Handbook of Research on Socio-Technical Design and Social Networking Systems

Brian Whitworth Massey University-Auckland, New Zealand

Aldo de Moor CommunitySense, The Netherlands

Volume II

Information science reference Hershey • New York Director of Editorial Content: Kristin Klinger Senior Managing Editor: Jamie Snavely Managing Editor: Jeff Ash Assistant Managing Editor: Carole Coulson Typesetter: Michael Brehm Cover Design: Lisa Tosheff Printed at: Yurchak Printing Inc.

Published in the United States of America by Information Science Reference (an imprint of IGI Global) 701 E. Chocolate Avenue, Suite 200 Hershey PA 17033 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: [email protected] Web site: http://www.igi-global.com and in the United Kingdom by Information Science Reference (an imprint of IGI Global) 3 Henrietta Street Covent Garden London WC2E 8LU Tel: 44 20 7240 0856 Fax: 44 20 7379 0609 Web site: http://www.eurospanbookstore.com

Copyright © 2009 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher. Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark.

Library of Congress Cataloging-in-Publication Data

Handbook of research on socio-technical design and social networking systems / Brian Whitworth and Aldo de Moor, editors. p. cm. Includes bibliographical references and index. Summary: "Every day throughout the world, people use computers to socialize in ways previously thought impossible such as e-mail, chat, and social networks due to emergences in technology. This book provides a state-of-the-art summary of knowledge in this evolving, multi- disciplinary field"--Provided by publisher. ISBN 978-1-60566-264-0 (hardcover) -- ISBN 978-1-60566-265-7 (ebook) 1. Online social networks. 2. Internet--Social aspects. 3. Information technology--Social aspects. I. Whitworth, Brian, 1949- II. Moor, Aldo de. HM742.H37 2009 303.48'33--dc22 2008037981

British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library.

All work contributed to this book set is original material. The views expressed in this book are those of the authors, but not necessarily of the publisher. 715

Chapter XLVII Resolving Wicked Problems through Collaboration

Peter J. Denning Naval Postgraduate School, USA

Abstract

Wicked problems (messes) are tangled social situations that are too costly to stay in and too intransigent to get out of. Collaboration is essential to resolving them. This chapter examines five main ideas: (1) Messes and wicked problems are the most difficult in a hierarchy of difficult problems.(2) Why mess resolution usually involves disruptive innovation. (3) Why collaboration is essential and hard to achieve. (4) Collaboration is a practice generated in six kinds of conversations. (5) Someone who understands the practice of collaboration will find many information technology tools to help with the process:exchangers, coordinators, and games, and can design better tools.

Every revolutionary idea seems to evoke three stages of reaction: It’s completely impossible. It’s possible, but it’s not worth doing. I said it was a good idea all along.

—Arthur C. Clark

The Americans can be counted on to do the right thing, after they have exhausted all the alternatives.

—Winston Churchill

Introduction to stay in and has no obvious way out?” Messes are also called wicked problems. The various players The question is simple enough: “How can we resolve cannot agree on the nature of the problem or on a mess, a tangled social situation that is too costly solution approaches. Their search for solutions

produces few results and seems open-ended amidst through experiments (Category III). When the constantly shifting constraints (Denning 2007, system of interest is complex and adaptive, it tends Roberts 2000, 2001). The end state is a moving not to have reproducible behaviors; it adjusts its re- target (Reeves 1991). The purpose of this chapter sponses and neutralizes repeated probes (Category is to shed light on effective strategies for resolving IV). The last category is the abode of messes and messes and the technologies available to support wicked problems. those strategies. These categories blend together ideas from Kurtz We will investigate four main ideas: the nature and Snowden (2003) and Roberts (2001). Kurtz of messes, why mess resolution is likely to involve and Snowden discuss the notions that Category III disruptive innovation needs collaboration, strategies problems may be complex natural systems gov- for organizing collaboration to confront a mess, and erned by unknown laws waiting to be discovered, technology tools to support collaboration. Most and that Category IV problems are complex social existing “collaboration technologies” are good for systems. Roberts lumps our Categories I and II information sharing but not true collaboration. into a single class (“simple problems”) and uses However, someone who understands the practice the terms “complex problems” for our Category III of collaboration will find many tools to help with and “wicked problems” for our Category IV. These the process. categories represent the degree of agreement among the social power centers about the problem and its possible solutions. The simple problems are those SOLVING HARD PROBLEMS IN in which everyone agrees on the problem definition SOCIAL SYSTEMS and there is a power center that can implement the change. The complex problems are those in which Let us begin by considering messes as a category everyone agrees on the problem definition, but there within a hierarchy of difficult problems. We use the is no consensus among power centers on how to word “system” to mean either a social or natural proceed. The wicked problems are those for which system. there is no consensus on the problem definition or Problems come in four categories of difficulty on the solution approach, and partisan interests (Table 1). The simplest are the ones where the block collaboration. solution knowledge already exists, either in one’s These categories suggest a strategy for solving own domain (Category I) or in another (Category a problem of unknown difficulty. We start with the II). The more difficult require the construction of hypothesis that our problem is of Category I, and new knowledge. When the system of interest is then work our way upwards through the categories complex and governed by fixed (but unknown) until we find a solution or know that we confront laws, its reproducible behaviors can be discovered a mess. If our problem is Category I or II, we will

