designing smart artifacts for smart environments

9
0018-9162/05/$20.00 © 2005 IEEE March 2005 41 COVER FEATURE Published by the IEEE Computer Society Designing Smart Artifacts for Smart Environments A n integral part of our environment, com- puters contribute to the social context that determines our day-to-day activities while at the office, on the road, at home, or on vacation. The widespread avail- ability of devices such as desktop and laptop com- puters has fueled our increasing dependency on a wide range of computing services. The technolog- ical advances that underlie the laptop, PDA, or cell phone also provide the foundation for nontradi- tional computer-based devices such as interactive walls, tables, and chairs—examples of roomware components that provide new functionality when combined with innovative software. 1 Two complementary trends have resulted in the creation of smart environments that integrate infor- mation, communication, and sensing technologies into everyday objects. 2 First, continual miniatur- ization has resulted in computers and related tech- nological devices that are small enough to be nearly invisible. Although they are not visible, these devices still permeate many artifacts in our environment. Second, researchers have augmented the standard functionality of everyday objects to create smart artifacts constituting an environment that supports a new quality of interaction and behavior. In our work, we distinguish between two types of smart artifacts: system-oriented, importunate smart- ness and people-oriented, empowering smartness. System-oriented, importunate smartness creates an environment in which individual smart artifacts or the environment as a whole can take certain self- directed actions based on previously collected infor- mation. For example, a space can be smart by having and exploiting knowledge about the persons and artifacts currently situated within its borders, for example, how long they have occupied the space and what actions they have performed while in it. In this version of smartness, the space would be active, in many cases even proactive. It would make decisions about what to do next and actually exe- cute those actions without a human in the loop. In a smart home, for example, the control system auto- matically performs functions such as adjusting the heating system and opening or closing the windows and blinds. In some cases, however, these actions could be unwelcome or ill-timed. Consider a smart refriger- ator that analyzes the occupants’ consumption pat- terns and autonomously orders replacements for depleted menu items. Although we might appreci- ate suggestions for recipes we can make with the food that is currently available, we would proba- bly resent a smart refrigerator that ordered food automatically that we could not consume because of circumstances beyond the refrigerator’s knowl- edge such as an unanticipated absence or illness. In contrast, people-oriented, empowering smart- ness places the empowering function in the fore- ground so that “smart spaces make people smarter.” This approach empowers users to make decisions and take mature and responsible actions. Smart artifacts promise to enhance the relationships among participants in distributed working groups, maintaining personal mobility while offering opportunities for the collaboration, informal communication, and social awareness that contribute to the synergy and cohesiveness inherent in collocated teams. Norbert A. Streitz Carsten Röcker Thorsten Prante Daniel van Alphen Richard Stenzel Carsten Magerkurth Fraunhofer IPSI Darmstadt, Germany

Upload: c

Post on 22-Sep-2016

221 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Designing smart artifacts for smart environments

0018-9162/05/$20.00 © 2005 IEEE March 2005 41

C O V E R F E A T U R E

P u b l i s h e d b y t h e I E E E C o m p u t e r S o c i e t y

Designing SmartArtifacts for SmartEnvironments

A n integral part of our environment, com-puters contribute to the social contextthat determines our day-to-day activitieswhile at the office, on the road, at home,or on vacation. The widespread avail-

ability of devices such as desktop and laptop com-puters has fueled our increasing dependency on awide range of computing services. The technolog-ical advances that underlie the laptop, PDA, or cellphone also provide the foundation for nontradi-tional computer-based devices such as interactivewalls, tables, and chairs—examples of roomwarecomponents that provide new functionality whencombined with innovative software.1

Two complementary trends have resulted in thecreation of smart environments that integrate infor-mation, communication, and sensing technologiesinto everyday objects.2 First, continual miniatur-ization has resulted in computers and related tech-nological devices that are small enough to be nearlyinvisible. Although they are not visible, these devicesstill permeate many artifacts in our environment.Second, researchers have augmented the standardfunctionality of everyday objects to create smartartifacts constituting an environment that supportsa new quality of interaction and behavior.

In our work, we distinguish between two types ofsmart artifacts: system-oriented, importunate smart-ness and people-oriented, empowering smartness.

