Wearable communities: augmenting social networks with wearable computers

1536-1268/03/$17.00 © 2003 IEEE � Published by the IEEE CS and IEEE Communications Society PERVASIVEcomputing 71

Wearable Communities:Augmenting Social Networkswith Wearable Computers

Wearable computers enhance per-sonal computing with continu-ously worn, intelligent assistantsthat augment memory, intellect,creativity, communication, and

physical senses.1–3 As wearable computing technol-ogy pervades our daily lives, we must ask how itinfluences social behavior. Will it merely make peo-ple “smarter” by providing seamless, context-awareaccess to information, or will it also enrich theirsocial interactions?

To advance the wearable com-puting paradigm for personalcomputing, we must carefully con-sider this technology’s social mech-anisms, potential, and constraints.Some important research ques-tions we must consider include

• What social potentials do wearable computershave?

• What system requirements and characteristics willrealize these potentials?

• How can we systematically and effectively buildsuch systems?

• What are this technology’s social impacts?

The preliminary state of wearable computingtechnology and the low number of systems in activeuse keeps us from satisfactorily answering all thesequestions. To advance the wearable computing par-adigm, however, we must start investigating these

issues and develop a socially acceptable vision ofwearable computing. This article presents a frame-work for wearable technology support for socialnetworks. Our WearCom wearable communitydesign methodology facilitates application creationand provides a framework for investigating thesocial and technical issues involved.

Wearable communitiesWe use the research framework of computer-

mediated communities4–6 to investigate wearablecomputers’ social potential. Wearable communitiesdenote the social networks that might emerge whenenough people use wearable computing technologythroughout their daily lives.7 Using a particular tech-nology, namely wearable computers, defines such acommunity just as Internet use defines an onlinecommunity. Like online communities, wearablecommunities are multiparty conversations organizedaround affinities and shared interests, bringingtogether people who don’t necessarily know eachother personally. Unlike online communities, how-ever, they are based on embodied, real-world humanencounters augmented by wearable computers.

The combination of wearable computing devicesand wireless personal area networks presents inter-esting opportunities.8 WPANs such as Bluetoothenable seamless ad hoc communication over short-range radio links (up to 10 meters) and let people’swearable computers communicate during face-to-face encounters. This permits new forms of sponta-neous social interactions among collocated people.

Wearable communities—social networks based on computer-augmented face-to-face encounters—present both social and technical designchallenges. Our WearCom design methodology permits rapid prototypingof wearable community systems to facilitate community building.

W E A R A B L E C O M P U T I N G

Gerd KortuemLancaster University

Zary SegallUniversity of Oregon

Unfortunately, wearable computingresearch generally ignores the crucial inter-play between technology and social behav-ior. Current systems and applicationsemphasize intellectual and sensory capa-bilities over interpersonal interactions andsocial competence. The danger of this trendis that wearable systems may end upinhibiting interaction. We’ll eventually over-come some of the potentially negative socialimplications that stem from their currentstate (such as bulky head-mounted dis-plays), but others are inherent to the wear-able computing paradigm. By attending tothe device and the information it presentsthem, wearable users have less time andattention to offer people nearby. Suchbehavior can negatively influence outsiders’views of wearable computers and their use.

Augmenting social spaceThe idea of wearable communities is

based on the notion of augmented socialspace, where WPANs generate a sphere-likedigital field that envelops a wearable com-puter and its user (see Figure 1). Social spacedefines the space in which humans interact.Edward Hall suggests four distinct inter-personal communication proximity lev-els—intimate, personal, social, and publicspace9—each defining a region around thebody with invisible boundaries determinedby the body’s spatial characteristics, cul-tural conditions, and social relationships.

Touch happens in intimate space, conver-sation in personal space. Groups of peopleset up social distances, usually 10 feet orless; public space extends beyond this.

Information broadcast by the wearablecomputer fills a digital social space whosesize depends on the wireless transceivers’transmission range and can range from afew inches to several feet. The human eyecan’t see the digital field, but the wearablecomputers can. When two computers’ dig-ital social spaces overlap, they becomeaware of each other. We see the digital fieldas the alter ego of human social space: justas the concept of social space describes fac-tors that influence interpersonal activities,the concept of digital social space describesfactors that influence interactions betweenwearable devices. Interactions in the socialrealm can initiate interactions in the digi-tal realm, and vice versa.

