innovation.
Even on internal networks, Web 2.0 tools can improve the efficiency and effectiveness of government operations. After receiving criticism for not sharing surveillance findings across organizations, the U.S. government's intelligence community developed Intellipediaa wiki site for coordinating and sharing information among intelligence, diplomatic, and military organizations on top secret, secret, and unclassified networks. Search tools make previously inaccessible information available to personnel around the world.
Beyond the benefits of efficiency and innovation, the government must use these new technologies just to maintain credibility. Individuals, particularly young people, raised with social media and Web 2.0, see it as the norm. They are comfortable with user-contribution sites like Digg and reddit for getting their news and Wikipedia for researching topics. This generation may not trust an institution that communicates with a monolithic voice, nor will they want to work for an institution that deprives them of social media and other commonplace Internet communication tools. Government cannot fall too far behind this technology curve, or a large segment of the populace will lose confidence in its effectiveness.
Though there is a desire, and even excitement, in many quarters to explore Web 2.0 and social media, current government policies create huge roadblocks. The government is fundamentally different from other institutions because it is subject to federal laws that pertain only to or more strictly to government. Difficulties arise because many laws and policies put in place before the advent of the Web are antithetical to the development of open collaborative systems. For instance, privacy laws that prohibit the government from tracking an individual's activities have resulted in policies that prohibit the use of persistent cookies. This makes it is difficult to create user experiences comparable to those in the rest of the online world. Citizens used to such sites as Amazon.com will be disappointed when their government agency website doesn't "remember" their name and settings the next time they log in.
The legal issues and implications are numerous and often arcane. To cite just a few more, Section 508 of the Rehabilitation Act requiresnot just advisesgovernment information technology systems to provide comparable access to all employees and the public, including those with disabilities. This makes it difficult for government to procure new technologies that are not mature enough to offer complete accessibility features. The Office of Management and Budget (OMB) regulations require agencies to go through a lengthy process to obtain a control number when requesting information from the public. This requirement potentially inhibits the ability of Web managers to use online surveys for collecting customer feedback . Additionally, government moderation of a blog or social network page may violate First Amendment rights and user contributions to government websites would seem to violate the government's restrictions on accepting free labor.
Managing data and systems security is another major challenge in implementing Web 2.0. Access to government data has two sides. Just as government has a responsibility to provide accurate information, it also has the responsibility to protect sensitive data. Government can face unfortunate consequencesfrom lawsuits to putting national security at riskif their systems become vulnerable or proprietary information is shared inappropriately. Unlike Google, which risks fiscal consequences if company information leaks, disclosed or hacked government information may cause irreparable harman extreme example would be the deaths of U.S. soldiers. Thus, government systems are designed to limit information disclosure, which can be a major barrier in the world of Web 2.0
Matthew Linton, an IT security specialist at NASA Ames Research Center, phrases the problem this way: "Because decision-making authorities in government are risk averse and want to clarify policies down to easily digestible sound bites, there is a tendency to go with the highest level of security as a blanket policy ." While understandable, this unfortunately, doesn't allow for any nuancesome agencies may not deal in sensitive information, or a project may include sensitive and non-sensitive data. We can design systems to address these nuances, but such systems are complex and developing them may take more time, money, and effort than decision makers may find worthwhile.
In the past year, there has been a flurry of activity as representatives from different agencies meet to brainstorm about Web 2.0 and social media use. In December 2008 the Federal Web Managers Council produced a white paper on social media in government with specific policy recommendations . Earlier this year, in April, the Center for Strategic and International Studies brought together a panel of "open government champions" from federal agencies to discuss their success and roadblocks . In the future, these efforts will at some point lead to important changes that will make it easier to incorporate Web 2.0 into government operations. Still, waiting for policy changes to trickle down from above will be slow and in fact may be counterproductive.
We spoke with Rob Padilla, legal counsel at the NASA Ames Research Center, about the legal issues surrounding Web 2.0 development. He has worked with several open government community websites at NASA Ames, helping them to develop appropriate community policies and design compliant sites. Instead of focusing on individual laws, he looks at the larger context. He says it is crucial to remember that the government is held to higher standards than other organizations. He mentioned privacywhereas other organizations are ethically bound to protect individual privacy, the government must do so by law. There is also a higher standard, he says, since government must maintain the trust of the people. When someone goes to a government website, there are certain expectations. ".gov means something...It means the [information on the] site is legitimate," explains Padilla . Users have a high level of confidence they won't be deceived or misled by information on a government site.
This helps to explain why the government is conservative about user contributions. It also means the government cannot operate like Facebook and allow just anyone to create an account and post text and pictures. The public understands that Facebook is not responsible for what users post (though there is some expectation of moderation, as evidenced by the recent outcry to remove offensive groups advocating holocaust denial .) Unlike Facebook, the government can and will be held accountable for what appears on federal websites, even if the content was not posted by government officials.
DASHlink is an official government website for open collaboration in research and development. It is a unique government site in that it utilizes two key aspects of the social media movement, social networking and rapid dissemination of information, to further scientific inquiry. It is also an early exploration into the planning and development of an open government website. We were part of DASHlink's user experience and community design team. DASHlink is an example of setting goals for Web 2.0 and social network use, and working through challenges to affect bottom-up policy change and establish best practices.
DASHlink was built under the auspices of the Integrated Vehicle Health Management (IVHM) project in NASA's Aviation Safety Program . IVHM is a branch of systems engineering encompassing a range of technologies, methods, and principles for improving the safety and performance of complex vehicles such as airplanes and spacecraft. IVHM research is, by its very nature, interdisciplinary, with investigations ranging from aerodynamics to structural mechanics to data mining. NASA scientists and engineers collaborate with researchers at various organizations scattered throughout the country, including several NASA field centers, academic institutions, and industrial labs. NASA's IVHM project coordinates myriad research efforts to achieve one overarching goalimproved aviation safety.
DASHlink was initiated as a response to two major project needs: to increase partnership and collaboration among different research communities and to more widely disseminate research results, algorithm codes, and NASA-generated data sets to the scientific community and general public. Web 2.0 technologies were well suited to meet these needs. Lightweight content generation and management technologies would make it possible for the scientists themselves to post information and discussions of their work with ease, while social networking and communication applications could facilitate collaboration. Though having an open community was not an initial goal, we decided that it was crucial community featureallowing the widest possible audience direct access to NASA's scientists and research.
The decision to make DASHlink an official NASA ".gov" site exposed the project to a greater risk of failure than if we kept it on internal networks or put it in the ".org" domain. Building the site required approval from NASA officials who had little experience with open websites, and we knew of no precedent for a public, community-moderated (as opposed to officially moderated) NASA website. The project also faced the risk that NASA researchers would be unwilling to use the site because either it was not useful or they did not want to share work-in-progress on an open site. We used a mix of customer-centered and participatory design methods to discover and address user needs and concerns. While this seems like a routine part of Web design, government agencies must often gloss over designing for the end user due to monetary and time constraints.
Because of unique limitations, it takes more effort and planning to develop new information technology systems in government compared with the private sector. Project planning and our iterative development process took into account the effort required to work through the approval process. To address challenges from government policies and regulations, we did a lot of up-front work to identify the policy compliance officials and legal staff who had the responsibility to examine and approve our site design and online community policies. From the earliest development iterations, our design team met with NASA officials and legal advisors to interpret policy, understand security issues, refine the design, and craft our terms of service. When there was a conflict between official policy and our design, we first sought to change policy. When this was not possible, we generated multiple design alternatives to respond to different possible policy-negotiation outcomes and worked with a legal advisor to evaluate these alternatives. In the course of the project our legal advisor became a virtual member of our design team.
The posting policy and registration process provide examples of how decisions from NASA officials affected our website design. Traditionally, NASA researchers go through a formal review and approval process before they release information to the public via publications, presentations, or even public websites. Our user research indicated that users rarely updated information on official websites due in part to an inconvenient approval process. To make the DASHlink effective, it was critical to allow users to post directly to the site without an intermediate approval step. To enforce preapproval would make DASHlink essentially no different from any other government website and ineffective for real collaboration.
Rightfully, NASA policy makers and legal staff were concerned about letting just anyone post on a NASA website with only community moderation in place. No matter how many disclaimers we may add to a government website, the .gov domain still implies that all of the content is government sanctioned. We had to find a middle ground.
We contrived an elegant solution that would meet both DASHlink's needs and government policies. Only DASHlink community members can create content, and membership is limited to an approval process. NASA civil servants determine a potential DASHlink member's eligibility, with approval occurring during the registration process. (The site is open to any NASA civil servant, and they are exempt from the approval step.) The requirements for membership are liberal: expertise appropriate to the subject matter and the potential for affiliation with NASA. The latter means that if, at some point in the future, a potential user or the user's organization signs a contract with NASA, then that user is eligible for membership.
This made the community open enough so users without any current formal connection to NASA would feel welcome. At the same time, the restrictions created a community of peers, since there are no anonymous users and everyone knows someone on the site. This provided assurance to our policy and legal teams that our terms of service effectively reduced the likelihood of spam or inappropriate content. It also ensured the quality of the community, making DASHlink membership more attractive to researchers.
The site made its public launch in June 2008, and it has shown some success as an example of the benefits of Web 2.0. One of the keys to acceptance by our users was ease of use afforded by simple to use content-generation tools. For instance, to present findings from a data analysis, a scientist can easily create a page, add text that describes analysis objectives, upload visualizations of the findings, and include links to the algorithm and data set used. Because the scientist controls his or her content on the site, they are more likely to contribute new information as their research grows. Surveys indicate that users find the site useful for staying up-to-date on their colleagues' research. Users also report that they use the member-to-member messaging service to establish connections with potential collaborators. Like most websites, DASHlink continually evolves in response to its actual use.