System-oriented, importunate smartness createsan environment in which individual smart artifacts

or the environment as a whole can take certain self-directed actions based on previously collected infor-mation. For example, a space can be smart byhaving and exploiting knowledge about the personsand artifacts currently situated within its borders,for example, how long they have occupied the spaceand what actions they have performed while in it.

In this version of smartness, the space would beactive, in many cases even proactive. It would makedecisions about what to do next and actually exe-cute those actions without a human in the loop. Ina smart home, for example, the control system auto-matically performs functions such as adjusting theheating system and opening or closing the windowsand blinds.

In some cases, however, these actions could beunwelcome or ill-timed. Consider a smart refriger-ator that analyzes the occupants’ consumption pat-terns and autonomously orders replacements fordepleted menu items. Although we might appreci-ate suggestions for recipes we can make with thefood that is currently available, we would proba-bly resent a smart refrigerator that ordered foodautomatically that we could not consume becauseof circumstances beyond the refrigerator’s knowl-edge such as an unanticipated absence or illness.

In contrast, people-oriented, empowering smart-ness places the empowering function in the fore-ground so that “smart spaces make people smarter.”This approach empowers users to make decisionsand take mature and responsible actions.

Smart artifacts promise to enhance the relationships among participants indistributed working groups, maintaining personal mobility while offeringopportunities for the collaboration, informal communication, and socialawareness that contribute to the synergy and cohesiveness inherent in collocated teams.

Norbert A.StreitzCarstenRöckerThorstenPranteDaniel van AlphenRichard StenzelCarstenMagerkurthFraunhofer IPSI

Darmstadt, Germany

Page 2: Designing smart artifacts for smart environments

42 Computer

In this case, the system also collects andaggregates data about what goes on in thespace, but it provides and communicates thisinformation intuitively so that ordinary peo-ple can comprehend and determine the sys-tem’s subsequent actions. This type of smartspace might make suggestions based on the information collected, but users remain in the loop and can always decide what to do next.

This type of system supports its occupants’smart, intelligent behavior. In an office sce-nario, for example, the smart space couldrecommend that current occupants consultwith others who worked on the same con-tent while occupying the same space earlier

or it could direct them to look at related documentscreated earlier in the same space.

The system-oriented and people-orientedapproaches represent the end points of a line alongwhich we can position weighted combinations ofboth types of smartness depending on the applica-tion domain. Although in some cases it might bemore efficient if the system does not ask for a user’sfeedback and confirmation at every step in anaction chain, the overall design rationale shouldaim to keep the user in the loop and in controlwhenever possible.

FROM INFORMATION TO EXPERIENCEMuch work on smart things and environments

focuses on intelligently processing the data andinformation that supports factory and home con-trol and maintenance tasks or productivity-orientedoffice tasks. We considered another promisingdimension, however: designing experiences viasmart spaces.

We sought to design smart artifacts that users caninteract with simply and intuitively in the overallenvironment. This includes extending awarenessabout the physical and social environment by pro-viding observation data and parameters that—inmany cases—are invisible to unaugmented humansenses. Revealing this information thus enables newexperiences.

This process of capturing and communicatinginvisible parameters is applicable to both knownexisting action contexts and to newly created situ-ations and settings. Known examples include pol-lution or computer network traffic data that usuallyescapes detection by the human senses.3 Presentingthis data can provide a new experience that givespeople a deeper sense of what occurs around them.Depending on the particular application, this capa-

bility could raise public awareness and potentiallytrigger changes in people’s behavior.

Our work in creating augmented social archi-tectural spaces in office settings culminated in theAmbient Agoras environment (www.ambient-agoras.org).4 We are now applying this knowledgeto other domains, including interactive hybridgames, home entertainment, and extended homeenvironments.

We focus here on computer-based support foractivities beyond direct productivity, in particularinformal communication and social interactionbetween local and remote teams in an organizationthat are working at different but connected sites.5,6

Because these activities are important to an orga-nization’s overall progress and success, they meritmore technology-based support.

Ambient AgorasWithin this overall context, we used the Ambient

Agoras environment as a test bed for developingfuture applications of ubiquitous and ambient com-puting in smart workspaces. This required a widerange of expertise and a highly interdisciplinaryapproach involving not only computer scientistsand electrical engineers but also psychologists,architects, and designers.