Based on the notion of augmented socialspace, we define a wearable community asa social network that people create or main-tain using wearable computing devices. Acollection of users becomes a communitywhen enough people use their wearablecomputers to form webs of personal rela-tionships. This notion of wearable com-munities acknowledges and is based on theunique value of random encounters andface-to-face interactions. We believe com-munity building depends on fully embodied“human moments.” While the Internet and

online communities have separated physi-cal place and social space, our work onwearable communities tries to reunitethem—rather than replacing face-to-faceinteractions, we intend to augment them.

For example, when two people convers-ing exchange electronic business cardsbetween their PCs, social interaction leads todigital interaction. On the other hand, inter-actions between wearable devices can facil-itate or even augment social interactions.For example, wearable computers caninform their users about another person’spresence nearby, autonomously exchangeinformation, quietly gather informationabout encounters, or suggest a conversationtopic of mutual interest. They might evenspeculate and inform the user about affinityrelationships from repeated encounters.10

ApplicationsTo illustrate the interactions that might

take place within a wearable community, wedescribe several implemented applications.

Genie. We designed this simple applicationto improve knowledge exchange within adispersed group of wearable computerusers—for example, students attending thesame college. Genie makes the group’s com-bined expertise available to every groupmember. Some might be soccer experts, oth-ers might be hobby gardeners with exten-sive horticultural knowledge; but someonewho needs to answer a particular questionoften doesn’t know who can help them.Genie members use their WPAN-equippedwearable computers to automatically findfellow community members who can pro-vide just-in-time expertise.

Using Genie, each user defines a set ofquestions that are stored on the wearabledevice. Whenever two or more wearableusers meet, their computers exchange ques-tions. This might happen during informalchance encounters occurring throughout aday, for example, during lunch or at a cof-fee shop. After receiving a question, thedevice displays the question and asks the

72 PERVASIVEcomputing http://computer.org/pervasive


Human interaction

Digital interaction Wearable computerWearable computer

Figure 1. Augmenting social space withwearable computers.

user whether he or she knows the answerand is willing to provide it to the other user.Genie relays the response from one com-puter to the other in real time. The systemtransmits only an indication of a person’swillingness to engage in a conversation, not the answer itself. After a successfulexchange, the users can approach eachother and discuss the topic of interest faceto face. To help users identify each other,the computers also exchange personalinformation such as user names or photos.

Although simple in many respects, this example illustrates many importantaspects of wearable communities. First,interactions in wearable communities aresituated and involve a rich social context:users are aware of whom they are inter-acting with and thus can observe impor-tant social cues such as sex, appearance,and gestures. The addition of social con-text shapes people’s willingness to engagewith strangers and the particular mannerin which they interact.

Second, wearable communities may ormay not be built on top of existing socialnetworks. In particular, wearable commu-nities can facilitate face-to-face interactionsbetween strangers. The only prerequisitefor a wearable community is a group ofwearable users willing to cooperate throughautomated interactions of their respectivewearable computers, regardless of pre-existing social ties.

Third, you cannot implement wearablecommunity applications with traditionalmobile devices such as laptops and PDAs.They require devices that are

• Constant: always on and running• Presence-aware: aware of the presence

of nearby devices and people• Communicative: able to communicate

with other collocated devices • Proactive: able to perform tasks auton-

omously and proactively without requir-ing explicit user intervention (althoughinteractivity might also be supported)

This definition is consistent with previousdefinitions of wearable computers1–3,11 butadds important functional requirements.Specifically, it emphasizes the ability to

communicate with nearby devices andreplaces a generic context awareness witha more specific presence awareness.

Other examples. Genie is a simple appli-cation that encourages direct, face-to-faceinteraction. We have also built severalapplications in which digital interactionsoccur implicitly—that is, without users’knowledge but with their consent.

The Piraté collaborative music guide12

borrows ideas from peer-to-peer file-sharingapplications such as Napster and movesthem to the wearable domain. Piraté letsusers exchange MP3 playlists and musicrecommendations during brief randomencounters. Serendipitously exchangingand accumulating information throughoutthe day helps users discover new musicbased on the playing habits of those theymeet most often, provides awareness of thecommunity’s favorite music titles, and letsusers discover with whom they share acommon taste in music. In contrast toGenie, members of the Piraté communitydo not engage in direct face-to-face inter-actions: exchanges occur implicitly when-ever two or more people come physicallyclose. Users aren’t aware of interactionsbetween their wearable computers.

