designing ubiquitous computing games – a report from a ... · designing ubiquitous computing...

Designing Ubiquitous Computing Games – A

Report from a Workshop Exploring

Ubiquitous Computing Entertainment

Staffan Bjork1, Jussi Holopainen2, Peter Ljungstrand1 and Karl-PetterAkesson3

1PLAY Studio, Interactive Institute, Sweden; 2Nokia Research Center, Finland; 3ICE, Swedish Institute ofComputer Science, Sweden

Abstract: We report from a Research Atelier that explored how ubiquitous computing could be applied to fun and entertainment. TheAtelier lasted for five days, starting with two days of scenario development and brainstorming activities. This led to three fairly concrete –though very different – game ideas. The background and motivation for the Atelier is described, as well as the method used and the gamesdeveloped.

Keywords: Computer-based entertainment; Computer games; Game design; Ubiquitous computing

1. Introduction

The rapid spread of mobile phones throughoutthe 1990s and the recent investment in 3Gtechnology has sparked a significant interest inmobile gaming, including games that rely oninformation about players’ location as part of theexperience (cf. Nokia Game, Bluefactory, Pico-fun and Botfighters). Other handheld devices(e.g. Cybiko, Gameboy Advance and Game ParkGP32) providing a wide range of availableplatforms for networked, mobile games. At thesame time, computer game developers have beenexploring how to incorporate live data from thephysical world into virtual environments (e.g.Black & White, where the virtual weatherchanges according to the actual weather wherethe game is being played). Attempts at trulymultimedia games have been launched bycomputer game companies (such as Majestic byElectronic Arts) that make use of multipleheterogeneous devices such as mobile phones,fax, email and the Internet.

We see all these developments as the firststeps towards creating fully ubiquitous comput-ing games in which participants on the groundand online participants share a common experi-ence. However, we also believe that we need to

go beyond using only very specific end-userdevices, such as mobile phone handsets andhandheld gaming consoles, if the true potentialof ubiquitous gaming is to be unleashed. Even aJava-enabled mobile phone has inherited most ofits affordances from a traditional telephone, andit is mostly optimized for making phone calls, notmulti-user games intertwined with the realworld.

Besides coarse positioning, most currentcomputer-augmented games make little use ofthe real-world environment as an intrinsic andmeaningful game element. Wishing to explorethe possibilities of new sensor and actuatortechnology in a game context, we organized anextended workshop where researchers from thefield of ubiquitous computing worked for severaldays with the aim of creating a number ofproof-of-concept designs. The workshop aimedat bringing together people interested inexploring ubiquitous gaming, games whereplayers’ real-world context is influencing the(computer-controlled) game play. While ubi-quitous games are situated and played in a realenvironment, much in the same sense astraditional games, their game play is augmentedin real time by computational services, toenhance and leverage the overall gaming

443

Ownership and Copyright# Springer-Verlag London Ltd 2002Personal and Ubiquitous Computing (2002) 6:443–458

experience. In this article we report the resultsfrom that workshop, starting with backgroundto the workshop, methods used and relatedwork. We also describe some observations madeat the workshop, and the three different work-ing groups each describe the games theydeveloped during the event.

2. Background

Since January 2001, the EU has funded theDisappearing Computer (DC) Initiative, aproactive initiative of the Future and EmergingTechnologies (FET) activity of the InformationSociety Technologies (IST) research program toinvestigate how ubiquitous computing andsimilar technologies can affect all aspects ofpeople’s lives. As stated on the project home-page, ‘‘The mission of the initiative is to seehow information technology can be diffusedinto everyday objects and settings, and to seehow this can lead to new ways of supportingand enhancing people’s lives that go above andbeyond what is possible with the computertoday.’’ (http://www.disappearingcomputer.org).Seventeen research projects have been fundedby the DC initiative to fulfil this mission. Tofurther reap benefits from the projects, a specificbudget has been create to enable collaborationand information transfer through a number ofdifferent intra-project activities such as work-shops and one-day meetings. No project withinDC specifically addresses entertainment orgames and we therefore proposed to explorethe area within the context for the DC in ajoint workshop, a so-called Research Atelier.The workshop would focus on merging ubiqui-tous computing and games with the goal ofproducing a number of actual game designs andprototypes. A Research Atelier was ideallysuited for this, as ‘these are small budget andshort time scale projects (<3 months, typically7–10 days) which merge ideas from differentprojects, help build new consortia, or encouragenew developments. These ateliers reflect aphilosophy of ‘‘putting the work back in work-shop, where real tangible results of mutualinterest and benefit to the DC community canbe developed’’ (http://www.disappearingcompu-ter.org/activities.html).

3. Goals and Methods of theResearch Atelier

The goal of the workshop was to bring togethercreative and driven people to exchange experi-ences and ideas about expanding the domain ofubiquitous computing to computer entertain-ment. The long-term goal was to find a groupthat could form the nucleus for a researchcommunity. More explicitly, the following fourgoals were set for the workshop:

1. Gain experience in the designing of ubiqui-tous computing games by the creation of anumber of prototype systems.

2. Create a network of researchers interested inthe field within the DC community.

3. Gather and disseminate information aboutacademic and commercial developmentsrelated to ubiquitous computing games andentertainment.

4. Exchange methods, techniques and techno-logies usable for future research in ubiquitouscomputing games.

Our aim at having a number of game concepts,and even prototypes, was the primary require-ment that directed the working process of theworkshop. Thus, we wished to quickly get theparticipants to work together in small groupstowards a common goal and to work on apractical level. A scenario session would give usthis focus and also be a good starting point. Theorganizers had previously used this method [1,2]in workshops where a number of people fromdifferent disciplines quickly have to create acommon mental model of a possible future.Briefly summarized, the method consists of thefollowing points:

. Identify a focal issue and determination ofvalid time frame.

. Identify key factors.

. Search for the ‘unknown’ driving forcesbehind the key factors.

. Organise driving forces in scale of importanceand uncertainty.