Fraunhofer IPSI in Darmstadt, Germany, pro-vided the scientific, technical, and administrativeproject coordination. Fraunhofer’s AmbienteResearch Division also employed product design-ers and architects to develop some of the artifacts.

Electricité de France, the French electrical powerutility, served as the consortium’s user organiza-tion. As part of its R&D division, the Laboratoryof Design for Cognition in Paris provided the testbed for the evaluation studies and contributed tothe observation and participatory design methods.

Wilkhahn, a German office furniture manufac-turer, contributed to the design and developmentof some artifacts, leveraging its experience indesigning the second generation of Roomwarecomponents developed in cooperation withFraunhofer IPSI in the Future Office Dynamicsconsortium.7

Social marketplace of ideas and informationWe chose as the guiding metaphor for our work

the Greek agora, a marketplace. In line with this,we investigated how to turn everyday places intosocial marketplaces of ideas and information wherepeople could meet and interact.

In our particular context, we addressed the officeenvironment as an integrated organization located

Developing futureapplications of ubiquitous and

ambient computingin smart workspaces

required a widerange of expertise

and a highlyinterdisciplinary

approach.

Page 3: Designing smart artifacts for smart environments

in a physical environment and having particularinformation needs, both at the organization’s col-lective level and at the worker’s personal level.

Overall, we sought to augment the architecturalenvelope to create a social architectural space thatsupports collaboration, informal communication,and social awareness. We achieved this by provid-ing situated services and place-relevant informa-tion that communicate the feeling of a place tousers. Further, augmented physical artifacts helppromote individual and team interactions in thephysical environment.

Specifically, we used a scenario-based approach,starting with a large number of so called bits oflife—short descriptions of functionalities, situa-tions, events, and so on—that we aggregated intoscenarios and presented to focus groups usingvisual aids such as video mock-ups. This, in com-bination with extensive conceptual work based ondifferent architectural theories,8 served as the basisfor developing a wide range of smart artifacts andtheir corresponding software that, together, pro-vided users with smart services.

Design, development, and evaluation followed aniterative and rapid-prototyping approach. For theAmbient Agoras environment, we coupled severalinteraction design objectives, including the disap-pearance and ubiquity of computing devices; sens-ing technologies such as active and passive RFID;smart artifacts such as walls, tables, and mobiledevices; and ambient displays. We then investigatedthe functionality of two or more artifacts workingtogether. In particular, we addressed the

• support of informal communication in organi-zations, both locally and between remote sites;

• role and potential of ambient displays in futurework environments; and

• combination of more or less static artifactsintegrated in the architectural environmentwith mobile devices carried by people.

Mobility and informal communicationSeveral trends are changing how large organiza-

tions work. For example, organizations increas-ingly organize work around teams that changedynamically in response to the temporary natureof projects. People working in these organizationsalso experience a large degree of personal mobilityin two dimensions:

• local mobility within the office building as aresult of new office concepts such as the lossof personal office space because of shared desk

policies and wide-open office landscapeswith movable walls and furniture thatcan be adapted on the fly to changingrequirements and new project team con-stellations; and

• global mobility achieved by using mobiletechnologies while traveling or workingat different sites at the municipal,regional, national, or international level.

Although increased mobility offers severalbenefits, it also has implications that demandnew responses, especially at the global mobil-ity level.

At the local level, the usually recognized channelsof communication between people working togetherinclude face-to-face conversations, formal meetings,phone conversations, e-mail messages, and docu-ment sharing. In addition, informal communicationincludes interactions such as chance encounters atthe copying machine, hallway chats, and conversa-tions while relaxing in the lounge. These interac-tions help participants stay on top of things,anticipate future developments in the organization,and exchange gossip and rumors. Like explicit ver-bal communication, implicit communication occursin terms of a mutual awareness through which peo-ple can determine who’s who and assess theircoworkers’ overall mood and morale.

Design recommendations for the workplace fre-quently conclude that both informal awarenessabout ongoing activities in the local work environ-ment and a sense of community play vital roles inthe workplace.9 Teams that share the same physicalenvironment generally benefit from increased infor-mal awareness because the team members havehigher mobility within the shared workspace.

When looking at global mobility, the situationchanges fundamentally. The increased mobility ofteam members usually leads to poor communica-tion and lack of group cohesion, which negativelyaffects the teams’ performance. This holds true forindividual global mobility caused by intensive trav-eling and for group global mobility in the case ofdistributed teams that have subgroups working atdifferent sites.

One empirical study that addressed this topic con-firmed the trend toward the formation of virtualteams, but noted that such teams reduced interper-sonal relations to a minimum.10 Further, this studyshowed that it is exactly these relationships betweenteam members that have the strongest effect on performance and work satisfaction. The poor com-munication and lack of group cohesion often expe-

March 2005 43

Informal awarenessabout ongoing

activities in thelocal work

environment and asense of communityboth play vital rolesin the workplace.

Page 4: Designing smart artifacts for smart environments

44 Computer

rienced in virtual and distributed teams has consid-erable negative effects on team performance.11

Building on this work, we developed the con-stituents for a smart environment in a corporatesetting that augments existing local and distributedarchitectural spaces, transforming them into spacesthat make people “smarter” by supporting socialawareness and informal communication.

POPULATING AMBIENT AGORASEach of the artifacts and software components

we developed to populate the Ambient Agorassmart spaces, including InfoRiver, InforMall, andthe SIAM-system, meets different aspects of ouroverall design goals.12 We focus here on theHello.Wall, ViewPort, and Personal Aura.

Calm technologyWhile working on the ideas that the Ambient

Agoras embody, we set another complementarygoal for implementing the technology. We felt thatthe implementation should correspond to and becompatible with the nature of informal communi-cation, social awareness, and team cohesion. Ourconceptual analysis combined with informationgathered from focus groups showed that traditionalapproaches to communicating using desktop tech-nology did not achieve this goal and would notmeet expectations.

Therefore, we took a different route based on thenotion of ambient displays and lightweight supportwith mobile devices. An observation by MarkWeiser helped inspire our decision to move thecomputer to the background and develop a calmtechnology: “The most profound technologies arethose that disappear. They weave themselves intothe fabric of everyday life until they are indistin-guishable from it.”13 The “The DisappearingComputer” sidebar describes Weiser’s influence onproactive research exploring how smart artifactscan support and enhance everyday life in a people-friendly environment in which there is no role forthe “computer as we know it.”

Ambient displaysWe decided that a calm, ambient technology best

supports the informal social encounters and com-munication processes within a corporate building.The ambient displays that exemplify this approachgo beyond the traditional notions of the typical dis-plays found on PCs, notebooks, PDAs, and evenmany interactive walls or tables.

Some ambient displays employ nature-likemetaphors to present information without con-stantly demanding the user’s full attention. Theyusually achieve this implicitly by making the dis-plays available in the periphery of attention.

Designers envision that ambient displays willspring up all around us, moving information off con-ventional screens and into the physical environment.They will present information via changes in light,sound, object movement, smell, and so on. HiroshiIshii and his colleagues at the MIT Media Lab devel-oped several early examples of this technology.3,14

Given that awareness of people’s activities canstrengthen social affiliations, ambient displays canbe used to trigger the attention of team members sub-tly and peripherally by communicating a location’satmosphere, thus providing a sense of place.

Ambient displays provide only one aspect of theimplementation, however. Another aspect is sens-ing people and collecting the parameters relevantto achieving the goal of providing location- and sit-uation-based services.

Hello.WallWe developed the Hello.Wall, our version of an

ambient display, for the Ambient Agoras environ-ment. This 1.8-meter-wide by 2-meter-high com-pound artifact has integrated light cells and sensingtechnology. As Figure 1 shows, this display facili-tates communication via dynamically changing lightpatterns. The current version uses 124 light-emit-

The Disappearing Computer

The European Commission funded the proactive The DisappearingComputer research initiative. Launched by the Future and EmergingTechnology section of the Information Society Technology program,The Disappearing Computer initiative seeks “to explore how everydaylife can be supported and enhanced through the use of collections ofinteracting smart artifacts. Together, these artifacts will form new peo-ple-friendly environments in which the ‘computer-as-we-know-it’ hasno role.”

The initiative has three main objectives:

• developing new tools and methods for embedding computationin everyday objects to create smart artifacts;

• investigating how new functionality and new uses can emergefrom collections of interacting artifacts; and

• ensuring that people’s experience of these environments is bothcoherent and engaging in space and time.

These objectives require research in ambient intelligence, pervasiveand ubiquitous computing, and new forms of human-computer inter-action. Researchers have undertaken a cluster of 17 related projectsunder the umbrella theme of The Disappearing Computer initiativeto pursue these three objectives. The Ambient Agoras project is one ofthese projects.

For more information about The Disappearing Computer initiative,visit www.disappearing-computer.net or contact Norbert Streitz, chairof the DC-Net Steering Group.

Page 5: Designing smart artifacts for smart environments

ting cells organized in an eight-row array structure. A standard computer hidden in the background

uses a special driver interface to control theHello.Wall artifact. To adjust the LED clusters’brightness, we developed a new control unit thatuses pulse-width modulation. The system’s generaldesign captures a range of parameters as input andmaps them to a wide range of output patterns.

The Hello.Wall provides awareness and notifi-cations to people passing by or watching it.Different light patterns correspond to differenttypes of information. Using abstract patterns allowsdistinguishing between public and private or per-sonal information. Although everyone knows themeaning of public patterns and can therefore inter-pret them easily, only the initiated can access themeaning of personal patterns. This makes it possi-ble to communicate personal messages and infor-mation in a public space without worry that otherswill catch their meaning.

In the Ambient Agoras environment, theHello.Wall functions as an ambient display thattransmits organization-oriented information pub-licly and information addressed to individuals pri-vately. We can think of it as an organism thatradiates the breath of an organization’s social body,making it perceivable to the organization’s mem-bers on the inside as well as others on the outside.

The Hello.Wall does more than communicateinformation and atmosphere, however—its appear-ance also has an effect on the atmosphere of a placeand thus influences the mood of the social bodyaround it. While the artifact serves a dedicatedinformative role to initiated members of the orga-nization, visitors might consider it as simply anatmospheric decorative element and enjoy its aes-thetic quality.

As an integral part of the physical environment,the Hello.Wall constitutes a seeding element of asocial architectural space that provides awarenessto the members of an organization. In this way, theHello.Wall is a piece of unobtrusive, calm technol-ogy that exploits people’s ability to perceive infor-mation via codes. It can stay in the background, atthe periphery of attention, while those around itconcern themselves with another activity, such asa face-to-face conversation. The Hello.Wall’sunique blend of unobtrusive, calm technology andits continual display of high-quality aesthetic pat-terns make it informative art.15

Sensing and different zones of interactionBeyond developing a new ambient display, we

also sought to make the type of information and

how it is communicated context-dependent. Theartifact should provide services that are location- orsituation-based depending on the proximity of peo-ple passing by. As Figure 2 shows, depending onthe distance from the display, the Hello.Wall hasthree different communication zones: ambient,notification, and interaction.

To cover different ranges, we used integrated sen-sors that can be adapted according to the sur-rounding spatial conditions. Using these sensorsintroduces a distance-dependent semantic, whichimplies that the distance of an individual from the

March 2005 45

Figure 1. Hello.Wall. The ambient display combines unobtrusive, calm technologyand a continual display of high-quality aesthetic patterns to convey the idea ofturning everyday spaces into agoras—social marketplaces where people canmeet and interact.

Figure 2. Communication zones. Depending on the distance from the display, the Hello.Wall has three communication zones: ambient, notification, and interaction.

Page 6: Designing smart artifacts for smart environments

46 Computer

smart artifact defines the kind of informationshown and the interaction offered.

People passing through the ambient zone con-tribute to and experience the ambient patterns con-tinuously displayed on the Hello.Wall. Thesepatterns concern, for example, general presenceinformation. People in the notification zone are iden-tified as individuals and agree to have the Hello.Wall-enriched environment react to their personalpresence. This can result in personal notification pat-terns being displayed on the Hello.Wall. People inthe interaction zone can get directly involved withthe Hello.Wall environment. The artifact reflects thisby showing special interaction patterns.

ViewPortWe designed a complementary mechanism for the

Hello.Wall that can “borrow” the displays of otherartifacts to communicate additional informationthat complements the Hello.Wall’s display. As

Figure 3 shows, each of these mobile ViewPorts con-sists of a WLAN-equipped PDA-like handhelddevice based on commercially available componentsthat are mapped to a new form factor. Furthermore,we integrated RFID readers and transponders sothat a ViewPort can sense other artifacts and besensed itself.

The Hello.Wall can borrow the ViewPort’s dis-play to privately show more explicit and personalinformation that can be viewed only on a personalor temporarily personalized device. Depending onaccess rights and current context, people can useViewPorts to learn more about the Hello.Wall, todecode visual codes on the wall, or to access a mes-sage announced by a code.

Personal AuraPeople adopt different social roles in daily life,

such as mother, client, or customer. In some situa-tions, communication reveals a lot about a person,while in others it reveals very little.

In a corporate organization, employees have dif-ferent professional roles that might change evenduring the course of a single workday. An employeecan be the project manager on one team and laterparticipate in another meeting as a regular taskforce member.

Based on these considerations, we wanted to pro-vide a similar mechanism for sensor-based environ-ments. We sought to design an easy and intuitiveinterface that would let users control their appear-ance in a smart environment. They could decidewhether to be visible to a tracking system and, if so,they could control the social role in which theyappeared. This mechanism contributes to theincreasing discussion of privacy issues that the imple-mentation of smart environments has generated.16

Figure 4 shows the Personal Aura, our firstinstantiation of this concept. The artifact consistsof two matching parts: the reader module and the

Figure 4. Personal Aura. (a) The reader module and two ID sticks, (b) connecting reader module and ID stick, and (c) an active Personal Aura.

Figure 3. ViewPort.(a) The implementedprototype and (b) adesign prototypeshowing the nextversion’s new formfactor.

(a)

(b)

(a) (b) (c)

Page 7: Designing smart artifacts for smart environments

ID stick, which contains a unique identity andoptional personal information. Each person hasmultiple ID sticks, with each stick symbolizing adifferent role. If people want to signal their avail-ability in the connecting remote team application,for example, they can do so by connecting a spe-cific ID stick to the reader module.

CONNECTING REMOTE TEAMSAside from opportunistic chance encounters in the

hallway, gathering in a lounge area offers people thehighest accessibility to informal communication.Although a person’s mood and availability for par-ticipating in a chat can be detected easily in a face-to-face situation, identifying similar information ina remote setting is difficult. People must be called ore-mailed to determine their receptiveness to an en-counter. When they use standard videoconferencingsystems, people usually must plan the encounter andprepare the setup in advance.

To evaluate how the Hello.Wall and its support-ing artifacts could facilitate communicationbetween two remote teams, we ran a living-lab eval-uation in the fall of 2003. This scenario addressedthe issue of extending awareness information andfacilitating informal communication from withina corporate building to distributed teams workingat remote sites.

We built and installed Hello.Wall ambient dis-plays and the corresponding sensing infrastructuresin two lounge spaces, one at EDF-LDC in Paris andthe other at Fraunhofer IPSI in Darmstadt. Figure5 shows the Hello.Wall in one of the lounge areas.

We used different media to allow the continuousexchange of information about the availability ofpeople for chance encounters and to provide a start-ing point for initiating spontaneous video-basedcommunication between the two remote sites.

We mapped the zone model to the floor plans ofthe lounges at each site. While people in the ambi-ent zone only contributed to the ambient presencepatterns, people entering the notification zone wereidentified via their Personal Aura, and their per-sonal sign was displayed at the Hello.Wall at theopposite remote lounge space. Thus, the Hello.Wallcontinuously presented a combination of patternscommunicating what was going on at the remotesite. People could perceive this information in anambient way without having to explicitly focustheir attention on it.

When they became aware of the presence of par-ticular people and had the feeling that it was a goodtime to engage in a spontaneous encounter, peopleneeded a way to communicate their interest in an

intuitive way. The request was triggered by push-ing a button, which resulted in a specific patternthat overrode all other patterns on the Hello.Wallat the remote site. The remote site could reject thisrequest or accept the invitation, and the informalvideo-based communication could proceed.

We used dynamic light patterns to communicatedifferent types of information: the presence andnumber of people at the opposite site, their generalmood, the presence and availability of specific teammembers, and their interest in communicating withthe remote team.

We designed a specific pattern language that dis-tinguishes between the

• ambient patterns representing general infor-mation like mood and presence;

• notification patterns handling individual orpersonalized messages; and

• interaction patterns handling direct commu-nication requests, such as a request for engag-ing in a spontaneous video communicationwith a remote team member.

To give them an aesthetically pleasing and non-monotonic appearance, we purposely designed thepatterns to appear abstract. The Hello.Wall con-tinuously displays these dynamic patterns as theyinterweave with each other. To reduce complexityand facilitate peripheral perception, as Figure 6shows, the wall displays the presence and mood pat-terns at only three levels—low, medium, and high.

In addition, the Hello.Wall can apply overlays tothese patterns. Static personal signs display when aspecific team member appears in the lounge area.Figure 7 shows that each person has a specific sign,controlled by the Personal Aura. As a dedicatedexample of privacy-enhancing technology, thePersonal Aura provides users with control overRFID-based identification in smart environments.6

March 2005 47

Figure 5. Hello.Wallin lounge area. Theexperimental settingconsisted of twolounge areas, eachenhanced with aHello.Wall: one atEDF-LDC in Paris and the other atFraunhofer IPSI in Darmstadt,Germany.

Page 8: Designing smart artifacts for smart environments

48 Computer

U sing questionnaires to provide feedbackrevealed that our approach was effective infacilitating workplace awareness and group

communication. Our evaluation demonstrated thatparticipants could learn how to identify and inter-pret the Hello.Wall patterns correctly in a shortperiod of time. The participants indicated that theyperceived the Hello.Wall as being an appropriatemeans of establishing awareness of people whowere working at a remote site, thus overcoming the

isolation of not being physically present withoutcausing privacy problems.

The study participants described the Hello.Wallas providing a playful experience while interactingwith the remote team. They commented that dueto the Hello.Wall, interactions with the remote sitetook place more often, spontaneous video confer-ence interactions were less formal, and videocon-ferencing became a daily routine.

The Hello.Wall patterns and their smooth move-ments when flowing over the display were consid-ered to be aesthetically pleasing. People mentionedthat the Hello.Wall caused positive feelings andinduced a good mood.

Our future work will exploit the results gained inthis study as we focus on building awareness sup-port for distributed remote home environments in anew EU-funded project, Amigo–Ambient Intelli-gence for the Networked Home Environment (www.ipsi.fraunhofer.de/ambiente/amigo). �

AcknowledgmentsWe thank the European Commission for its

extensive support of The Disappearing Computerinitiative (contract IST-2000-25134). Thanks alsoto our EDF partners: Saadi Lahlou and his team,DALT, Wilkhahn, wiege, Daniela Plewe, SebastianLex, and the members and students in the AmbienteResearch Division (www.ipsi.fraunhofer.de/ambi-ente) at Fraunhofer IPSI. For more informationabout the Disappearing Computer initiative, visitwww.disappearing-computer.net.

References1. T. Prante, N. Streitz, and P. Tandler, “Roomware:

Computers Disappear and Interaction Evolves,”Computer, Dec. 2004, pp. 47-54.

2. N. Streitz and P. Nixon, “The Disappearing Com-puter,” Comm. ACM, Mar. 2005, pp. 33-35.

3. C. Wisneski et al., “Ambient Displays: Turning Archi-tectural Space into an Interface between People andDigital Information,” Cooperative Buildings: Inte-grating Information, Organization, and Architecture,N. Streitz, S. Konomi, and H. Burkhardt, eds., LNCS1370, Springer-Verlag, 1998, pp. 22-32.

4. N. Streitz et al., “Ambient Displays and MobileDevices for the Creation of Social ArchitecturalSpaces: Supporting Informal Communication andSocial Awareness in Organizations,” Public and Sit-uated Displays: Social and Interactional Aspects ofShared Display Technologies, K. O’Hara et al., eds.,Kluwer, 2003, pp. 387-409.

Figure 6. Hello.Wall patterns. The patterns express (a) three different levels ofmood and (b) three different levels of presence—low, medium, and high.

Figure 7. Static personal signs. The Personal Aura privacy-enhancing technologycontrols different personal signs, each indicating a specific person’s presenceand role.

(a)

(b)

Page 9: Designing smart artifacts for smart environments

5. T. Prante et al., “Connecting Remote Teams: Cross-Media Integration to Support Remote InformalEncounters,” Adjunct Proc. 6th Int’l Conf. Ubiqui-tous Computing (UbiComp 04), Univ. Nottingham,UK, 2004.