The WALID (Wearable AugmentedTask-List Interchange Device) communityapplication digitizes the timeworn tradi-tion of borrowing butter from your neigh-bor.7 You help others because you knowthat one day they will return the favor.With WALID, two people use their wear-able devices during random encounters tonegotiate and exchange real-world tasks,such as dropping off someone’s dry clean-ing, buying stamps at the post office, orreturning a library book. WALID uses per-sonal-agent software to find nearby com-munity members and negotiate the ex-change of favors. The agents maintain auser’s task list, becoming fully aware of the

locations and activities involved. When anencounter occurs, the agents produce anegotiation. If both users approve, a dealis struck. In a negotiation, the agent eval-uates the value of favors and keeps scores:having to run across town just to dropoff someone’s mail compares unfavorablywith buying milk for someone if the gro-cery store is just a block away. Agentsemploy ideas from game theory to ensurethat negotiation results are mutually ben-

eficial: they cooperate only if doing soenhances the users’ goals.

We envision many other wearablecommunities:

• Computing communities: Members sharecomputational resources such as networkbandwidth and computing cycles.

• Helper communities: Members pledgeto assist each other when in distress.

• Bargain-hunter communities: Memberscollectively search out sale items in localstores.

• Marketplace communities: People ex-change goods without money.

• Job market communities: Free applica-tions offer services to passersby.

• Knowledge communities: Members col-lectively accumulate information to cre-ate shared understanding.

• Political communities: Members createnew forms of instant democracy andlocal activism.

From online to wearablecommunities

Wearable communities have precursorsin other types of computer-mediated com-munities. One of the most powerful fea-tures of today’s Internet is its ability toenable total strangers to interact on a per-sonal and sometimes even intimate level.People who have never met (and areunlikely to ever do so) use the Internet todiscuss personal matters related to health,

JANUARY–MARCH 2003 PERVASIVEcomputing 73

The notion of wearable communities

acknowledges and builds on the unique value of

random encounters and face-to-face interactions.

raising kids, romance, and many other top-ics of shared interest. In effect, the Internetprovides the connectivity for the “globalinformation community” that J.C.R. Lick-lider envisioned in 1968 as being “not ofcommon location, but of common inter-est.”13 Howard Rheingold, who coined theterm “virtual community,”14 described theInternet as a place that enables a new typeof social network, mediated by computerterminals and networks. He defined virtualcommunities as “social aggregations thatemerge from the Internet when enoughpeople carry on public discussions longenough and with sufficient human feelingto form webs of personal relationships.”

Technologies such as mobile phones andInternet-enabled PDAs have begun toextend online communities’ reach to mobile

users. For example, online community sitessuch as eBay give users access to commu-nity tools from their mobile devices, lettingmembers stay in touch with their commu-nity at any time from any place. Also,mobile communities are proliferating inwhich interactions between members hap-pen exclusively with mobile devices (mostlythrough SMS messages and mobile phones).Examples of mobile communities includeUpoc, mopilot, and cosmiccupid.

Despite their widespread and successfuladoption, however, mobile phones andInternet-enabled PDAs have serious flawsas enablers of new communities. Theydon’t support many-to-many communica-tion very well and don’t facilitate randomencounters (either virtual or real). Conse-quently, it is difficult to create new rela-

tionships using mobile phones. Moreover,the use of traditional mobile devices canpotentially decrease real-world randomencounters and face-to-face interactions:by attending to people located elsewhere,mobile users possibly reduce the amountof time and attention they pay to strangersnearby. These factors make current mobiletechnologies well suited for maintainingexisting community relationships but lessapt at creating new ones.

How wearable communities differLike online and mobile communities,

wearable communities consist of webs ofrelationships that grow from computer-augmented social interactions. (See thesidebar “Supporting Face-to-Face Com-munication” for current wearable com-



T he notion of wearable communities defines a research

framework for investigating computer support for face-to-

face social interactions. Such interactions are often spontaneous

(driven by chance encounters of mobile people) and situated

(embedded in the real-world context in which people interact).