Table 1. Categories of problem difficulty

Name Category Characteristics Actions I Solution knowledge exists in your own domain Redirect attention. Simple Problems II Solution knowledge exists in another domain Find an expert. Become an expert and design own solution. III No solution exists in any domain; system is very Explore for recurrent patterns by probes and Complex complex but responds the same way to repeated experiments, design resolution around recurrences Problems stimuli discovered. Wicked IV No solution exists in any domain; system is chaotic Organize collaboration in a local part of system, then Problems and adaptive, does not repeat patterns under the same spread the new organization to the whole. (messes) probes

716 Resolving Wicked Problems through Collaboration

discover an expert to help us. If our problem is Cat- The systems embodying wicked problems tend egory III, there is no expert, but we will discover and to resist and defy attempts at change. Change oc- exploit the system’s recurrences for a solution. If our curs only when leaders achieve consensus among problem is Category IV, we will find no recurrences; power centers to enact new social agreements and we can employ the strategies discussed shortly for new organization within the system. The leaders resolving messes. look for ways to bring about local solutions that Category III problems tend to appear around can be propagated to the whole. undiscovered aspects of natural systems. They de- Candace Lightner founded Mothers Against mand a level of skill comparable to a cutting-edge Drunk Driving in 1980 as a response to a wicked scientist who is capable of inventing new hypotheses problem: the widespread tolerance of drunk driv- and validating them with experiments. Sometimes ers and their annual carnage. Lightner and her current methods and instruments are not powerful colleagues showed great skill in attracting media enough to discern the patterns needed to solve the attention and in gaining the support of politicians problem; the solution may have to wait for a later for new laws. (See www.madd.org.) age with finer instruments. Throughout most of the 1800s, for example, physicists hypothesized that light traveled in a medium called “ether”. They could not MESSES AS SOCIAL TANGLES verify this because they lacked the instruments to measure ether. In 1887 the Michelson-Morley ex- Here are fifteen contemporary examples of periment provided the instrument. That instrument’s messes: failure to detect any ether influenced Einstein’s 1905 inspiration for relativity: he postulated there • Spam is no ether and light travels at the same speed in all • Identity theft frames (Einstein 1916). • Information overload Category IV problems tend to appear in conflicted • Sustainable versus secure infrastructures social systems. Nancy Roberts refers to them as • Getting dependable, reliable, useful, safe, and wicked problems: secure software • Drug resistant bacteria Government officials and public managers are en- • Preventing a pandemic countering a class of problems that defy solution, • Global warming even with out most sophisticated analytical tools. • Quality education in public schools These problems are called “wicked” because they • Planning for affordable housing have the following characteristics: (1) There is no • Obesity epidemic definitive statement of the problem; in fact, there is • Health care cost crisis broad disagreement on what “the problem” is. (2) • Poverty The search for solutions is open ended. Stakehold- • Thwarting terrorist plots ers champion alternative solutions and compete to • Reconstructing society after war or disaster frame “the problem” in ways that directly connect their preferred solution and their preferred problem The first eight of these seem like technology definition. (3) Resources and political ramifications problems and the last seven like social problems. are constantly changing. (4) Constraints constantly However, they are all social problems. The first change as interested parties come and go. (Roberts, eight are the social consequences of pushing tech- 2001, p353) nology beyond its limits; their resolutions lie in the social domain. Messes cannot be resolved without untangling the social situation. The signature signs

717 Resolving Wicked Problems through Collaboration

Table 2. Signs of a mess

Threat Something of great value is threatened in a large community; many stakeholders are involved. No progress Little or no progress despite huge effort; improvements haven’t worked; existing solutions are ad hoc, incompatible, and ineffective. Social paralysis No agreement on problem statement, causal relationships, or solution strategies. Active resistance Multiple stakeholders have social and political means to block actions that do not support their agendas. They distrust or resent one another. Negative moods Frustration over disordered conditions, feeling of being stuck, confusion, discord, conflict, turmoil, controversy, distrust, resentment. of a mess—all social—are summarized in Table 2 accompanying great scientific revolutions lasted (Denning 2007). 50-100 years each (Burke 1995). We prefer the term “mess” for these exception- There are three main observers in this structure. ally difficult situations because the word “problem” Alice (A) is embedded in the mess and lives under its (as in wicked problem) carries a connotation that mindset and rules. To Alice, the mess looks normal we can articulate the nature of the concerns and and impossible to change. Bob (B) is embedded in conflicts sufficiently well to anticipate an “answer”. the settlement and its mindset and rules. To Bob, the Messes don’t have answers. With a mess, the prob- settlement looks normal and the ways of the mess lem is that we don’t know how to characterize it as archaic. Chris (C) straddles the transformation and a problem. We may even disagree about whether sees both the mess and the possibility of resolution. there is a mess at all. We will design a strategy for C shortly. These three observers personify the stages in Arthur Clark’s quip about revolutions. Alice says, MESSES AND INNOVATION “Change is impossible.” If Alice meets Chris, she will say, “Your proposed change is not worth doing.” If History tells us that solutions to messes are likely to Alice survives and becomes like Bob, she will look require disruptive innovations (Christenson 1997). back and say, “It was a good idea all along.” The reason is that the paradigm (belief system) of Sometimes a chain of sustaining innovations the mess dwellers has already proved itself incapable will collectively create sufficient disruption to alter of resolving their difficult situation. Only a belief- the mess. Computational science gives an example. changing innovation will succeed. This is why many In the 1970s, scientists and engineers articulated in the mess feel threatened about the prospect of a many grand challenge problems—such as designing solution. The solution may challenge everything aircraft with computer simulation of flight instead connected with the mess, including social power of with wind tunnel tests. These problems were in- structures and deep beliefs. solvable with the supercomputers of the day, which Figure 1 depicts a temporal structure to a mess. performed around 1 million operations per second. The horizontal line represents time. The mess con- Scientists estimated those problems would yield to dition builds in the social system and exists for a supercomputers of 1 billion operations per second—a period of time. A transformational event provides thousand times faster—but such supercomputers the key to a resolution. The social system integrates were dauntingly expensive. Twenty years later, the resolutions and settles down with the mess gone. Moore’s Law had given us 10 doublings of comput- Smaller social systems resolve more quickly than ing power—the required thousand gain—enabling bigger ones. Highly uncomfortable messes resolve solutions to those grand challenge problems. The more quickly than less uncomfortable ones. James Boeing 777 aircraft was a product of this advance. Burke showed that the messes and settlement periods

718 Resolving Wicked Problems through Collaboration

It is worth noting that a mess is not a necessary Roberts begins the class by posing a wicked precondition for disruptive innovation. Many dis- problem and asking everyone to come up with a ruptive innovations arise from other conditions and solution to it. When they come together and report motives. For example: (1) Someone serendipitously their proposals, the group judges no solution satis- stumbles on something wonderful (e.g., penicillin). factory. Their proposals typically involve getting an (2) Someone envisions a whole new potential (e.g., appropriately high authority to make and enforce key Alan Kay’s Dynabook led to laptop computers). (3) declarations. For example, a green infrastructure is Someone creates a new social entity that unleashes best achieved by establishing a new cabinet-level creativity and new values (e.g., MySpace). (4) “infrastructure czar” who can set sustainability goals, Someone seeks a cheap way to give to the many create timetables for their completion, and inflict what only the well-to-do can afford (e.g., Unix, punishments on those who do not comply. Internet telephony, low-cost inkjet printers); accord- After this failure, Roberts asks the students to ing to Clayton Christenson (1997), many disruptive try again. Once again, when they come together and innovations arise from this source. report their proposals, the group judges no solution satisfactory. This time their proposals involve various forms of competition: the best prevails in COLLABORATION IS NOT OUR some sort of contest. For example, the green and FIRST CHOICE anti-green advocates both present their cases to the public, who vote on referenda to adopt one scheme Given the discordant nature of a mess, it would after a period of debates and campaigning. seem obvious that collaboration is essential to Roberts sends the students back to try a third make progress. Otherwise the different groups will time. In their frustration over their recalcitrant continue the stalemate of mutual opposition. Yet, instructor they start meeting as a group. They when faced with a messy problem, most people do discover they can invent solutions that take care of not automatically fall into a mode of collaboration. multiple concerns. Together they find a solution to Our colleague, Nancy Roberts, has confirmed this the wicked problem. from her work and uses it to teach a class on “coping Roberts notes that they eventually got to col- with wicked problems”(Roberts 2001). laboration, but not before they had exhausted the alternatives of authoritarianism and competition. These two approaches do not work because they do not show how individual concerns will be taken Figure 1. The mess and its observers. The transfor- care of. Roberts observes, as did Winston Churchill, mational event marks the beginning of adoption of “People fail into collaboration.” new practices that eventually resolve the mess The situation in the US after Hurricane Katrina in August 2005 followed this pattern. The wicked problem was to restore infrastructure in a region where most of the residents had permanently fled after the storm knocked out all power, communications, water, transportation, food distribution, sewage, and waste removal. The President’s first proposal (FEMA takeover) was authoritarian. Local authorities asserting regional rights rebuffed that approach. Thereafter, the situation devolved into numerous competitions (including disputes and fin- ger-pointing) between federal and local jurisdictions.

719 Resolving Wicked Problems through Collaboration

Two years after the disaster, the region remained think of other groups as obstructionist. gridlocked by local rivalries, fewer than half the 3. Messes may be intractable because many residents had returned, disaster reimbursements groups in the social system have enough power were held up by enormous tangles of red tape, and to block action they dislike but insufficient very little rebuilding had even started. Most of the power to forge consensus around action they progress that was made came from the grass-roots favor. Resistance from disaffected groups is level, such as businesses, churches, voluntary as- a major obstacle to change. sociations, and neighbors. 4. Resolution requires a transformation of think- So the political system tried and failed at ing and practice in the community. authoritarianism and competition and got stuck, 5. Given the differences of opinion about solu- while the grass roots fell into collaboration and tion approaches, no resolution will be attained made progress. The political system, in its desire without developing a sense of solidarity in the to manage everything, did little to empower the community. grass roots. The Katrina mess is one of many where 6. Someone has to take the lead to promote grass-roots movements have outperformed govern- adoption of a solution. ments. There is a worldwide movement to empower local grass roots groups for humanitarian assistance These six themes suggest the following six prac- (Hawken, 2008). tices for resolving messes. These practices extend We are not saying that authoritarian solutions or the eight foundational practices of all innovations competitive solutions never work. Of course they (Denning and Dunham, 2006). do. They tend not to work for wicked problems because authoritarian solutions provoke resistance 1. Declare. Begin by declaring that you see a and competition produces local winners at the ex- mess and intend to do something about it. Your pense of the whole. Our familiarity with these two declaration is needed because many people approaches draws us to them first. Roberts is saying find the mess to be normal and see no point that when we encounter a wicked problem, we are in changing or fighting it. Your declaration most likely to find a solution by going straight to will mobilize others who may be willing to collaboration. join you in the struggle. Clearly it will take some work and practice on 2. Learn. Few people appreciate the full com- our part to understand how collaboration works and plexity of the mess. Most see only their parts how to achieve it. and think of other groups as obstructionists. Do not take sides. Instead, make yourself a student of the mess; learn everything you can PRACTICES FOR RESOLVING THE about it; become an expert on the mess. Read MESS what has been written, talk to people about what they know, and perform experiments. What form shall the collaboration to resolve a mess When you accomplish this, you will see pat- take? It will be shaped around six themes running terns that no one else has seen, which may through all the examples of messes: help you lead the stakeholders to a resolution. Becoming an expert is challenging because 1. Many people in the system see the mess as many people are unable to articulate all their normal. They are resigned to the apparent concerns: you must listen for what is not said impossibility of change. as well as what is said. 2. Few people in the system see the full complex- 3. Blend. This is Terry Pierce’s advice (Pierce ity of the mess. Most see only their parts and 2004). Your proposed innovation to resolve the mess is certain to be resisted. Many groups in