. Pick important and uncertain forces andcreate a scenario matrix or a few scenariosby combining driving forces.

. Evaluate the focal question in each scenario.

. Identify indicators which tell in which direc-tion the environment is heading.

444

Staffan Bjork et al.

Typically, most of these steps are done insubgroups, which present their results for theother groups before proceeding to the next step.For this workshop, we chose to modify themethod by leaving out the two last steps andinstead letting the participants engage in adesign phase. Further, the first step was set byus before the workshop started to be ‘‘What willyou do to entertain yourself after work/school/day-care in 2010?’’ We justified the modificationby the fact that the scenarios were not to be usedto answer a specific question or to try to foresee apossible future but rather to provide a backdropfor the design process.

The use of the scenario method had twoadvantages. First, the creation of future scenarioswas something that was needed as the ubiquitouscomputing games were to be set in futureenvironments and letting the participantscreate these themselves would let them feelinvolved in them. If the scenarios had beenprepared before the workshop, we would havehad to present and explain the scenarios to theparticipants, and would have to hope that theywould find them appealing as potential settingsfor ubiquitous computer games. Secondly, theactivity let the participants work together tocreate something that was outside their normalline of work, thereby avoiding risks of conflictsbased upon the methods and theories used intheir day-to-day work, at least until the groupshad worked together for some time. Also, thishighly cooperative work process between peoplefrom many different countries and from differentresearch fields led to a development of a commonmindset or mental platform for the followingwork with more specific game designs. It alsohelped to overcome some language problems andallowed for the participant to get acquainted.

The specific design processes used for creatingthe games were chosen by the groups themselves,although an overall framework and a set of toolswere provided by the organizers. The participantswere given the task of designing a game andmaking a half hour presentation during the lastday of the workshop.

The organizers provided tools to facilitate anumber of different design processes: rangingfrom whiteboards and post-it notes for brain-storming, to digital cameras and video camerasfor documentation and scenario presentations, tohardware (microcontrollers, weather stations,alarm clocks, fingerprint readers, etc.) for actual

building of working prototypes. The availabilityof these tools was made possible due to thelocation of the workshop at the IT University inGothenburg where a design studio has beencreated for the Masters program in Human-Computer Interaction/Interaction Design. Inaddition, a collection of recreational gameswith miniatures and toy figures, collectable cardgame cards, and toys were provided to functionas well as props and as sources of inspiration.

Finally, the organizers of the workshop madethemselves available as extra ‘muscle’ to thevarious groups during the whole design process, anotion that can compare to how designers arefacilitators while users are players in the ‘design-by-playing’ method [3].

4. Related Work

Researchers have identified computer gaming asa useful prototyping area for exploring innova-tive metaphors, modalities, and hardware with-out having to fully develop infrastructures thatwould be need in other applications [4]. Indeed,within the international research communitythere is a growing interest for studying howcomputers can be used for leisure relatedactivities. Over the last few years, severalinternational workshops have been organizedaround this theme: the ‘Future of Fun’ series[5,6], the ‘Designing Ubiquitous ComputingGames’ [7], the ‘Computers and Fun’ series [8–10], and ‘Funology: designing enjoyment’ [11].However, these have been shorter workshops(typically, one or two day sessions) where thefocus have been on the conceptual level ratherthan grounded on the presently available tech-nology.

Several projects have explored how to uselocation information to enhance computergames. Can You See Me Now? is a game thatinvolves up to 20 members of the public asplayers logged in over the Internet, moving anavatar across a map of a city. The author ischased through the streets of an actual city bythree runners equipped with handheld computerswith GPS receivers communicating over awireless network (Can you see me now?). Pirates!is a multi-player PDA-based game that wasdemonstrated at HUC2k in Bristol in 2000 andat SIGGRAPH in 2001 [7]. Radio-based proxi-mity sensors are used to detect proximity to otherplayers or ‘islands’ in the game. By making

445

Designing Ubiquitous Computing Games – A Report from a Workshop Exploring Ubiquitous Computing EntertainmentDesigning Ubiquitous Computing Games – A Report from a Workshop Exploring Ubiquitous Computing Entertainment

players move through the physical environmentto travel within the game, the presence in thegame was superimposed on the presence in thereal world.

Similarly, a number of experimental gameshave been created using virtual reality oraugmented reality techniques. The MIND-WARPING system [12] uses computer gameapplications to explore aspects of wearablecomputing and augmented reality. In an examplegame, players used one of two forms ofaugmented reality, one with a first-person viewand one with a third-person view, giving twodifferent forms of presence within an augmentedreality. The Mixed Reality Systems Laboratoryhas developed three augmented reality games:AR2 Hockey, RV-Border Guards [13] andAquaGauntlet [14]. Using high-precision bodytracking and see-through head-mounted displays,virtual game elements are superimposed onto theplayer’s worldview. Players see each otherthrough their displays, with virtual helmets andguns rendered on top of their real-worldappearance.

PingPongPlus [15] explores the concept ofComputer-Supported Cooperative Play (CSCP),in which traditional games and sports aretransformed and augmented. Using a sound-based tracking system and a ceiling-mountedprojector to show graphics on a ping pong table,several applications have been designed thatsupport both cooperation and competition.Although restricted to the location of a table,the system shows how one can augment aphysical activity without having to equip userswith hardware. False Prophets [16] is a hybridgame that uses a computer-augmented boardgame and handheld computers to create a gamethat requires non-mediated social interactionbetween the players. Computation is used toprovide private information to the handhelddevices and to create an action-for-energy systemthat frees the players from traditional turn-taking.

Most of the projects above developed gamesto explore the functionality of some technology.The purpose with our workshop was to exploregaming and entertainment, rather than a specifictechnology. Once the idea has emerged, atechnology that suites the purpose is used orseveral are explored to find the one that suitebest. In contrast to the other gaming andentertainment workshops that had the same

focus, how to create better games and entertain-ment ours were very practically oriented. Wewanted to create mock-ups and prototypes andlearn from those experiences.