6. C. Röcker et al., “Using Ambient Displays and SmartArtifacts to Support Community Interaction in Dis-tributed Teams,” Proc. OZCHI 2004, Univ. Wol-longong, Australia, 2004.

7. N. Streitz et al., “Roomware: Toward the Next Gen-eration of Human-Computer Interaction Based onan Integrated Design of Real and Virtual Worlds,”Human-Computer Interaction in the New Millen-nium, J. Carroll, ed., Addison-Wesley, 2001, pp. 553-578.

8. C. Alexander, S. Ishikawa, and M. Silverstein, A Pat-tern Language: Towns, Buildings, Construction,Oxford University Press, 1977.

9. J. Tanis and F. Duffy, A Vision of the New WorkplaceRevisited, Site Selection, Sept. 1999, pp. 805-814.

10. J. Lurey and M. Raisinghani, “An Empirical Studyof Best Practices in Virtual Teams,” Information &Management, vol. 38, 1999, pp. 523-544.

11. R. Blake, J. Mouton, and A. McCanse, Change byDesign, Addison-Wesley, 1989.

12. T. Prante et al., “Ambient Agoras: InfoRiver, SIAM,Hello.Wall,” Proc. Human Factors in ComputingSystems (CHI 2004), ACM Press, pp. 763-764.

13. M. Weiser, “The Computer for the 21st Century,”Scientific American, Sept. 1991, pp. 94-104.

14. H. Ishii et al., “AmbientROOM: Integrating Ambi-ent Media with Architectural Space,” Proc. HumanFactors in Computing Systems (CHI 98), ACM Press,1998, pp. 173-174.

15. J. Redström, T. Skog, and L. Hallnäs, “InformativeArt: Using Amplified Artworks as Information Dis-plays,” Proc. Designing Augmented Reality Envi-ronments (DARE 2000), ACM Press, 2000, pp.103-114.

16. S. Lahlou, M. Langheinrich, and C. Röcker, “Privacyand Trust Issues with Invisible Computers,” Comm.ACM, Mar. 2005, pp. 59-60.

Norbert A. Streitz is head of the Ambiente–SmartEnvironments of the Future research division atFraunhofer IPSI, Darmstadt, Germany, where healso teaches in the Department of Computer Sci-ence at the Technical University Darmstadt. Hisresearch interests include cognitive science, com-puter-supported cooperative work, and interactiondesign for ubiquitous computing. Streitz receiveda PhD in physics from the University of Kiel and a

PhD in psychology from the Technical UniversityRWTH Aachen. He chairs the steering group of theEU-funded The Disappearing Computer initiative.Contact him at [email protected].

Carsten Röcker is a PhD candidate in computerscience at the Technical University Darmstadt anda scientific staff member in the Ambiente–SmartEnvironments of the Future research division atFraunhofer IPSI, Darmstadt. His research interestsinclude ubiquitous and ambient computing andawareness and privacy in sensor-based environ-ments. Contact him at [email protected].

Thorsten Prante is the deputy head of the Ambi-ente–Smart Environments of the Future researchdivision at Fraunhofer IPSI, Darmstadt, where healso teaches in the Department of Computer Scienceat the Technical University Darmstadt. His researchinterests include context-aware information man-agement and computer-supported cooperative work.Prante is a PhD candidate in computer science. Con-tact him at prante@ipsi. fraunhofer.de.

Daniel van Alphen is a staff member in the DesignDepartment at the Corporate Development Cen-ter of Steelcase North America, Grand Rapids,Michigan. He contributed to this work as a con-sultant to Fraunhofer IPSI after receiving a diplomain industrial design from the University of Arts(UdK), Berlin. Contact him at [email protected].

Richard Stenzel, a scientific staff member at Fraun-hofer IPSI, Darmstadt, and a PhD candidate incomputer science at the Technical UniversityDarmstadt, contributed to this work while he wasa staff member in the Ambiente division. Hisresearch interests include information filtering anddistributed systems. Contact him at [email protected].

Carsten Magerkurth is a PhD candidate in computer science at the Technical UniversityDarmstadt and a scientific staff member in theAmbiente-Smart Environments of the Futureresearch division at Fraunhofer IPSI, Darmstadt.His research interests include ubiquitous comput-ing, user-interface design, and pervasive gaming.Contact him at magerkurth@ ipsi.fraunhofer.de.

March 2005 49