720 Resolving Wicked Problems through Collaboration

the social system have enough power to block 6. Lead. All the declarations, learning, question- action they dislike but insufficient power to ing, and thinking will come to naught unless forge consensus around action they favor. You someone steps up to lead the change. If that will probably have to use politics and media is not you, you had better convince someone to forge healthy consensus and keep large else to do it. The primary work of the lead- numbers of people involved in the new game ership is provoking people to question and until they embody it. You want a critical mass learn, facilitating collaboration, and managing of people to buy in to the innovation before the large-scale coordination. resistance solidifies. Think of Amazon.com and iTunes as examples; they blended with the These practices are not linear “steps” to be per- copyright protection interests of traditional formed. They are areas of action and skill. They are publishing by looking like an on-line version performed in parallel. of a conventional store, but they wound up Because multiple stakeholders are involved disrupting the traditional publishing houses throughout, collaboration is an essential ingredient by allowing authors to self-publish through of all six practices. Collaboration is a synergistic them. coordination in which the collaborators create new 4. Question the paradigm. The “paradigm” observers, new possibilities, new futures, and new is the belief system in which everyone is concerns (London 1995, Straus 2002). It is the only operating. The existence of a mess is strong way that the stakeholders will come together, come evidence that the paradigm is not able to resolve to understand the nature of the mess, blend with the the problem, and in fact may be the cause. resistance, transform their thinking and practice, Therefore, try to identify all the assumptions develop solidarity, and coordinate their resolution- in the belief system and diagnose which are generating actions. questionable in the current situation. Pay Collaboration does not mean consensus. Consen- special attention to anomalies; they reveal the sus is the enemy of collaboration. Consensus means limitations of the paradigm. Looking outside to make a minimally disagreeable compromise that the current paradigm is quite difficult because may be unsatisfying to many but not so bad as to most stakeholders don’t know what “outside” provoke serious opposition. It is a lowest common looks like; they lack the language to discuss it denominator. It appeals to a deep resignation about or even think about it. Synergistically combin- the mess, a feeling that the status quo is better than ing their multiple perspectives is the way to an attempted change. Consensus narrows possibili- overcome this blindness. ties; collaboration opens possibilities. 5. Develop a “we”. Bring together representatives of all the different views and interests in the system, who are willing to talk it through COLLABORATION PRACTICE together. (Nancy Roberts calls this “Getting the system into the same room”.) Lead them to Let us now examine how to organize a collaboration experience solidarity by helping them generate that may resolve a mess. new observers of the mess and new possibili- First, let us acknowledge that organizing groups ties for resolving it. Chances are that the group who have been in opposition is not an easy task. will see something together that no individual Lewis Perelman cites infrastructure renewal as a saw alone. They may find a new perspective messy problem involving the clash of “green” and that the various power centers can accept and “blue” agendas (Perelman 2008). Green represents move with. In other words, collaboration may the sustainability movement, which aims at envi- find a solution where serendipity, coercion, or ronmental protection and resource efficiency; its competition cannot.

721 Resolving Wicked Problems through Collaboration

main concerns include energy-neutral designs for pants and designs action teams to address the mess. buildings and other infrastructure. Blue represents The sponsor coordinates the follow-through by help- the security movement, which aims to protect against ing the teams find people, allies, and resources. attacks and disasters; its main concerns include criti- At the workshop, the facilitator leads the group cal infrastructure. The various players do not agree through five stages that culminate in collaboration on the relative importance of the two perspectives. and the design of follow-on projects. These stages Each perspective reaches different conclusions about implement portions of the six practices needed to infrastructure renewal and best use of resources. bring the participants into collaboration on the design The struggles among the perspectives often of follow-on actions. lead the opposing groups to distrust and resent one another. Perelman notes that in the infrastructure 1. Declare: The sponsor declares the question for issue, blue and green advocates tend to avoid each the group to consider. The question emphasizes other. When they do make contact, their interac- new possibilities rather than current deficits. tions often end with legal battles, such as the one in Each group member declares acceptance of California between the Navy (wanting to test new the need or desire to work together on the is- sonar systems) and National Resources Defense sue, and openness to the perspectives of the Council (wanting to protect marine wildlife). The others. Without the agreement of everyone in opposing groups also form political movements the group to cooperate with the process, egos that try to influence public opinion in their direction can get in the way and hijack the process. and impose a solution at the ballot box. In such an 2. Connect: The members take time to become atmosphere, it is difficult get the parties talking about present and engaged with each other. They collaborating. They fight over a choice between a say what concerns bring them to the gather- “blue space” and a “green space” but do not work ing. They say their aspirations and what is at together to create a combined “blue-green space” stake for each of them. They say why they (Denning, 2002). see a need for collaboration. They look for Recent experience at the grass roots is more and acknowledge connections such as mutual optimistic. People are tired of failed public projects friends, business interests, or education. in parks, development, affordable housing, climate 3. Listen and learn: Now the group speaks change, and infrastructure renewal. They are turning and listens, as openly as possible, to the to facilitated processes that guide them to collabo- concerns motivating each member on the ration. Prominent examples include Appreciative issue. The goal is to expose all the concerns Inquiry (Barrett 2005), Straus Method (Straus 2002), and learn how and why each matters to some and Charrettes (NCI). These successful methods member. Members tell stories showing how have a common structure, exhibited in Figure 2 concerns affect their worlds. For example, (Denning 2008). “Low wattage light bulbs matter to me. My The sponsor is a credible entity who declares company replaced a thousand incandescent the mess and convenes stakeholders to engage in bulbs and saved $5000 on our electric bill in the six mess-coping practices. The sponsor invites the first year. That’s a lot of cash for our little the design team to propose a question for inquiry company.” The listening must be open and and an invitation list to a collaboration workshop. inclusive—seeking to gather many different All key players, generations, and interests must be perspectives, and avoid any initial judgment represented at the workshop. The sponsor provides a that one is better than another. Conversation facilitator for the workshop and leads them through is for clarification—not justification or argu- the mess-coping strategies. The workshop fosters a ment. Comments beginning “What if...” and sense of community—a “we”—among its partici- “I wish...” fit, but not “That won’t work.” This

722 Resolving Wicked Problems through Collaboration

Figure 2. Structure of messy problem solving

A sponsor declares the mess and its intent to resolve it. The sponsor facilitates a three-stage practice where (1) a design team creates the agenda and broadly-representative invitation list for (2) a facilitated workshop that fosters collaboration among the representatives, and (3) follows through with action teams carrying out projects designed at the workshop. The workshop facilitator leads the group through five stages, building their collaboration

stage is complete when no one has any further and take care of their interlocking concerns. ideas to express; everyone appreciates that Occasionally, the mess will evaporate in the the group has multiple concerns to consider; light of the reconfigured concerns of “we”. many may see a common core of concerns 5. Design: Now the group engages with the ac- the group can work with. tual work of creating projects. Some will be 4. Promote “we”: Members of the group con- variations of the tentative earlier proposals, tinue the conversation about what matters others new. Members offer to lead projects; for as long as necessary until they develop other interested parties join the project teams. the experience of a “we”. This is the hardest Projects addressing multiple concerns are the part. The early signs of group identity and most likely to attract teams. The facilitator solidarity are members making tentative guides members with doubts about a proposed proposals that recognize, respect, and even project to question in a “we” mood of explo- own the interests and concerns of the other ration, clarifying objectives and exploring members. A later sign is reconfiguration of consequences. For example, instead of saying, concerns—for example, someone favoring “This project cannot work,” the member could authoritarian, protective, anti-terrorist gov- ask, “In my experience the resources to do this ernment might reconfigure into a concern will be considerable. Can we reformulate in a for strong, safe, resilient community. The less expensive way? All proposals that attract facilitator keeps the proposals tentative and sufficient teams can move forward for action. the mood exploratory. The conversation will The group’s final agreement on projects to take evolve into a shared feeling that we are all in forward cements its solidarity and service the same mess together, and by staying together to a larger cause. The mood of this stage is we can resolve the mess. The mess may start ambition. to unravel as the members become aware of

723 Resolving Wicked Problems through Collaboration

Throughout these five conversations, the facilita- We will discuss them below. We will also discuss tor maintains a background mood of appreciation emergent systems of practice that appear in the and openness. Openness encourages everyone to conversation spaces opened by these technologies, contribute ideas and disclose concerns. Apprecia- and the design considerations for technologies that tiveness invites creativity. This is the hallmark of support true collaboration. the highly successful Appreciative Inquiry process Other discussions of work-supporting technolo- (Barrett 2005). The contrasting mood of problem- gies use different categories of behavior: information fixing tends to be narrow; it focuses on what’s sharing, coordination, cooperation, consensus, col- wrong rather than what could be; it discourages laboration, and collective action. These categories group solidarity. are usually stated as various degrees of control The facilitator also displays all new points over behaviors of people in a group. It is often learned, proposed, or created on shared computers difficult to classify a given technology into one of or wall posters. David Straus has found that this these categories. Our three categories make more form of group memory helps everyone recall ideas explicit the kinds of conversations and practices a belonging to the group as a whole (Straus 2002). technology supports. Consider what might happen if this process were A caveat: These technologies are only able to applied to Lewis Perelman’s blue-green clash cited help users who are willing to enter the conversations earlier. Suppose that a group of green and blue in- they support. If the opposing groups in messes try frastructure advocates decide to collaborate together to avoid each other, these technologies will be of despite the clash between their perspectives. Their little help. facilitated discussion might evolve as follows. They discover that some of their members are motivated Exchangers green because beloved family members succumbed to lung diseases. They discover that others are Exchangers support the sharing and transfer of in- motivated toward security because their businesses formation. The most common examples include: have been robbed at gunpoint and because one of their companies went out of business in a blackout. • blog They discover that all of them are hesitant to back • chat a centralized government solution because of the • content streaming government’s poor track record; they do not want • corporate directories to risk locking in a bad solution. They start specu- • database sharing lating about grass-roots solutions that make it cool • discussion board and fashionable to be both green and secure. They • document sharing agree on committees and working groups that will • email sponsor contests for well-designed energy-efficient • file servers products and stimulate research into personal home • instant messaging power plants that don’t depend on the grid being • live presentation operational all the time. • personal computer access • personal info sharing • photo sharing (e.g., Flickr) TECHNOLOGIES • recording • remote blackboard Over the past several decades many impressive • RSS “collaboration technologies” have become available • screen sharing in the Internet. They can be grouped in three main • version control systems categories: exchangers, coordinators, and games. • remote screen sharing

724 Resolving Wicked Problems through Collaboration

• VoIP Games • VPN A game is a system of interactions among players Coordinators seeking to achieve a specified outcome through their play together. The players are free to make individual A coordinator contains a workflow representation choices as long as they abide by the game’s official of the network of commitments of a group and a rules and standard strategies. Besides sports and means to observe when participants make new com- entertainment games, many social systems in po- mitments or move existing commitments closer to litical science, social science, economics, business, completion (Winograd and Flores, 1987). It allows biology, ecology, computer science, and psychology only those actions that align to the workflow. It have been interpreted as games. Games bring the tracks workflows and can answer questions about players into a set of shared practices, at which they workflow states. Examples of coordinators: develop skill by repeated play. Many on-line games have been developed that • auction system allow people to assume roles in the games. The • business process manager current generation of these technologies is the • classroom manager “massively multiplayer online role playing game” • collaboratory (MMORPG). A previous generation was called • concurrency controller “multiuser virtual environment” (MUVE). Examples • coordinators of these technologies include: • Coordina • .00.tor email system • America’s Army • creation net • Active Worlds • decision support system • Dungeons and Dragons • discussion forum • Flight simulator • interactive voice recognizer • River City • Internet protocols • Road simulator • network meetings • Second Life • newsgroups • SimCity • online payment system • Socially beneficial games • operating system • There • project manager • Training games • shopping cart • World of Warcraft • service oriented architecture (SOA) • social network systems (MySpace, Facebook, The list mentions “socially beneficial games”, LinkedIn, etc.) a class defined by Luis von Ahn (2006). Socially • support center beneficial outcomes are a side effect of their regular • telescience (remote lab) play. In esp.com, for example, random pairs of play- • voting systems ers try to label images with keywords by guessing • wiki discussions the keyword the other player will use. Players have • Wikipedia fun and accumulate points and national ratings. The • workflow manager side effect is that the images get good keyword descriptors, which helps search engines find them. Von Ahn says that these games mobilize brains to do computations that we do not yet know how to program into computers.

725 Resolving Wicked Problems through Collaboration

Game theorists study the same kinds of games practices in a virtual world, that can be taken later analytically. They introduce payoff matrices that into the real world. award points to pairs of players according to the Impressive systems of practice often arise around kinds of decisions they make in encounters during the simplest technologies. The Faulkes Telescope the game. They seek to discover, through analysis (www.faulkes-telescope.com) is a facility that pro- and simulation, which strategies produce the highest vides free access to robotic telescopes and an educa- long-term payoffs for the players. Robert Axelrod, tion program to encourage teachers and students to for example, learned that large-scale cooperation is engage in research-based science education. John likely to evolve in a social system if the players use Hagel and John Seely Brown (2006) see this as a a “tit for tat” strategy when they interact (Axelrod fine example of a creation net, a (possibly collabora- 1984); tit-for-tat means that, in your next encounter tive) community that learns and invents together. with a person, match that person’s last move of Clay Shirky (2008) gives numerous examples of cooperation or non-cooperation. groups coming together spontaneously in a cause or Two aspects of these games make them very movement using the simplest of information sharing interesting for innovators. First, when people join tools. Thus, a community practice can flourish even these games, they quickly become immersed in if no technology has been specifically designed to the practices of the game. The virtual world of the support it. game becomes “reality” and they often forget they Sometimes these impressive systems are cap- are playing in a game. This aspect of games makes tured into a new technology. Technologies that them very useful for training. The players learn how support the work of organizations illustrate this. to perform in the real world by developing their skill Initially, organizations used simple information in a simulation of that world. sharing tools such as email and online records to The second interesting aspect for innovators is support their work. Workflow management systems that people tend to develop trust for their fellow were invented later to support standard practices, community members in the game. This happens such as mapping interactions among roles and because they share in the same practices, giving tracking the commitments made by persons in those them a strong connection. roles. More recently, multiplayer role-playing games Many innovators work with a game interpreta- have been invented to enable users to practice the tion. They begin by interpreting the mess as the dynamic creation of workflows and roles. result of a game. Through careful examination of the history of attempted innovations, they find the Designing to Support Collaboration rules and strategies of the game. Then, in the question-paradigm practice, they speculate about how Collaboration is an emergent practice. There is as they might change the game so that the mess will yet no category of technology that fully supports disappear. The multiuser role-playing game may be the collaboration practice we described earlier. a useful tool to develop the new practice in a virtual People learn the collaboration practice in various world and then propagate it to the real world. ways including coaching and immersion in an already-collaborating community. Once they know Emergent Systems of Practice the practice, they carry it out with the help of exchangers, coordinators, and games. Each category of technology explicitly supports Designers of systems to support collaboration certain practices. Exchangers support practices for use a three-part strategy: sharing and exchanging information, coordinators support the practices of a community in doing their 1. Declare the unifying principle or theme for work, and games support people in learning new the collaboration,

726 Resolving Wicked Problems through Collaboration

2. Interpret the social system as a network of conversations, commitments, and practices, • repositories (articles, case studies, pre- 3. Assemble a suite of tools that enable con- prints) versations, enact commitments, and support • digital libraries (access to ACM, IEEE, wiki- individual practices in the network. pedia, etc.) • real time communication such as teleconfer- The users can then use the tools to participate in ence the process and achieve the unifying purpose. • internet connected blackboards The collaboratory is a social process designed • community discussion boards, RSS feeds, in this way. A collaboratory is a virtual center that blogs, and wikis supports collaborative scientific research among • distance learning systems geographically distributed researchers (Wulf 1989, • systems to record group results at collabora- 1993). The collaboratory is envisioned as a means tion workshops (charrettes) to solve complex natural system (Category III) • coordinator systems to track follow-on projects problems, but not wicked problems (Category IV). after the summit workshop Wikipedia reports at least nine collaboratories in various fields. Some have been successful, others not. The overlap with collaboratory systems is strik- They each have a social model for their community ing. and have selected tools to facilitate research in that Chauncey Bell (2005) describes a coordina- community. Despite their differences of purpose, tor system for financial management that could all the collaboratories employ similar technologies be adapted for supporting follow-on projects. His (Bly 1998): system recognizes three roles: proposer, investor, and manager. It provides tools that support the main • repositories (technical papers, preprints) actions of each role. It enables sophisticated report- • digital libraries (access to ACM, IEEE, wiki- ing on the status of investments and the expected pedia, etc.) returns. • real time communication such as teleconfer- In time, we will be able to design additional tools ence that will help facilitate collaboration and extend its • Internet connected blackboards reach into larger communities. • community discussion boards, RSS feeds, blogs, and wikis • distance learning systems LIMITATIONS OF THIS STRUCTURE • remote instruments • remote data collection and analysis It is doubtful that the process of Figure 2 could ever • integration with supercomputers and grid be fully automated and the facilitators sent home. computing The facilitator’s main job is to manage the mood of the group, maintaining a sense of appreciation With our model of the social collaboration pro- and moving toward the experience of solidarity (the cess for a mess (Figure 2), we can infer the kinds of “we”). Building computer systems that monitor tools that would have to be assembled into a mess- and manage moods is hard. Chauncey Bell (2005) resolving center. The National Charrette Institute, points out that financial management systems, which has developed a suite of web-based technology which record every commitment and every action to help their clients with architectural design and leading to its fulfillment, enable auditors to make infrastructure issues, perhaps comes the closest to powerful inferences about participant moods and this goal. The most useful tools are: probable wrongdoing.

727 Resolving Wicked Problems through Collaboration

How far does the messy problem collaboration issue. There is no sense of a “we” to work process scale? We know that it works for workshop- together on the issue. It must be developed size groups (say 50-200 people). It extends to larger through collaboration. communities if the workshop represents them well 2. Falling into authoritarian or competitive and if the sponsors can support the follow-through approaches. Authoritarian and competitive teams created by the collaborating group (London strategies often fail with messy problems. 1995). What about messy problems that affect mil- Without everyone’s coming to a mutual lions of people? How do we bring about enough understanding of all the other concerns and collaboration to influence so many? interests, and learning together, it is unlikely This of course is the central question in efforts that a design will be found that wins enough to deal with large-scale wicked problems such as acceptance to resolve the mess. sustainable infrastructure, global warming, or health 3. Trying to do it alone. Messes cannot be re- care. We don’t yet know how to make the collabo- solved without collaboration. Moreover, the ration process scale up to enlist millions of people action teams will require experts in various in a solution. Currently, problems of such scale areas including technology, media, social is- tend to be resolved by strong leaders who combine sues, and politics. technology with political and media savvy to inspire 4. Technology-only solutions that do not ad- collaboration. For example, Candy Lightner and dress the social issues. The mess is a social Cindy Lamb established Mothers Against Drunk issue even if originates with technologies. Driving (MADD) as an international movement. US Considerable collaboration-building is needed Senator George Mitchell established the “Mitchell to bring about social agreement. Principles” that created a workable framework for 5. Being unprepared for resistance. It is a dialog that ultimately led to the peace agreement mistake to discount the resistance that will in Northern Ireland. Amory Lovins, who focuses surely come from stakeholders who see no on technical facts and avoids moral judgments, benefit in the proposed solution. has helped clients as diverse as Wal-Mart and the Department of Defense deal with energy issues. CONCLUSION

COPING WITH FAILURES Messes are intransigent social situations that people want to exit but feel stuck in. While some messes The social process depicted in Figure 2 does not may be irresolvable, we can often find ways out of always lead to a solution of the mess. There are five messes through six basic practices. Collaboration common failures that the participants must cope is at their core. with as the process unfolds. Professional facilita- Collaboration is a practice of creating new ob- tors, who are trained to cope with these failures, servers and new possible actions together. Through significantly improve the odds of success. The five collaboration, a community creates a solution to a failures are: messy problem that takes care of all their concerns at the same time. Collaboration does not mean that 1. Not developing a shared interpretation of community members give up or comprise their dear- the problem. It is easy to blame the obvious est concerns. It means they design a solution that lack of consensus on obstructionists giving recognizes their concerns. The process often leads preference to their own interests over the to a reconfiguration of everyone’s concerns. The common good. But the lack of consensus on hallmark of successful collaboration is the experi- problem definition or approach is the central ence of solidarity and new energy: a “we”.