5. The Research Atelier

The Research Atelier took place between the18th and 22nd of February 2002 at the ITUniversity in Gothenburg. Thirteen peopleparticipated which represented six differentcountries and seven different projects withinthe Disappearing Computer initiative with theaddition of two master-level students from the ITUniversity.

The workshop started after some adminis-trative issues with a presentation of the scenariomethod to be utilized and the schedule for theweek. This was followed with the actualscenarios session focusing on ‘‘What will youdo to entertain yourself after work/school/day-care in 2010?’’ The session was intended tobroaden the mindset of the participants and useda scenario method (cf. Holmquist and Maze [12]for a similar use) that would form a backgroundenvironment for the actual game designs. Duringthe second step of the scenario method a largenumber of key factors were identified and fromthese a number of driving forces were selected.Taking the driving forces that was foreseen asthe most important ones (introvert/extrovert,long-time/short and brief relationships, strong/weak self expression, social/ economical/politi-cal/educational gap, mass consumption/everyoneis a producer) each group created a two-dimensional matrix with one driving force oneach axis indicating if the influence was strong ornot. Each quadrant in these matrixes made up afuture society where the driving forces playeddifferent roles (see Figs 1–4). Each group decidedon one of all these societies, which would thenprovide the context in which the game orentertainment system was designed.

From this point of the workshop, the settingfor the participants’ design was set. Threedifferent groups had to come up with a game orentertainment idea and continue to develop itfurther during the coming days. Each group haddifferent ways of pursuing their work and in thefollowing section each is described in moredetail.

Much of the material provided was used, e.g.the game miniatures were used both as actual

446


game components and for representing players ingames that took place on a city-wide scale. Thisuse of props can be compared to the use ofcardboard mock-ups used by Ehn and Kyng [17]and Brandt and Grunnet [18].

In the last day each group had to make apresentation. Although no specifics were givenfor how the presentation should be done, thegroups invested significant effort into the pre-sentations: one group created a detailed imagescenario based upon pictures taken in the city ofGothenburg while another created a 10-minutevideo scenario showing the prototype in use.

6. Results

The atelier resulted in three very differentgames. The SpyGame is basically a catch theball game, but on a worldwide basis, while theMulti Monster Mania is an extension of thetoday popular trading card games. The Guild isnot a game in the sense of a competition, butrather a social game, it creates the base forfictional guilds among which different games orinteractions can take place. In this section ofthe report, each group describes their game interms of the initial game idea, how it was

447

Fig. 1. Long-time versus short and brief relationships.

Fig. 2. Introvert versus extrovert.

Fig. 3. Massive consumption versus everyone produces.

Fig. 4. Strong versus weak self expression.


developed and issues that came up during thedevelopment.

6.1. The SpyGame

Rob Anastasi1, Steve Benford1, Martin Flin-tham1, Dimitris Riggas2 and Tobias Rydenhag3

1Mixed Reality Laboratory, University of Notting-ham, UK; 2Computer Technology Institute, Greece;3The IT University, Gothenburg, Sweden.

The SpyGame is a game played between a largernumber of different teams that has one goal incommon, to take control and deliver a suite-case.

6.1.1. The game ideaThe SpyGame is based around one of thefuturistic social scenarios that were suggestedduring the brainstorming session. The scenariosuggested that there was a wide socio-politicalgap in a futuristic society that had evolved intotwo distinct groups. The first group wereeffectively the ruling class – they were affluent,well educated, had a large amount of money tospend on leisure time, but also not a hugeamount of time for leisure, as they were too busyworking. This distinguished them from thesecond group, who were said to be quite theopposite of the first group, in that they werepoorly educated, had poor health and housing,and very little money, however as most of themwere unemployed, lots of free time.

The first idea for the game was that it wouldbe good to provide some way of allowing the twogroups to interact, while at the same timeproviding the desired segregation between thetwo groups – the ruling class would not want, wedecided, to mix with the other class, and wouldstill want to exercise a certain degree of controlover them. So, the basic premise of the game isthat one group of people interact on a physicallevel, but are remotely ‘controlled’ in some wayby a second group of people, to achieve acommon objective. The common objectivewould be the ‘delivery’ of a parcel, with theremote users receiving more information as tothe contents of it, and why it needed delivering.The second group of people, the physical players,only receive limited information, via theircontrolling equivalents in the first group.

This basic idea was developed into thefollowing scenario: There are two teams ofusers, each divided into two groups, virtual oronline players and physical players. The game

revolves around one object, or package. The aimis to deliver the package from one physical playerto another in a chain, in such a way that thepackage travels from one side of the game area toanother – the game area could be a city, forexample. One team is trying to make thepackage travel in one direction, while theother team is trying to make it travel back inthe opposite direction. The team only has alimited amount of time to make the handover ofthe package, otherwise it will lock the currentplayer out of the game, effectively handing thepackage over to the opposing team.

Each virtual player is twinned with a physicalplayer, who they can talk to via mobile phones.The physical player receives instructions onwhere to go, and what to do, by the virtualplayer. The virtual player can coordinate theiruse of their own physical players with othervirtual players in a virtual chat environment.This environment also provides extra informa-tion as to the whereabouts of the parcel to all ofthe virtual players, information that is notavailable to the physical players. This informa-tion is detected by the parcel, and then relayedto the chat environment.

As the parcel changes its physical location,different physical players are called in to play thegame – obviously only players in the currentlocale would be able to take part, so the game

448

Fig. 5. The main areas of play in the game, and the flow ofinformation between the areas.


locates players that are in the right area, anddistributes their contact details to the virtualplayers, their phone numbers for example.Physical players would register their locationsand times they were available to play –obviously, it is beneficial for one’s team in oneis available to play as much as possible.

Standard physical players play the part ofspies, or scouts, merely relaying information backto their team. After the accumulation of gamerewards, i.e. successfully taking part in the game,physical players may rent detection equipment –devices that give them more information as tothe location of the package, and the ability tolock out its current carrier when nearby, againgiving their team the chance to acquire theparcel. The standard physical players also havethe ability to trigger the parcel’s lockingmechanism, but only by reading a secret codeon the parcel, which would be unlikely unlessthe carrier was very careless, and then relayingthis code to the system.

The information broadcast from the parcel isnot very precise, so as not to easily give away itslocation. It relays information including acontinuous audio stream from its location, plussensors give data on its ambient surroundings, forexample, how warm it is, whether it is in light ordarkness, the humidity, etc. There is a sensor onthe handle of the parcel, so it can tell if it isbeing held or not. There is a fingerprint reader,which the courier must register their fingerprintwith before being able to pick up the parcel. Thisenables the parcel to lock out users who have runout of time. Finally the parcel contains apositioning system, so it knows where it is.This information is not for the players directly,but to allow the system to figure out whichplayers to involve in the game, based on theparcel’s location.

The chat environment consists of a publicchat room, which virtual players from both teamsinhabit. The only way two physical players canmeet to exchange the parcel is by arranging ameeting through their controlling virtual players.The virtual players are given information aboutthe area the game is taking place in, in the formof a map. The parcel broadcasts the data aboutits ambient surroundings into this chat room.The idea of the game was to involve someconcept of espionage, or spying. The physicalplayers have certain aspects of this in their halfof the game, as they do not know which members

of the public in the same place are playing thegame, or, if they are playing the game, whichteam they are on. To create this atmosphere forthe virtual players, participants in the chat roomcan spy on other players while they talk. Weproposed to facilitate this as follows.

Virtual players can send private messages toone another; however, there is a catch, some ofthe messages are also available for viewing by thechat room as a whole. From a one line message,up to half of the words, chosen randomly, arebroadcast in their original order to the publicchat area. This gives the opposition chance to tryand work out where the meeting is going to bearranged. Obviously, this gives the team who areoperating as the couriers an opportunity tospread disinformation, and thus enhancing theidea of not being able to trust any informationyou may receive from those around you for thevirtual players.

6.1.2. Game IssuesTo try and find out whether the above ideawould be suitable for an actual game, we rapidlyprototyped parts of the system, and then roleplayed an example game, at first as a board gamethen to make a video.

To run through the game idea as a boardgame, we constructed a tabletop representationof a physical and virtual space, and then eachmember of the group took the role of a playerplaying the game. We then attempted to play agame, noting the interactions between players,and what information was transferred.

To see whether the technical infrastructurefor the game was roughly feasible, we prototyped

449

Fig. 6. Testing a fictional scenario with role playing.


two key pieces of the system. The first was thechat environment for the virtual players. Thiswas a relatively trivial piece of software en-gineering, but it allowed us to test the mechan-ism for eavesdropping on private messages. Thechat room client provided an interface that agame server client could join, that randomlypicked physical player details from a database,and then distributed this information to theother clients.

The second was the parcel itself. This was asmuch an exercise in prototyping the physicaldesign of the parcel as making sure all of thefunctionality worked. A briefcase was acquired,and enhanced in the following ways:

. Adding a kitchen timer – this is the time thatthe courier has remaining to deliver theparcel. The idea was to have this controlledby the lap top, but time restrictions did notallow this.

. Adding a flashing light – this is the alarm,when the timer reaches zero, the alarm willsound and the flashing light will flash.

. A laptop was stored within the briefcase. Theaim eventually was to have this controlling allof the devices mounted on the briefcase, to actas a game controller. For this demonstration,the laptop was merely a client connecting to awireless network, rather than acting a basestation itself. For this a wireless wave LANantenna was inserted in the skin of thebriefcase, to allow good connectivity. Thisworked very well within the building, but itshould be realized that the building hadseveral wave LAN base stations installedanyway.

. An omni-directional microphone was discre-tely mounted in the skin of the briefcase,which was then connected to the laptop. Thelaptop relayed the audio stream over thewireless network to the chat environment,using the MASSIVE-3 audio server.

. A portable wireless weather station wasinserted in the front of the briefcase. Thisrelayed information to a base unit that in turncould be connected to a laptop computer. Thelaptop again implemented the chat clientdescribed earlier to crudely publish this data inthe chat room.

. A finger print reader was mounted on thebriefcase, but was not integrated into the restof the system.

Our experiments showed the sensors in thebriefcase did give some indication as to itslocation. Users in the chat room could here thevoices of people in the same room as thebriefcase, so it gave adequate audio coverage.Taking the briefcase outside caused a significantdrop in the temperature, and placing the brief-case next to a running water tap caused anoticeable change in the humidity level regis-tered.

6.1.3. The ResultThe prototyped game seemed to meet many ofthe initial goals. The division of players into twogroups, the virtual and physical players, fittedwell with the concept of the social/economicgap. The virtual players make the high leveldecisions, and control the physical players andthe overall flow of the game. The physicalplayers are highly dependent on their virtualminder, while being the mechanism throughwhich the game progresses.

The game provides opportunities for varyinglevels of involvement in the game. Virtualplayers have the opportunity to join the gameat any time and from any place, and can becomemore involved by forging relationships withother online players and gaining recognition.Physical players can join whenever the game isclose to their physical location and requirenothing more than a mobile phone to play.Over time they can build a relationship withonline players and are rewarded for playing withdevices that assist them.

Using a mobile node (the briefcase) as thefocus of the game supports this short-term,location based involvement, and also allowsthe game to run for a very long period of time,potentially indefinitely. It is, however, impossi-ble to verify this since the paper mock-up onlyallowed a very limited subset of the game to beplayed, and the prototype was not playable. Thatsaid, they did allow for many elements of game-play to be explored and developed and (with thenotable exceptions of tracking the briefcase andidentifying players to the briefcase) most of thesoftware and hardware was functional.

The game, in its current state, demonstratedthe basic concept well, but without furtherdevelopment it is impossible to say whether thegame would function over a large area, over along period of time, and whether the game-playcan be balanced so that the game can move from

450


area to area to involve as many people and variedlocations as possible without making it too hardto intercept the briefcase.

6.2. Multi Monster Mania

Mads Haahr1, Sus Lundgren2 and Niels Reijers1

1Distributed Systems Group, Department of Com-puter Science, Trinity College, Dublin, Ireland;2The IT University, Gothenburg, Sweden.

6.2.1. The IdeaMulti Monster Mania is a card based gamesystem featuring monsters that is set in aGibsonian future with quite large social, poli-tical, educational and economical gaps. One ofthe ideas behind it is therefore that it should bepossible for smart streetkids to create their owngames or artwork for the game, earning somemoney on it. One also has the option to breedown monsters, selling the cards. The variousincitements to engage in the game are thus toearn money, to have fun, to get an outlet forcreativity and/or to collect cards. These incite-ments can attract players from any background;they will be attracted by different incitements, orrank them differently, but will nevertheless play.

Over the last couple of years, Personal DigitalAssistants (PDAs) have decreased in size whileincreased in functionality and now come withcolor touch screens, short range wireless networkconnectivity (e.g. IR or Bluetooth) and areasonable battery life as standard. Over thenext years, we expect a further decrease in sizecombined with an increase in hardware com-plexity in the form of sensors that allow thedevices to be aware of their immediate surround-ings, such as the presence, location and orienta-tion of other devices. The present game conceptis based on the assumption that devices withtouch-sensitive color screens and computationalpower equivalent to a current PDA in 6-8 yearswill be the size of a playing card, a couple ofmillimeters thick and cheap (on the order of afew euros).

Multi Monster Mania is a card game wherecards have computational power that allowsthem to be dynamic and limited communicationcapabilities that allow them to interact overshort distances. The game concept combineselements from traditional trading card games(such as Magic the Gathering) with that of theTamagotchi to form a flexible game platform. In

addition, the game contains an evolutionaryaspect that allows players not only to use thecards to play games but also evolve new cardssuitable for particular games. Cards deemedvaluable can also be traded.

There are four different types of cards: gamecards, location cards, monster cards and modifiercards. Game cards specify game logic for aparticular game, such as chess-type strategygames, Pokemon-style fighting games, peacefuldoll-house type games or more traditional boardgames such as the Hatchery Game describedbelow. Location cards embody a location (e.g. adesert, the magic forest, or Dracula’s Castle) inwhich a game can be set. Monster cards containthe actual entities with which the games areplayed according to the rules specified in thegame rule cards. Modifier cards can be used tomodify any of the other cards and are used torepresent for example weather (to modify loca-tion or game logic) or items (to modify theabilities of monsters). An actual game consists ofa single game card and a set of cards of the othertypes. During the game, the cards interactaccording to the rules specified by the gamerule card. All cards also carry a visual animatedrepresentation of their content on the front.

The monster cards are of particular interest.Each monster is identified by a genome thatdefines the monster’s appearance and abilities,such as the number and types of arms, legs, eyes,etc. These characteristics determine how good amonster is at certain games. A monster withmany legs and at least a couple of arms couldperhaps be suitable for an American footballgame. In addition to the genome, a monster cardcarries a biography of the monster in the form ofgames played, training activities performed,ancestry, offspring, etc.

6.2.2. Design issuesInitial ideas included a game board carrying allthe logic, combined with ID-tagged cards. Thiswas abandoned, since it would be harder toexchange cards, since the code had to betransferred also. A kind of fantasy combat gamewas tried out; but turned out to be quite similarto existing games. The concept of a fantasyfootball game was also explored which led to theidea that card orientation could be significant insome games.

When we had come up with the idea ofmonsters whose physical appearance are deter-

451


mined by their genome, we experimented with aboxing application for monsters. This led us toconsider how various non-physical properties,such as courage, could be derived from thegenome. This is an interesting idea, that wenever had time to develop.

6.2.3. The Monster GenomeA genome consists of a sequence of gene groups.Each group contains two types of genes: instancegenes and modifier genes. The instance genesdefine the degree to which a certain character-istic is present in a monster. For a given type ofinstance gene, modifier genes can modify thatcharacteristic. For example a monster with threeeyes on sticks would have three ‘eye’ instancegenes, and at least one ‘eye on stick’ modifiergene. There is an order of dominance in themodifier genes. For example if the ‘human eye’gene is the dominant eye modifier, a creaturewith human eyes can also have the ‘eye on stick’gene, but this would be suppressed by thedominant ‘human eye’ gene.

The genome is of variable length, and morecomplex monsters can have longer gene stringsto describe their appearance. In trading cardgames is it common that periodically expansionsets of cards are released. These contain newcards, adding new options to the game. Ourvariable length genome gives us a great deal offlexibility for expanding the game after theinitial release. Besides new games, locationsand modifier cards, an expansion can containnew creatures that possess new properties. Ifthese properties are useful, this will be anincentive for players to keep breeding monstersand integrate this new property into theirexisting group of monsters.

Breeding monsters is done by placing twomonster cards and a blank card beside eachother. (Monsters do not have genders as such.)

When this is done, there is a chance that the twomonsters will mate and produce offspring. Thelikelihood of the monsters mating depends onhow similar they are. Similar monsters are morelikely to reproduce than dissimilar monsters. Todetermine the chance, we examine both mon-sters’ genome and count the number of commongenes and the number of genes only found in oneof the two. We divide the number of differentgenes over the number of common genes. Thelower the resulting number, the more similar themonsters are. We determine the chance ofreproduction by feeding this number into afunction that only returns a high probability forlow genome differences.

If two monsters mate, the following algorithmis used to determine the genome of the child. Foreach ‘gene group’, the genes of both parents arecombined in a pool, and from this we select arandom number of genes for the child. Theprobability of selecting a certain number of genesis determined by a bell curve. This ensures thatif, for instance, both parents have two arms, thechild will most likely have two arms but couldhave any number between 0 and 4.

The intention is that the reproduction algo-rithm will produce results that seem natural tothe player, while still allowing for minorvariations and the occasional freak result. Bykeeping the bell curve steep, we make sure thatproperties shared by both parents are copied inthe offspring, while still allowing a slight chancethat two two-armed parents will produce a four-armed child. When this happens, this may bevery valuable to the player for certain games, orcompletely useless. We think this will keepbreeding monsters an interesting aspect of thegame.

The grouping of the genes is necessary tomake the bell curve work. If the genes were notgrouped, two two-armed and two-legged mon-sters could easily get a four-armed and zero-legged child. Of course, the fact that some genesmay be recessive can lead to interesting resultswhen the child turns out to have a property thatboth parents lack.

6.2.4. The ResultEffectively, Multi Monster Mania is a flexibleand extensible game framework rather than anactual game. Games can be implemented in theform of game rule cards, and the monsters can beused as pieces in such games. An example of this

452

Fig. 7. Examples of monsters with different genomes.


type of game is the Hatchery Game describedbelow. The framework’s evolutionary aspectprovides a type of meta-game in its own right,the goal of which is to breed monsters that areeither suitable for particular games or just plaininteresting.

The framework is extensible in many ways.First, new games can be implemented andexisting monsters used in them, even if themonsters did not exist when the game wasdeveloped. Similarly, new monsters can be bredand used in existing games. In addition, newgenes can be invented by the frameworkdesigners. Although such genes cannot beinserted into living monsters, the breedingalgorithm assures that monsters with new andold genes remain compatible and that offspringcan have characteristics from both.

The remainder of this section describes theHatchery Game, an actual game designed as anapplication of the framework, and discusses howthis game uses the framework to implementfunctionality that would not be possible other-wise.

The Hatchery GameThe game is played with teams of monsters and isset in an egg-hatching factory. Each player ownsfour eggs stored in an egg-hatching machine atthe centre of the board, and the objective is tohatch one’s eggs before the other players. Eggscome in four colours, and each team has one ofeach. When in the hatching machine, eggs growor shrink depending on the colour of themachine – the closer this is to an egg’s owncolour, the better it grows, and finally it hatches.Monsters can change the colour of the machineby feeding pellets of different colours into it.Pellets grow in pellet fields located at the cornersof the playing area.

Points are scored for hatching eggs – thesooner the better – and by cooperating withother players. The game evolves around thefollowing:

. Getting pellets of the right colour and feedingthem into the factory.

. Cooperating to get more pellets and points.

. Saving shrinking eggs by temporarily takingthem out of the hatching chamber.

The game ends when all eggs have been hatched,and whoever has the most points wins.

To play, one needs the following cards:

. A rule card containing the rules for the game.

. Location cards; one game field entrance, pelletfields where pellets grow, as many hatchingchambers as there are players.

. Monster cards; each player picks a teamconsisting of three monsters. Monsters withmany legs and/or wings move faster, monsterswith many arms can harvest and carry manypellets at a time.

The hatching chambers are laid out in a row inthe centre to represent the egg-hatching ma-chine. Other cards can be placed anywhere onthe table. Distances between cards are virtual,and there is the same distance between alllocations. In addition to the locations alreadymentioned, there is one extra location called thevoid, which represents the state of not beingnext to a location card. When a monster shall bemoved, the player physically moves the monstercard from one location card to another. At thispoint, the cards compute how long time it willtake for that particular monster to perform themovement. Similarly, all activities take a certaintime to carry through.

In addition to moving a monster card,activities can be indicated by dragging pelletsin-between cards or an egg from the hatchingchamber to a monster or vice versa. Typicalactions include harvesting pellets from the fields,trading pellets, saving eggs, cooperating withother teams, and stealing pellets. While perform-ing an action, the monster is considered to bebusy and cannot be disturbed. Thus, the game isnot turn based; players can activate any one oftheir non-occupied monsters any time.

The hatchery game contains many dynamicaspects that cannot be easily implemented usingtraditional board game methods. First, thedynamic and individual nature of the monstersis a vital aspect of the game. The timing ofmonster actions (such as travelling virtualdistances, juggling eggs, harvesting pellets, etc.)subject to individual monster characteristicswould not have been possible without compu-ter-enhanced cards. The game environmentitself also contains dynamic features thatcannot easily be implemented with conventionalmethods: the continually changing colour of thehatchery, the constant effect of this colour onthe growth and shrinkage of the eggs, the growthrate of pellets in the fields, etc. Also, it would not

453


be straightforward to keep the points attained byplayers secret for everyone until end of game.

6.3. The Guild

Jennica Falk1, Asa Harvard2 and JonasSoderberg3

1Media Lab Europe, Dublin, Ireland; 2NarrativityStudio, Interactive Institute, Sweden; 3ICE, Swed-ish Institute of Computer Science, Sweden

6.3.1. BackgroundOne of the discussions brought up during thescenario exercise dealt with the question ofbroadening the interest and participation incomputer games. Media companies, researchersand designers all seem to agree that computergames hold an enormous potential in the future.Certainly, new technologies have to be put inplace in order for this potential to be realized.However, there is also another angle to this:computer games tend to remain a pretty welldefined genre, with a huge and committedaudience, but also with large parts of thepopulation not particularly interested in them.

In terms of the Future Worlds defined in thescenario phase, the Guild did not use one of thedeveloped worlds but envisioned a future worldwith the following characteristics from several ofthe worlds: Strong self expression, Everyone produ-cing, Extrovert and Short-term relationships.

6.3.2. The Guild ObjectThe Guild concept took its starting point in thenotion of ‘Guild Object’, as explored in one ofthe previous exercises during the workshop. TheGuild Object is a physical token worn by players.The Guild object helps players to recognize eachother as friends or foes, and it enables players tobring the game with them out into everydaysituations or environments. The Guild objectcould range from a piece of jewelry to a bodyimplant, with some computational part, like aRFID tag. However, in the discussions aroundthe Guild concept we came to realize that a‘plain’ non-computational artifact could fulfilmany of the uses of a Guild object.

6.3.3. Play CommunitiesIn continuation of the discussions about targetgroups for computer games, we started bydiscussing motivational factors in play andgaming – with a focus on those factors that areexternal to the game itself, such as need for

excitement or relaxation, identity and socialnetworking. We spent quite some time listingvarious examples of ‘play communities’ i.e. socialgroupings around a shared repertoire of playactivities. How do they work? Why do peopleparticipate? The list of organizations had func-tions ranging from social clubs, such as OddFellow and other free masons type organizations,to golf clubs, role playing organizations, religioussects and societies, such as the Moon Sect or theChurch of Scientology, or organizations likeHells Angels and al’Qaida (the latter being notexactly ‘play communities’, but still interestingfor the comparison).

In our informal brainstorm around these playcommunities some characteristics were recurrent:

. One of the main activities in a play commu-nity is to manage how new members getintroduced and gradually become integratedin it. Secrets and the process of revealingthem is important as a part of the process ofinitiations.

. The progression of getting to know thecommunity, as well as rising in rank withinthe community, is an important motivationalfactor in participating.

. Shared stories, and even more, shared secretsare important in keeping the play communitytogether. Secrets can be of many shapes: a sectmember may be prohibited to communicatethe revelations of the sect to a non-member,members of clubs for extra rich or talentedpeople are advised not to display their talents/riches in public, communities engaged inpractices that are taboo or illegal sticktogether around knowledge that would hurtboth members and association if it was madeknown to outsiders.

. Members of a play community form friend-ships and structures for mutual support that gobeyond the realm of the play activity itself.

We also listed media and events these commun-ities use in order to keep the organization runningand progressing: Uniforms or various costumes,which often signify rank and status within theorganization, parties and other social gatherings,initiation rituals for petitioners, secret hand-shakes or identities, symbols and tattoos to markaffiliation, address books or web sites givingmembers access to other members’ personalcontacts, various facilities such as club houses,systems for favors granted by fellow members.

454


6.3.4. Game ideaWhen thinking up our game, we wanted toexploit the very strong forces that are at playwhen people join and maintain memberships inthese kinds of various organizations and net-works. We also wanted to explore the notion ofubiquitous gaming, and create a game that isplayed by engaging with physical spaces andobjects, using the everyday settings of the playersas a part of the game experience.

The backbone in the Guild game is anarrative framework and a social structure thatproposes a number of events and interactions tothe players. We imagine heterogeneous uses fortechnological devices in the game – but thesecan be determined locally.

We explored a game idea along the followinglines:

. a narrative backbone structure that proposes aseries of events and interactions to the players;

. a network of ‘guilds’ where new guilds can beadded to the game;

. a LARP (live action role-playing game) wherethe players’ everyday environments and ac-tions take on a second meaning within thegame context;

. Use of portable devices and new technology inthe game (whether it will be a part of thestructure or a part of the content is to bedetermined);

. Two overriding themes: competition betweenenemy guilds – management of new memberswithin each guild.

The narrative frameworkThe following narrative framework for the Guildgame highlights interesting issues but is not adetailed story, rather a set of logical devices tomake if not believable so at least acceptable – inthe terms of fiction – the assumptions of thegame. Two rivalling alien species land on earth,both with the mission to conquer humanity andtake possession of its planetary resources, and todo so before the rivaling clan. Both clans striveto create a new society extending their own, andbuilt on their own values as adopted by theirhuman followers.

They begin their quest by abducting membersof the human race. The abducted humans aremodified as to their physical and mentalessences, and they start to work for the finaltakeover of power. However, waiting for the new

society, the abducted humans continue to livetheir everyday life as usual, and the signs that aperson has been taken over by the aliens aresubtle. Only an initiated person can read them.

For the new abductee, a second life in parallelto their everyday life starts up. The first event isthe abduction itself. Later, the new abductee getsinitiated in the rituals and routines of the alienclan. When the abductee becomes sufficientlyknowledgeable, he or she may be sent out onoperations aiming to prevent the other clan fromtaking power, or to abduct new humans.

The ‘promise’ to the abductee is that they willrise in rank and position when the new leaderstake over the world. You begin as a low rankcitizen, and work your way through missions togain higher positions. This rise is also expressedin the artefacts, clothes and symbols worn by theabducted persons.

User scenarioThe following user scenario describes the con-cept not from the fictional framework but therealm of players of the game.

An interested person, Sheila, gets to hearabout the Guild Game and wants to join it.Through friend, websites, or other informationchannels she decides which Guild to join. Sheilalikes the graphical profile of the NKMRR guild,and she knows slightly some of the members –they are people she would like to hang out with.She contacts the Guild in question, and an‘abduction’ is arranged for by the Guild. Theabduction is seemingly a power overtake. In realterms, it can be described as a welcome party. Atthe Abduction, Sheila buys a token for theNKMRR Guild: a ring with an ID-tag containingher Guild Identity. It serves both to log in to the

455

Fig. 8. A new member being abducted by one of the guilds.


Guild website and to show her affiliation to otherGuild players.

After the abduction, Sheila attends toNKMRR events once in a while. The clubevenings are often mostly a social event, and anoccasion to use the NKMRR costume – startingwith the ring Sheila has gradually constructed/bought/collected a full costume for her NKMRRalter ego ‘Aliehs’. Some of it can be worn underregular street clothing without calling on theattention of non-players.

At her work, Sheila starts to notice some ofher colleagues wearing Guild tokens. Some workcolleagues also show interest in the game, andshe later on introduces them to the Guild, afterwhich they are ‘abducted’ as well. After a year,Sheila is responsible for arranging new abduc-tions, which puts her into contact with a lot ofnew people.

6.3.5. Design IssuesThe Guild concept was formulated in strongenjoyment. There are many issues in the conceptthat are not resolved – but this is also aconsequence of the intention to focus atmotivational factors external to the game itself.This division was of course difficult to maintain,and at some point we were all enjoying inventingnew characters and fantastic guilds – really doingthe work we had envisioned the players to dothemselves.

There are some problematic issues in ourconcept that we believe are relevant for thegenre of ubiquitous gaming overall, and that wewould like to describe to open up for a discussionaround them.

One issue is the idea of ‘giving up one’s freewill’. On one hand this is a very useful device in

order to have players enter the game, and to helpthem forget about everyday obligations for awhile. On the other hand, the idea of obeying aguild gives unpleasant resonance to totalitariansystem or religious sects.

Another central idea in the notion ofubiquitous gaming is that it should be possible– and interesting – to play a game at the sametime as doing some other activity. Ubiquitoustechnologies open possibilities to layer fictionallandscapes on top of real places. But do we havethe necessary intellectual and emotional cap-abilities to handle parallel lives/worlds? Doespresence in one context (real or fictional)require absence in other context? Can a splitmind enjoy a full experience?

In the workshop discussion there was ageneral agreement about identity building as animportant driving force in play and gaming. Butthere is an interesting issue in the role of fictionin relation to identity. How far can ‘play’identity building be pushed without spillingover into a persons ‘real’ identity? Are we readyto confront the real consequences of fictionalidentity building?

To summarize the different design issues:

. In designing a game there is a need to defineand address the social strategies/motivationalfactors of the ‘real’ players of the game, inaddition to that of the characters in the game.

. Ubiquitous gaming, i.e. game play not identi-fiable by a specific location or by a timeframeraises issues about how to distinguish playfrom reality.

. Any person’s choice of games (and of leisureactivities at large) as a highly significantaction in defining oneself and one’s socialcontext.

7. Discussion

The workshop showed that there is great interestin Europe in the research community forexploring the use of ubiquitous computing forentertainment. It also demonstrated that withrelatively few resources interesting concepts andideas could be quickly developed though withinsuch a short time it was impossible to perceiveany deeper knowledge within the area. Eventhough the scenario method used in the work-shop gave a structure for the workshop, in suchshort time the actual processes of designing the

456

Fig. 9. An example of a guilds graphical tag.


game concepts was difficult to formalize as wellas to find a common language to describe them.What methods can be used? What are thebenefits of different methods? How do youdescribe these results?

Several researchers within the field of parti-cipatory design have proposed the use of gameswithin the design process [3,19], and it couldseem, as this would have been ideal method forthe workshop. Indeed, many of the propsprovided were from games and most of theparticipants of the workshop were active gamers(within role-playing games, computer games,MUDs, board games, etc.). To use miniaturesto role-play the citywide game on a table couldbe seen as some kind of game, but the groupnever described it as such themselves. None ofthe participants suggested a method based onplaying a game to design the game. The use ofsetting up a game to create another game mayhave turn out to be counter-productive andconfusing, as one would have to remember whichgame is which.

Scenarios, structural modeling of game me-chanics, body storming, paper prototyping, rapidprototyping were all used but did not offer a wayof discussing interaction and game play in acommon language. It is our belief that theavailability of such a common language wouldhave made further refinement of the conceptspossible by speeding up design discussions toallow further design iterations or testing.

Fortunately, there seems to be an interestamong researchers and practitioners to developways of supporting the collaboration and com-munication when designing computer applica-tions and systems for entertainment purposes.The Computer Games and Digital CulturesConference (http://www.gamesconference.org)in 2002 presented work from the humanities tostudy games as its own subject with its ownmethods and terminology. Computer gamedevelopers have within their own forums pro-posed several theories and methods (c.f. Krei-meier’s [20] design patterns and Church’s [21]formal abstract design tools) to help design workand actively support research within the subject.By taking carrying on these projects and themore technically-oriented work presented in therelated work section, and learning from the otherfields, we believe that computer-driven enter-tainment can become as social and wide-spreadas traditional games.

The workshop demonstrated the growinginterest of ubiquitous gaming and entertainment;we hope to see more similar work in the nearfuture.

References

1. Bergqvist J, Bjork S, Borjesson M, Ljungstrand P.Scenarios for multidisciplinary design purposes: Experi-ences earned. Proceedings 22nd IRIS Conference(Information Systems Research Seminar In Scandina-via), University of Jyvaskyla Press, 1999

2. Holmquist LE, Maze R. Scenarios: Methods and Results.Project Deliverable No. 4. for Smart-Its IST-2000-254282000. (Available at http://www.viktoria.informatik.gu.se/groups/play/projects/smartits/

3. Ehn P, Sjogren D. From system descriptions to scripts foraction. . In: Greenbaum J, Kyng M (eds). Design atWork. Lawrence Erlbaum, Hillsdale, NJ 1991 pp 241–268

4. Starner T, Leibe B, Singletary B, Lyons K, Gandy M, PairJ. Towards augmented reality gaming. Proceedings ofIMAGINA 2000

5. Holmquist LE. Future of Fun. Viktoria Institute, May 81998. http://www.viktoria.se/play/fof/fun98.html

6. Holmquist LE, Thomas M. Future of Fun 2. Held inconnection with HUC 99, Karlsruhe, Germany 1999.http://www.viktoria.informatik.gu.se/groups/play/fof/

7. Bjork S, Falk J, Hansson R, Ljungstrand P. Pirates! Usingthe Physical World as a Game Board. Interact 2001, IFIPTC.13 Conference on Human-Computer Interaction,Tokyo, Japan 2001

8. Monk A et al. Computers and Fun 1. York 1998 http://www-users.york.ac.uk/~am1/Comp&Fun.html

9. Monk A et al. Computers and Fun 2. York 1999 http://www-users.york.ac.uk/~am1/candf2.html

10. Monk A et al. Computers and Fun 3. York 2000 http://www-users.york.ac.uk/~am1/candf3.html

11. Monk A, Hassenzahl M, Blythe M, Reed D. Funology:designing enjoyment. Extended Abstracts of CHI,Conference on Human Factors in Computing Systems.Minneapolis, MN 2002

12. Starner T, Leibe B, Singletary B, Pair J. MIND-WARPING: Towards creating a compelling collabora-tive augmented reality game. