The Reputation Society (edited by Hassan Masum and Mark Tovey; MIT Press, 2012) is a collection of essays discussing the benefits and risks of online reputation. It focuses on asking the right questions today, so that reputation is better used in society tomorrow.
Expert contributors offer perspectives ranging from philanthropy and open access to science and law. The 18 chapters are divided into 6 thematic parts. (The Table of Contents, sample chapters, and reviews are on the Reputation Society MIT Press web page.)
Citation: Masum, H., & Tovey, M. (Eds.). (2012). The Reputation Society: how online opinions are reshaping the offline world. Cambridge, MA, USA: MIT Press. ISBN 978-0-262-01664-3.
The Foreword and Introduction begin the book.
Foreword: Trust, Reputation Systems, and the Immune System of Democracy. Craig Newmark suggests that power is shifting toward those people with the best reputations and trust networks. Over time, reputation might help build "the immune system of democracy".
"Once the bugs are worked out, reputation and trust might be a key part of the immune system of democracy–a set of technologies and practices that help power, influence, and legitimacy to flow toward those who are willing and able to tackle the challenges that affect us all." (Newmark)
Introduction: Building the Reputation Society. Hassan Masum, Mark Tovey, and Yi-Cheng Zhang summarize the past, present, and potential futures of reputation systems, and survey the contents of the book.
"Properly designed reputation systems have the potential to reshape society for the better by shining the light of accountability into dark places, through the mediated judgments of billions of people worldwide, to create what we call the Reputation Society." (Masum, Tovey, and Zhang)
Part I lays a base for what follows by focusing on understanding reputation.
Chapter 1: Designing Reputation Systems for the Social Web. How can online reputation systems be designed to meet the needs of both users and system designers? Chrysanthos Dellarocas looks at the promises and challenges of online reputation design, using Yelp as a case study. He considers how to implement the right incentives for good behavior, truthful reporting, and resistance to manipulation. Four key design decisions are: what information should be included in one’s reputation profile? How can this information be obtained? How can it be aggregated and displayed? How should manipulation be dealt with?
"Reputation systems are ubiquitous in the social web, and their design principles surprisingly subtle." (Dellarocas)
Chapter 2: Web Reputation Systems and the Real World. When anticipating the future of society and technology, it is advisable to look to the past for relevant successes and failures. Randy Farmer shares some of the signposts, warnings, and potential solutions he has encountered over two decades of building reputation systems. The special role of personal reputation justifies a special name for it: "karma". There are advantages to focusing on positive karma, and keeping it context-limited. The socially transformative potential of digital reputation systems needs careful evaluation—for example, claims that karma will translate to wealth and offline power.
"As reputation systems increase in real-world influence, the importance of who keeps, calculates, and displays the scores will become a greater societal discussion. .... do we need a 'Good Housekeeping Seal of Approval' for reputation systems?" (Farmer)
Chapter 3: An Inquiry into Effective Reputation and Rating Systems. What are the fundamental principles that make a reputation system effective? John Henry Clippinger discusses connections between evolutionary biology, finance, and reputation systems. Honest social signals are hard to achieve, and reputational bubbles all too frequent. Clippinger questions whether today's search engines and rating agencies can be impartial and disinterested referees of reputation. He suggests design principles to make reputation systems more effective—for example, transparent and testable reputation metrics and processes.
"Reputation is not so much a thing as an ongoing contest." (Clippinger)
Part II takes up challenges of regulating societies.
Chapter 4: The Biology of Reputation. Online reputation rests on our instinctive traits, which have evolved over millions of years. John Whitfield shows how aspects of reputation have occurred in the evolution of our own species and in the animal world. For example, many animals use social learning and eavesdropping to rate other animals and their behaviors. Humans become more honest and generous when being watched, even when only by a stylized pair of eyes. Reputation has historically helped to defend co-operation, commerce, and honor. Whitfield suggests lessons for the future that can be learned from the past, such as punishing defection by giving negative ratings more weight than positive ones. The need for punishment itself can be reduced when reputations enable more trustworthy and competent partnerships.
"The job of reputation systems is to reward cooperation and reduce temptation." (Whitfield)
Chapter 5: Regulating Reputation. When should reputation systems be regulated? Eric Goldman explores challenges and future options for regulation of reputational information, and extends Adam Smith's famous "invisible hand" metaphor. He considers how well-functioning marketplaces depend on the vibrant flow of accurate reputations. When information about producers and vendors is costly, reputational information can help consumers make better decisions, acting as the invisible hand of the invisible hand (or the "secondary invisible hand"). But this secondary invisible hand itself can be hard to evaluate – it can be costly to find reputational information and to assess its credibility. Hence the reputation of reputation systems can be thought of as a "tertiary invisible hand". When the tertiary invisible hand fails to ensure useful reputational information, regulatory intervention may be necessary to protect the marketplace.
"I describe the reputation of reputation systems as a tertiary invisible hand – it is the invisible hand that guides reputational information (the secondary invisible hand) to guide the invisible hand of individual uncoordinated decisions by marketplace actors (the primary invisible hand)." (Goldman)
Chapter 6: Less Regulation, More Reputation. Can reputation complement regulation? Lior Strahilevitz considers this question in the context of landlord-tenant relations, antidiscrimination law, and commercial disputes. We often engage in statistical discrimination as a protective strategy when dealing with strangers. This type of discrimination could be reduced with better reputational information on each individual. Strahilevitz proposes that greater access to reputational information could be a twenty-first-century public good.
"If the state is strategic about the sorts of information it gathers and disseminates, then it will have at its disposal an additional, indispensable tool for combating discrimination, fraud, professional incompetence, and various other social ills." (Strahilevitz)
Part III looks at reputation's role in amplifying signals.
Chapter 7: The Role of Reputation Systems in Managing Online Communities. Reputation systems typically provide decision support to their users. However, Cliff Lampe discusses how they can also perform more subtle socializing functions, such as making desirable behaviors clear and supporting community norms. By providing feedback about behavior, penalizing negative actions, signaling desired outcomes, and rewarding users, reputation and recommender systems help organize online environments. Examples mentioned include Amazon, Slashdot, Everything2, CouchSurfing, LinkedIn, eBay, and Digg. Both explicit and implicit feedback can help to socialize new users and maintain online communities.
"[R]eputation and recommender systems can provide the feedback necessary to assist in the governance of online communities." (Lampe)
Chapter 8: Attention Philanthropy: Giving Reputation a Boost. Some ideas should have a stronger reputation. Without more recognition, their potential is lost to society. Alex Steffen discusses how "attention philanthropy" can amplify these faint signals. It can help important ideas to enter the mainstream by acting as a reputation booster that redistributes attention. Steffen makes the case for this approach, and envisions a future reputation-enabled attention philanthropy project.
"Experience suggests that attention philanthropy works best when preference is given to ideas and ratings by people who understand the deeper issues at play and the reasons why an idea has long-term potential–especially if such people have themselves been innovators." (Steffen)
Chapter 9: Making Use of Reputation Systems in Philanthropy. "Peer-to-peer philanthropy" leverages online systems to connect donors directly with charitable causes. However, it can be difficult to determine which worthy-sounding projects achieve real impact. Marc Maxson and Mari Kuraishi suggest how reputation information in philanthropy can overcome this gap, drawing lessons from their GlobalGiving platform. They suggest that greater trust is needed amongst NGO donors, clients, and staff. GlobalGiving connects NGOs which have community- and world-changing ideas with people who can support them. The reputational information that it considers from NGOs has evolved over time—it now includes project updates, feedback from partners and beneficiaries, social media footprint, and "visitor postcards" reporting what third-party volunteers saw and heard during their visit.
"In the longer term, making use of reputation systems in philanthropy can motivate more and smarter giving and suggest lessons for the introduction of reputation systems into other areas of society." (Maxson and Kuraishi)
Part IV explores reputation's functions in supporting science.
Chapter 10: The Measurement and Mismeasurement of Science. The public has a natural desire to maximize its return on investment for science funding. This has made reputation measures for scientific research increasingly influential. Michael Nielsen warns of the dangers of over-reliance on any single metric of success. "Centralized metrics" suppress cognitive diversity, create perverse incentives, and lead to the misallocation of resources. Nielsen suggests instead using diverse approaches to evaluating science—a perspective relevant to other fields involving innovation and discovery.
"If even Einstein demonstrably made enormous mistakes in judging his own research, why are the rest of us trying to measure the value of science systematically and even organizing the scientific systems of entire countries around these attempts? .... No one sensible would staff a company by simply applying an IQ test and employing whoever scored highest." (Nielsen)
Chapter 11: Usage-Based Reputation Metrics in Science. Citations are a valued currency in research. They are an easily measurable proxy for the attention of professional peers and the impact of scholarly work. However, the excessive use of citation-based metrics for judging research (and researchers themselves) has been criticized by many. Victor Henning, Jason Hoyt, and Jan Reichelt propose usage-based metrics, which would measure how articles are actually read and used. These metrics would be both harder to game and available immediately on a per-article basis. Collaborative filtering based on usage data could promote interdisciplinary research and connect people with similar research interests. It could also highlight controversial papers. Challenges include privacy, reputational bubbles, misuse, and social hurdles. The authors discuss existing approaches, including their Mendeley research service.
"By looking at longitudinal trend data, scholars might be able to spot whether a paper, topic, or theory is steadily gaining followers, is subject to sudden 'hype', or is already on the decline and is cooling off." (Henning, Hoyt, and Reichelt)
Chapter 12: Open Access and Academic Reputation. Open access aims to make knowledge freely available to those who would make use of it. John Willinsky considers the effect that open access is having on publication in academia. While scholars have been part of a reputation society for centuries, open access may impact the reputations of individual scholars—and the reputation of scholarship in general. High-profile open access journals like PLoS (Public Library of Science) have demonstrated a viable model: researchers pay a one-time fee upon publication so that the public at large can forevermore access this research without barriers. Open access publication can benefit researchers while creating public value.
"[O]pen access holds considerable promise for improving the standing–or reputation–of research and scholarship more broadly." (Willinsky)
Part V considers how reputation can help in improving policies.
Chapter 13: Reputation-Based Governance and Making States "Legible" to Their Citizens. How can government be made more effective for citizens? Lucio Picci argues that one way is to make government behavior more transparent and understandable, or "legible," to citizens. Legibility can enable "reputation-based governance"—citizen rating of government bureaucrats and policies. This can improve the incentives of governments and other large organizations. Reputation-enabled government legibility would allow citizens to interpret the vast and complex set of information by which governmental actions can be evaluated.
"Imagine a scenario in which citizens assess policies online, these assessments form the basis for reputational measures of public officials and other actors of governance, and these measures in turn influence governance decisions–for example, by determining bureaucrats' promotions and the choice of policies." (Picci)
Chapter 14: Trust It Forward: Tyranny of the Majority or Echo Chambers? If reputation systems weight all perspectives similarly, they may devolve into simple majority rule, and penalize people who think differently from the average. But if reputation systems only show each user perspectives from other similar users, they may become "echo chambers" in which like-minded people reinforce each others' views without being open to outside perspectives. Paolo Massa discusses design choices that are relevant to these two extremes. One conceptual distinction is between "global trust" (where a user gets the same reputation from the perspectives of all other users) and "local trust" (where a user can get different reputations from the perspective of each other user). Navigating these extremes has broad implications for our future societies.
"When, or if, there are no more cultural elements capable of bonding us together because all have become singletons with their own peculiar and totally personalized cultures, the very existence of our society may be at risk." (Massa)
Chapter 15: Rating in Large-Scale Argumentation Systems. How can reputation enable more effective and collaborative deliberation? Luca Iandoli, Josh Introne, and Mark Klein explain how "argument maps" allow the deliberations of many people on controversial topics to be visually represented. Argument maps can be extended by incorporating ratings. This allows mass collaboration to highlight ideas of collective interest. It may also allow automatic analysis of argument consistency, areas of consensus or conflict, and participant biases and strengths. Iandoli, Introne, and Klein report on applications in the areas of biofuels and climate change (the "Climate Collaboratorium"). They suggest that reputation-enabled online argumentation might help us to collaborate despite our varied biases and interests.
"Our approach to large-scale problem solving integrates community rating with large-scale argumentation systems to enable semistructured deliberation by very large groups. .... Ultimately, such a tool might help us to overcome our collective biases, to synthesize our increasingly balkanized social dialog, and to achieve the collective wisdom we need to solve the challenging problems we face today." (Iandoli, Introne, and Klein)
Part VI concludes the book by envisaging scenarios for the Reputation Society.
Chapter 16: Privacy, Context, and Oversharing: Reputational Challenges in a Web 2.0 World. Our online reputation can be at risk when personal information spreads farther than expected. Michael Zimmer and Anthony Hoffman address the twin risks of information spreading beyond its intended context and oversharing of personal information. New technologies have eroded control over information flows. Tools meant to increase convenience can also propagate information far beyond its intended context and audience. The idea of "contextual integrity" suggests informational norms that should be adhered to, and whose violation raises privacy concerns. Zimmer and Hoffman use examples from Facebook, and from instances of "oversharing" (inappropriately pushing intimate details to near-strangers). They suggest giving users more feedback on and control of their informational flows, and encouraging better design of future information-sharing and reputational platforms.
"Social networking sites should provide users with easier means to manage their identities across contexts, and provide users with visual feedback as to how far their information sharing might extend." (Zimmer and Hoffman)
Chapter 17: The Future of Reputation Networks. How will reputation systems evolve in practice? Jamais Cascio suggests that there are two basic issues to watch. Ratings can be given directly by participants, or be emergent from other factors. Participation can be constrained within particular communities, or be universal. Cascio discusses how combining these two issues gives four fundamental types of reputation networks: Direct-constrained, Direct-universal, Emergent-constrained, and Emergent-universal. Four short future scenarios then illustrate how each of the four types of reputation networks might unfold.