Several recently developed systems support such interactions.

These include match-making technologies (LoveGety1), awareness

devices to give roaming groups a sense of connectedness (Hum-

mingBird2), ad hoc games that incorporate real-world group mo-

bility into digital entertainment (Pirates,3 Pervasive Clue4), systems

for education and exploratory learning (Geney5), collaborative

augmented-reality systems,6 and messaging devices that adopt

“word-of-mouth” metaphors for proximity-based information

passing (ThinkingTags,7 MemeTags,8 iBalls9).

These systems represent important precursors of wearable com-

munity systems, yet key questions remain unexplored. First, ex-

cept for Richard Borovoy,8 few have investigated this technology’s

social implications over time. Most systems are one-off prototypes

that have seen little or no real deployment, so we don’t really

know how augmenting face-to-face interactions affects personal

relationships and communal behavior. Second, we have no sys-

tematic engineering process for such systems. Developers have no

infrastructures or tools to leverage but must build each system

from the ground up, in a manner dictated by the underlying net-

work technology and device platform. Many systems thus have a

limited feature set and cannot evolve to accommodate design

changes.

REFERENCES

1. CNN, Japan’s Lonely Hearts Find Each Other with “Lovegety,” www6.cnn.com/WORLD/asiapcf/9806/07/fringe/japan.lovegety.

2. L.E. Holmquist, F. Jennica, and J. Wigström, “Supporting Group Collab-oration with Interpersonal Awareness Devices,” Personal Technologies,vol. 3, no. 1, 1999, pp. 13–21.

3. S. Björk et al., “Pirates! Using the Physical World as a Game Board,”Proc. 8th IFIP TC.13 Conf. Human-Computer Interaction (Interact 2001),Laxenburg, Austria, 2001; http://viktoria.informatik.gu.se/groups/play/publications/2001/pirates.interact.pdf.

4. J. Schneider and G. Kortuem, “How to Host a Pervasive Game: Support-ing Face-to-Face Interactions in Live-Action Roleplaying,” Proc. Ubicomp2001 Workshop Designing Ubiquitous Computing Games, 2001; http://www.cs.uoregon.edu/research/wearables/Papers/how2host.pdf.

5. A. Danesh et al., “Geney: Designing a Collaborative Activity for thePalm Handheld Computer,” Proc. Conf. Human Factors in ComputingSystems (CHI 2001), ACM Press, New York, 2001.

6. M. Billinghurst et al., “Experiments with Face-to-Face Collaborative ARInterfaces,” Virtual Reality J., vol. 4, no. 2, 2002.

7. R. Borovoy et al., “Things that Blink: Computationally Augmented NameTags,” IBM Systems J., vol. 35, nos. 3 and 4, 1996; www.research.ibm.com/journal/sj/mit/sectionc/borovoy.html.

8. R. Borovoy et al., “Meme Tags and Community Mirrors: Moving fromConferences to Collaboration,” Proc. ACM Conf. Computer SupportedCooperative Work (CSCW’98), ACM Press, New York, 1998, p. 159.

9. R. Borovoy et al., “Folk Computing: Revisiting Oral Tradition as a Scaf-fold for Co-Present Communities, Proc. Conf. Human Factors in Com-puting Systems, 2001; http://web.media.mit.edu/~borovoy/papers/folk-computing.pdf.

Supporting Face-to-Face Communication

munity examples.) They differ from tradi-tional online communities, however, inthree important ways:• Social context. Wearable communities

are based on direct encounters ratherthan indirect, remote interactions.WPAN exchanges must occur over shortdistances—when people come face toface or are at least within close physicalproximity. Communication partnersthus can usually see whom they’re inter-acting with and observe such importantsocial cues as gender, clothing, and ges-tures. They might even be able to talk toeach other. The addition of social con-text shapes people’s willingness toengage with strangers and how theyinteract.

• Usage context. People use wearabledevices in different contexts than theyuse stationary computers at home orwork. User attention becomes scarcebecause, instead of sitting in front of acomputer where they can pay full atten-tion to the computer and its operation,wearable computer users also engage inreal-world tasks such as driving, oper-ating a machine, or simply conversingwith other people. Time also becomes acritical resource—when surfing theInternet at home or at work, people havemore patience with delays because theycan shift their attention to another task.This no longer holds true when ex-changes occur spontaneously betweenpeople who happen to meet during theirdaily activities.

• Technical context. WPANs offer limitedbandwidth and reliability compared towired networks. Piraté showed us thatwe should therefore design systems toexchange low-fidelity data or meta-information such as song recommenda-tions rather than large MP3 files.

When does a collection of individualsbecome a wearable community?

A community isn’t a perfectly organizedgroup where everyone who participatesgains equally. Among other things, people’s

individual actions primarily serve theirown interests. How people deal with con-flicting interests determines whether a com-munity is successful. We’ve identified sev-eral factors that influence people’s behaviorin wearable communities:

• Identity. Our sense of whom we’re inter-acting with shapes all interactions, eventhose with strangers. Our clothes, voices,bodies, and gestures send messages aboutstatus, power, and group membership.Wearable technology lets people defineand broadcast digital identities thatmight not be their true identities. In con-trast to online communities, these digi-tal identities augment rather than replacea person’s true identity, as both are visi-ble at the same time. Because observerscan match the digital identity to a per-son’s appearance and behavior, people inwearable communities can’t easily switchidentities.

• Privacy. Unlike security, which dealswith keeping information away fromunauthorized users, privacy gives peo-ple the right to control collection and useof their personal information, includingwhen, how, and to what extent infor-mation about them is communicated toothers. Most wearable communitiesrequire members to disclose certain per-

sonal information.• Trust. This aspect of human interaction

differs when people interact in personversus across the Internet. Being able tolook at someone, observe gestures, andmimic them tells you a lot about anotherperson. People place much importancein a handshake when completing a trans-action because it signals commitmentand mutual trust. No equivalent of ahandshake exists in online communities.

Wearable community systemsUnlike online communities, which have

a mature and proven technological founda-tion, wearable communities require com-plex, less mature technologies such as wear-able devices, wireless ad hoc networks,spontaneous networking mechanisms, andcontext technologies. To bring wearablecommunities together, we must build reli-able wearable community systems—collec-tions of wearable devices, distributed soft-ware infrastructure, and applicationsoftware that let people interact with fellowcommunity members (see Figure 2). Systemrefers to software and hardware that set thecontext for interaction, and wearable indi-cates the devices’ wearable nature andimplies that they’re always on, commu-nicative, proactive, and presence-aware.

Within a wearable community system,


W

W W

W

W

W

W

W

WW

W

Figure 2. A wearable community system.

a WPAN that enables seamless connectiv-ity among collocated devices establishescommunication. Thus, a wearable com-munity system is a highly dynamic, looselyconnected, potentially large-scale distrib-uted system made up of wearable hosts. Tosupport the formation of wearable com-munities anywhere, anytime, such a sys-tem must be independent of external com-munication and computing infrastructures,relying solely on device capabilities. Sys-tem- and application-level software per-forms essential functions, such as discov-ering nearby devices and users, initiatinginteractions across wireless links, manag-ing user identities, and handling user inputand output. Because of the environment’scomplexity, developing effective systemspresents a fundamental challenge to build-ing wearable communities.

An exploratory designapproach

In wearable communities, social con-cerns become difficult to separate fromtechnical practices. To design a wearablecommunity, we must simultaneously con-sider technology and social interactionsbecause wearable communities, like allcommunities, result from people’s self-organizing behavior. Relationships emergefrom interactions over time; we cannotconstruct them following a simple recipe.

Successful community design oftenhinges on what Etienne Wenger calls min-imalist design,15 which starts with a pro-visional design and facilitates communitygrowth over time. Designers must identifythe attributes that make communities suc-cessful and design technology that embod-ies these attributes. When building wear-able communities, we face the followingfundamental problem:

Wearable communities requirehardware and software to supportthem, but without real-world experi-ences we do not know what successfactors to look for and how to designsystems to embody them.

Solving the engineering problem and mak-ing it easier to develop wearable commu-nity systems will help us create and inves-tigate these communities.

To solve this problem, we developed theWearCom design methodology.16 WearComsupports an exploratory design approachbased on rapid prototyping of wearablecommunity systems. Rapid prototyping hasproved useful for creating several genera-tions of the Vu-Man wearable computer.17

Our work extends this by taking a system-atic approach to designing an entiresociotechnical wearable computing system.

The WearCom methodology integratessocial and technical concerns and guidesdesigners from scenario development to

implementation. WearCom provides

• A design language that allows the spec-ification of important design decisions

• A design process that outlines an itera-tive sequence of individual design activ-ities, each of which generates a specificdesign artifact (specifications and soft-ware components)

• The Proem software platform,18 whichsupports the implementation and exe-cution of proactive, presence-awarewearable community applications

WearCom specifies three design elements.A user profile—a typed data item thatdefines a user’s identity within a commu-nity—contains information the user dis-closes willingly to other community mem-bers along with that deemed useful ornecessary for community purposes. A com-munity protocol defines how wearable com-puters interact when people meet. Protocolsdeal with communication content, notmechanisms, and in essence formalize ritu-als of engagement. A community agent—an executable software component thatimplements the community protocol—runson the wearable device and functions asintermediary between user and community(see Figure 3).

Case studiesWe performed several case studies to test

our hypothesis that an exploratory designapproach facilitates wearable communitybuilding. Between spring 2001 and spring2002 we taught three software engineeringcourses at the University of Oregon inwhich students used the WearCom method-ology to design and implement wearablecommunity applications. In total, 35 stu-dents divided into 10 project teams workedon four projects:

• mClique, a community system designedto improve awareness among mobileusers by determining groups (cliques) ofmutually connected friends



Agent–agent interaction

Wearable computerwith community

agent

Human–agentinteraction

Human–agentinteraction

Wearable computerwith community

agentFigure 3. Community agents and community protocol in WearCom.

• mBazaar, a wearable community systemthat brings together buyers and sellersof goods

• Piraté, a collaborative music guide thatcreates community awareness (discussedearlier)

• Infomediation, a peer-to-peer informa-tion publishing and access system thatlets collocated users collaboratively com-pare prices at online bookstores

WearCom proved successful in guidingthe student teams’ design activities. Duringthe projects, we tracked data relating tolearning time, development time, complex-ity of developed applications, and supportrequirements. Within 10 weeks, studentslearned to design wearable communities,implement applications, and experimentwith the initial prototypes. It typically tookstudents less than a week to learn themethodology and write their first applica-tion using Proem. After a design phase ofabout three weeks, students took fiveweeks using an iterative process to developapplication prototypes. Teams deployedthe applications on a testbed consisting oflaptop, handheld, and wearable devices.The Proem platform not only simplifiedimplementation but also made it possibleto run applications on diverse hardwareplatforms.

For each project, we tracked supportinstances (how many times teachers aidedstudents) and, on average, counted fewerthan 3 instances per 10-week course. Con-tributing to this surprisingly low numberwas that we had encouraged students toshare their expertise across teams. In sum,we found that the WearCom methodologyeffectively supported rapid prototyping ofwearable community systems.

Despite this success, the overarching goalof building successful wearable communi-ties remains elusive. A true communityemerges over time and with regular tech-nology use, and practical considerations lim-ited our experiences with the applicationsdeveloped. A major obstacle is the lack ofadequate hardware—wearable computersthat are constant, proactive, communicative,and presence-aware aren’t yet available inlarge enough quantities to reach the critical

mass necessary for community building. Thishas limited us to highly experimental sys-tems confined to small groups of people atour lab and has prevented us from system-atically analyzing the communities’ socialdynamics. The projects have, however, givenus insight into what does and doesn’t workwhen building such communities.

Design principles for wearablecommunities

Sociologists have extensively researchedthe successes and failures of online19,20 andface-to-face communities.21 Although wehaven’t empirically studied wearable com-munity dynamics, we have identified sixpreliminary design principles that con-tribute to successful wearable communi-ties. Developers can apply these guidelinesto evaluate existing designs, guide thedesign process, and educate designersabout the characteristics of successfulwearable community systems.

Principle 1: Make users aware of the hid-den benefits of random encounters. Thisprinciple guided design of the WALID com-munity application. When two people meet,they usually aren’t aware of each other’sactivities and, unless they talk about theirrespective errands, will never find out thatthey could benefit from a trade. WALID’scommunity agents look for possible tradingpartners and alert their respective users if amutually beneficial trade opportunity arises.Once users become aware of this, they canact on or ignore the agents’ advice.

Principle 2: Reward users for social inter-actions. The frequency of social encountersper person should be proportional to thegain a user receives from using the system.

Principle 3: Help people to recognizeeach other. Without identity and mutualrecognition, a group of people will alwaysremain strangers. Promoting altruisticbehavior and cooperation necessitates asocial feedback mechanism that lets usersevaluate and track others’ behavior. Such afeedback mechanism requires identity andmutual recognition.

Principle 4: Support expressiveness. Live-action role-playing games have gainedwidespread international popularity. Suchgames become social events, bringing

together people who have taken on imagi-nary identities based on historical or myth-ical characters, and the social interactionscontribute much to players’ enjoyment.Similarly, wearable communities let mem-bers invent digital identities to augmenttheir true identities. Wearable communitiesshould promote such creativity and artisticexpressiveness as much as possible.

Principle 5: Use knowledge about pastencounters to enrich present interactions.An important aspect of social encounters isour ability to recognize other people and toremember if, when, where, and under whatcircumstances we met someone before.Wearable computers can track both theuser’s current context (defined by other peo-ple’s presence) and his or her interaction his-tory. Applying this knowledge in wearablecommunities can enrich interactions in sig-nificant ways.

Principle 6: Include social feedbackloops. Any community has disruptive orunruly members—in newsgroups, somepeople consistently and inappropriatelyflame others; on eBay, some members cheatby not paying for items they purchase. Suc-cessful communities can police their mem-bers’ behavior not by giving some mem-bers extraordinary powers but throughsocial feedback. For example, eBay usesreputations (aggregated signed opinionsfrom eBay users on a buyer’s or seller’stransaction history) to let members rateand judge other members’ trustworthiness.The community thus virtually excludes dis-rupting members. (Jay Schneider22 presentsa good discussion of trust and reputationin wearable communities.)

W illiam Gibson describedcyberspace as a “consensualhallucination experienceddaily by billions of legiti-

mate operators, in every nation.”23 Wear-able communities present an alternativemodel of human communication that reliesnot on shared imagination but on real-worldencounters and first-hand experiences.Mobile communication and computation


technologies have sufficiently permeated oursociety to shape social behavior, norms, andconventions. As mobile computers evolveinto wearable computers—constant, aware,communicative, and proactive—we canexpect to see even more dramatic changes.We believe that properly designed wearablecomputing technology can indeed begetcommunity by facilitating social networksbased on augmented interactions and face-to-face encounters.

Wearable communities face sociologicalas well as technical problems, which are dif-ficult to separate—they interact and coevolveintimately. We plan further empirical re-search on wearable communities and thesocial networks this technology creates.

ACKNOWLEDGMENTSThis work was performed at and supported by theUniversity of Oregon’s Computer and InformationScience Department.

REFERENCES1. T. Starner, Wearable Computing and

Contextual Awareness, doctoral disserta-tion, Mass. Inst. Technology, Cambridge,Mass., 1999.

2. S. Mann, “Wearable Computing: TowardHumanistic Intelligence,” IEEE IntelligentSystems, vol. 16, no. 3, May/June 2001,pp. 10–15.

3. B. Rhodes, “The Wearable RemembranceAgent: A System for AugmentedMemory,” Personal Technologies, vol. 1,no. 4, Dec. 1997, pp. 218–224; www.bradleyrhodes.com/Papers/wear-ra-personaltech/index.html.

4. N.K. Baym, “The Emergence ofCommunity in Computer-MediatedCommunication,” CyberSociety, S. Jones,ed., Sage Publications, Newbury Park,Calif., 1995, pp. 35–68.

5. P. Kollock and M. Smith, eds., Communi-ties in Cyberspace, Univ. Calif. Press, Ewing,N.J., 1998.

6. J. Preece, Online Communities: DesigningUsability, Supporting Sociability, JohnWiley & Sons, New York, 2000.

7. G. Kortuem et al., “When Cyborgs Meet:Building Communities of CooperatingWearable Agents,” Proc. 3rd Int’l Symp.Wearable Computers, IEEE CS Press, LosAlamitos, Calif., 1999, pp. 124–132.

8. IEEE 802.15 Working Group for WirelessPersonal Area Networks, http://grouper.ieee.org/groups/802/15.

9. E.T. Hall, The Silent Language, Anchor,Garden City, N.Y., 1973.

10. M. Terry et al., “Social Net: Using Patternsof Physical Proximity over Time to InferShared Interests,” Proc. 2002 Conf. HumanFactors in Computing Systems (CHI 2002),ACM Press, New York, 2002, pp. 816–817.

11. G. Kortuem, Z. Segall, and M. Bauer, “Con-text-Aware, Adaptive Wearable Computersas Remote Interfaces to ‘Intelligent’ Envi-ronments,” Proc. 2nd Int’l Symp. WearableComputers (ISWC ’98), IEEE CS Press, LosAlamitos, Calif., 1998, pp. 58–65.

12. G. Kortuem et al., “When Peer-to-PeerComes Face-to-Face: Collaborative Peer-to-Peer Computing in Mobile Ad-hoc Net-works,” Proc. First Int’l Conf. Peer-to-PeerComputing (P2P ’01), IEEE CS Press, LosAlamitos, Calif., 2001, pp. 75–91.

13. J.C.R. Licklider, “The Computer as a Com-munication Device,” Science and Technol-ogy, Apr. 1968; http://gatekeeper.dec.com/pub/DEC/SRC/publications/taylor/licklider-taylor.pdf.

14. H. Rheingold, The Virtual Community:Homesteading on the Electronic Frontier,Perseus Books, New York, 1998, www.rheingold.com/vc/book.

15. E. Wenger, Communities of Practice: Learn-ing, Meaning, and Identity, CambridgeUniv. Press, New York, 1998.

16. G. Kortuem, A Methodology and SoftwarePlatform for Building Wearable Commu-nities, doctoral dissertation, Computer andInformation Science Dept., Univ. Oregon,Eugene, Ore., 2002.

17. A. Smailagic et al., “Very Rapid Prototyp-ing of Wearable Computers: A Case Studyof VuMan 3 Custom versus Off-the-ShelfDesign Methodologies,” J. Design Automa-tion for Embedded Systems, vol. 3, nos. 2/3,1998, pp. 217–230.

18. G. Kortuem, “Proem: A Middleware Plat-form for Mobile Peer-to-Peer Computing,”to be published in ACM SIGMOBILEMobile Computing and CommunicationsRev., 2003.

19. M. Godwin, “Nine Principles for MakingVirtual Communities Work,” Wired, vol.2.06, June 1994, pp. 72–73.

20. P. Kollock, “Design Principles for OnlineCommunities,” PC Update, vol. 15, no. 5,June 1998, pp 58–60.

21. E. Ostrom, Governing the Commons: TheEvolution of Institutions for CollectiveAction, Cambridge Univ. Press, New York,1990.

22. J. Schneider et al., “Disseminating TrustInformation in Wearable Communities,”Personal and Ubiquitous Computing, vol.4, no. 4, Aug. 2000, pp. 245–248.

23. W. Gibson, Neuromancer, Ace Books, NewYork, 1984.

For more information on this or any other comput-ing topic, please visit our Digital Library at http://computer.org/publications/dlib.

the AUTHORS

Gerd Kortuem is a lecturer incomputer science at LancasterUniversity, UK. His researchinterests include collaborativetechnologies, ubiquitouscomputing, wearable com-puting, and soft-ware engi-neering. He received an MS in

computer science from the University of Stuttgartand a PhD in computer science from the Univer-sity of Oregon. He is a member of the IEEE Com-puter Society and ACM. Contact him at the Com-puting Dept., Lancaster Univ., Lancaster, LA1 4YR,UK; [email protected].

Zary Segall is the 2002–2003 Fulbright Distin-guished IT Chair at theRoyal Institute of Technol-ogy, Sweden, and profes-sor of computer science atthe University of Oregon.His research has focused

on theoretical and practical methods and sys-tems for parallel processing, highly dependablesystems, fault tolerance, real time, binary-to-binary translation, networking, and wearableinformation systems. He is a fellow of the IEEEComputer Society. Contact him at the Com-puter Science Dept., Univ. of Oregon, Eugene,OR 97405; [email protected].



Documents

Wearable communities: augmenting social networks with wearable computers