The takeaways from our experience fit into three main ideas: agencies should foster grassroots efforts and not wait for broad policy change from above; working through individual projects is a productive way to shape new policy; and grassroots efforts are most effective if projects begin with a clear goal.
It is neither feasible nor advisable for government to hold off on embracing these technologies. As new generations join the ranks of civil servants and contractors, these tools will be brought in whether policy makers are ready or not. By starting as soon as possible, government will have time to adapt and create a more suitable and mutually beneficial work environment. Following on this idea, agency officials should view small, grassroots projects as opportunities to begin thinking about legal and security policy changes. With these real projects in mind, they may avoid making broad policy declarations that could hamper future development of useful Web 2.0 systems and also understand the impact and benefit new technologies can have on workforce and operations.
The grassroots projects themselves should articulate definitive goals. In the world of Web 2.0 startups, it seems common to develop new technologies without a clear idea of their usefulness. This may work in Silicon Valley, but could be disastrous for government. Decision makers rightly feel a responsibility to shield taxpayer money from "schemes" and protect government resources from ill-considered risks. On the other hand, the government can minimize these risks. Government doesn't need to invent new Web 2.0 tools or be the next Googleit just needs to use already proven tools.
Though Web 2.0 proponents in government should not wait for policy change from above, they should accommodate policy and legal considerations. At first the system of laws, government traditions, and public expectations may inhibit new technology adoption, but all of these are malleable. A willingness to work with policy makers and include them in the design process early on is more likely to yield meaningful policy changes than ignoring or resisting legal issues altogether. By nurturing a shared understanding of how new technologies can be used to achieve definitive goals, Web 2.0 leaders and policy makers can manage risks and develop creative technical and policy solutions together. Inevitably, these are risks that must be faced, benefiting everyone in the long run.
The authors wish to thank Ashok Srivastava (NASA), Bryan Matthews (Stinger Ghaffarian Technologies), and Dawn McIntosh (NASA) for their thoughtful reviews. We are grateful to Rob Padilla (NASA) and Matthew Linton (NASA) for sharing their experience and wisdom. Chris Fattarsi (ASANI Inc), David Kluck (Mission Critical Technologies), and Eric Titolo (Mission Critical Technologies) taught us a lot about the development process. This project was funded in part by the IVHM Project of NASA's Aviation Safety Program.
2. WebContent.gov, "Matrix of Web 2.0 and Government," July 2008. Retrieved from usa.gov June 4, 2009, http://www.usa.gov/webcontent/documents/Web_Technology_Matrix.pdf
3. The White House, "Transparency and Open Government, Memorandum for the Heads of Executive Departments and Agencies," January 2009. http://www.whitehouse.gov/the_press_office/Transparency_and_Open_Government
4. Goodwin, B., Campbell, S., Levy, J., and Bounds, J., "Social Media and the Federal Government: Perceived and Real Barriers and Potential Solutions," Federal Web Managers Council, December 2008. http://www.usa.gov/webcontent/documents/SocialMediaFed%20Govt_BarriersPotentialSolutions.pdf
6. Center for Strategic and International Studies, "Open Innovation for Government, A Panel Featuring Open Government Champions from Federal Agencies," April 14, 2009. http://csis.org/event/openinnovation-government
8. Nickson, C. "Facebook Refuses to Ban Holocaust Deniers," Digital Trends, May 12, 2009. http://news.digitaltrends.com/news-article/19912/facebook-refuses-to-ban-holocaust-deniers
9. DASHlink. https://DASHlink.arc.nasa.gov
10. NASA Aeronautics Research Mission Directorate, Aviation Safety Program. http://www.aeronautics.nasa.gov/programs_avsafe.htm
Francesca Barrientos is an independent consultant in design research and user-centered design. Her approach to user-centered design is informed by her deep technical background in several engineering disciplines. Barrientos is also a design methods researcher and studied the work practice of mission design engineering teams while she was a research scientist at NASA Ames Research Center. Prior to becoming a researcher, she held several positions as a software developer; she was also a mechanical engineer. She holds a Ph.D. in computer science and a B.S. in mechanical engineering.
Elizabeth Foughty is a contractor (MCT Inc.) at NASA Ames Research Center. Her background includes Web design, project coordination, and public affairs work in politics and government. Foughty's social media expertise is of the first-hand varietyas a Gen Y user who grew up using these tools and as a designer and implementer of Web 2.0 tools at NASA and DARPA. She currently spends her days putting her youthful idealism and tech savvy to good use by working to create a more open and tech friendly government. She has co-authored papers and presentations for CHI 2009, Gilbane Boston, and the NASA Program Management Challenge. She holds a B.A. from Boston University.
Figure. The sequence (from left) shows a lightning bolt striking a metallic airplane then hitting the ground, as the flight takes off from Osaka, Japan. Instances such as this are useful to researchers at DASHlink. One of the goals of the IVHM project is to examine the direct and indirect effects of lightning on composite-based aircraft.
©2009 ACM 1072-5220/09/0500 $10.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.