"[F]or me, in thinking about the evolution of reputation networks, two issues loom large above the others. The first concerns process: how are the reputation ratings generated? Do they come from other participants offering explicit ratings, or are they the result of analysis of factors that aren't solely reputation-related? The second concerns participation: whom do the networks cover? Is participation a choice, and what are the implications of opting out?" (Cascio)
Chapter 18: "I Hope You Know This Is Going on Your Permanent Record". Cory Doctorow was one of the first speculative fiction authors to portray a reputation-based future society. Fellow author Madeline Ashby sets the stage for a Doctorow excerpt portraying an unusual university environment. She then gives a provocative scenario of how reputation systems could influence another educational arena.
"Relatives and supervisors may furrow their brows when you express an opinion they disagree with, but your friends online can vote you up and down, give you hearts or stars or gold, or nix you from their friends list. This is the charm of reputation economies. They codify the currents that once flowed invisibly beneath the surface of our conversations and transactions." (Ashby)
Our book's essays focus on long-term challenges and principles.
For detailed case studies of negative developments on the Internet, we suggest The Future of Reputation, The Offensive Internet, and The Net Delusion. These books discuss dangers such as digital mobs, abusive online behavior, and use of the Internet by totalitarian regimes.
On the system design and use side, Building Web Reputation Systems gives specific tactics for addressing some of the current design challenges that online reputation poses. Wild West 2.0 is a complementary book on protecting one's online reputation.
Rather than add incrementally to what exists, we chose to focus on longer-term challenges of reputation systems.
To take one example: if online reputation weights all perspectives similarly, it may devolve into simple majority rule. At the other extreme, if online reputation takes only similar users’ opinions into account, “echo chambers” may arise where like-minded people reinforce each others’ views without being open to outside perspectives.
The chapter by Paolo Massa discusses balancing these two extremes. As Massa notes, this issue can be traced back to political philosophers like John Locke, and even to the writings of Aristotle and Plato. We believe it is a fundamental design issue for reputation infrastructure—an issue that will still be important 10 or 20 years from now, long after today’s ephemera are forgotten.
Other authors in the book also take the long view. The questions they consider include: can reputation mitigate discrimination? How does open access shape the reputations of academics, and of research itself? What is necessary for individual government policies and projects to be rated by citizens? Can we make better choices about which scientific paths to pursue?
We agree with concerns about the dark side of Internet reputation. From our Introduction:
“Several authors have expressed grave reservations about what reputation systems may lead to in the long run: a mob mentality, vigilantism, a pitiless collective memory that never forgets a past misdeed …. Could the Reputation Society be nothing more than a conformity-enforcing Panopticon—one that could be hijacked by authoritarians as a more effective means of social control?”
We hope that our book contributes to avoiding these dangers, and achieving the positive potential of reputation systems for improving social progress and personal living standards.
This will require sustained engagement by society at large. Effective design can steer reputation systems away from pitfalls of online opinion sharing—motivating honesty, protecting privacy, and discouraging vigilantism.
This is not a book of easy answers. It is a book of hard questions.
Engaging with these hard questions is necessary. Only in this way can we achieve the insight necessary to guide the development of online technologies in a humane direction.
General reading:
Manifesto for the Reputation Society - discusses why we need a reputation infrastructure and where it might take us.
The Future of Reputation - book on "gossip, rumor, and privacy on the Internet".
People Will Talk - book on the science behind animal and human reputation.
TOOL: The Open Opinion Layer - a vision of an open infrastructure for sharing opinions and reputations.
Wild West 2.0 - book on protecting your online reputation.
Technical reading:
Building Web Reputation Systems - book and blog on the nuts and bolts of reputation system design.
Computing with Social Trust - book on computational approaches to trust in social contexts.
Designing Reputation Mechanisms - design considerations and practical implications.
Happier World with More Information - modeling mutually beneficial relations from increased information.
MIT Reputation Symposium Report (2003) - a milestone symposium on design of reputation mechanisms.
Information overload, challenges of evaluating quality, and the opportunity to benefit from experiences of others have spurred the development of reputation systems. Most Internet sites which mediate between large numbers of people use some form of reputation mechanism: Slashdot, eBay, ePinions, Amazon, and Google all make use of collaborative filtering, recommender systems, or shared judgments of quality.
But we suggest the potential utility of reputation services is far greater, touching nearly every aspect of society. By leveraging our limited and local human judgement power with collective networked filtering, it is possible to promote an interconnected ecology of socially-beneficial reputation systems - to restrain the baser side of human nature, while unleashing positive social changes and enabling the realization of ever-higher goals.
...from "Manifesto for the Reputation Society"