728 Resolving Wicked Problems through Collaboration

Collaboration is an ideal achieved far less Denning, P. (2007). Mastering the Mess. ACM often than it is invoked. Many people are drawn Communications, 50(April), 21-25. to more familiar authoritarianism or competitive Denning, P., & Yaholkovsky, P. (2008). Getting to strategies—which generally do not work for messy “We”. ACM Communications, 51(April), 19-24. problems. Collaboration is often confused with information sharing, consensus, cooperation, coor- Einstein, A. (1916). Relativity. dination, or collective action. Most “collaboration Hagel, J., & Brown, J. S. (2006). Creation Nets: technologies” are actually tools for information Harnessing the Potential of Open Innovation. sharing. The design strategy for tools is, first, un- Working paper; johnseelybrown.com . derstand the social process and, second, assemble a suite of tools to support the process. Hawken, P. (2008). Blessed Unrest. Penquin, reprint As we learn more about collaboration practice edition. and tools to support the collective actions of collaborating communities, we will be able to extend Kurtz, C. F., & Snowden, D. J. (2003). The New the known collaboration processes to much larger Dynamics of Strategy: Sense-making in a Complex scales, perhaps even to country or world sizes. and Complicated World. IBM Systems J., 43(3), Their designs will be based on deep knowledge of 462-483. the practices now used by the human facilitators of London, S. (1995). Collaboration and Community. today’s processes. An essay prepared for Pew Partnership for Civic Change (November); www.scottlondon.com/reports/ppcc.html References NCI, the National Charrette Institute; www.char- Axelrod, R. (1984). The Evolution of Cooperation. retteinstitute.org . New York: Basic Books. Perelman, L. (2008). Infrastructure Risk and Barrett, F., & Ronald, F. (2005). Appreciative In- Renewal: The Clash of Blue and Green. Work- quiry. Taos Institute. ing paper PERI Symposium; www.riskinstitute. org/PERI/SYMPOSIUM Bell, C. (2005). Wise organizations. In J. Courtney, J. Haynes, & D. Paradice (Eds.), Inquiring Organiza- Pierce, T. (2004). Warfighting and Disruptive Tech- tions (pp. 229-271). Idea Publishing Group. nologies: Disguising Innovation. Frank Cass. Bly, S. (1998). Special Section on Collaboratories. Reeves, B. N., & Lemke, A. C. (1991). The problem ACM Interactions, 5, 3. as a moving target in cooperative system design. HCI Consortium Workshop (January). Burke, J. (1995). The Day the Universe Changed. Back Bay Books. Roberts, N. C. (2000). Wicked Problems and Net- work Approaches to Resolution. The International Christenson, C. (1997). The Innovator’s Dilemma. Public Management Review, 1(1). Harvard Business. Roberts, N. C. (2001). Coping with Wicked Prob- Denning, P. (2002). Flatlined. ACM Communica- lems. In L. Jones, J. Guthrie & P. Steane (Eds.), tions, 45(June), 15-19. International Public Management Reform: Lessons From Experience. London: Elsevier. Denning, P., & Dunham, R. (2006). Innovation as Language Action. ACM Communications, 49, 5 (May), 47-52.

729 Resolving Wicked Problems through Collaboration

Shirky, Clay (2008). Here Comes Everybody. New Collaboration: a practice of working together York: Penguin. with others to produce new observers and new possibilities that no one could produce alone. Straus, D., & Layton, T. (2002). How to Make Col- laboration Work. San Francisco: Berrett-Koehler Collaboration Technology: information tech- publishers. nology that supports the practice of collaboration. See collaboration. Winograd, T., & Flores, F. (1987). Understanding Computers and Cognition. Addison-Wesley. Collective Action: people coordinating together inside a game, producing some result in the sum Wulf, W. (1989). Toward a National Collaboratory. total of their actions that cannot be seen from any Unpublished position paper at an NSF workshop, individual’s action. Rockefeller University, New York. Disruptive Innovation: a change of practice Wulf, W. (1993). The collaboratory opportunity. in a social system that requires new thinking, new Science 261(13 August), 854-855. beliefs, an alteration of the roles and their connec- Wikipedia. Article on Collaboratory; en.wikipedia. tions in the social network, and shifts of power org/wiki/Collaboratory among groups in the social system. Contrast with sustaining innovation, which means an improve- ment of performance in existing practices of a social system. KEY TERMS Exchanger: information technology that shares Consensus: people reaching an agreement for or transfers information among members of a group; action that is unsatisfying to many in the group but supports collaboration. not so bad as to provoke serious opposition. Game: a set of rules by which members of a Cooperation: people working together to social system interact to achieve some purpose achieve a common purpose. together. Coordinator: information technology that helps Mess: a tangled social situation that is too costly people move within a network of commitments, by to stay in and too intransigent to get out of. See also recording when they make commitments and track- wicked problem. ing their progress toward completion. Wicked Problem: a tangled social situation that is too costly to stay in and too intransigent to get out of. See also mess.

730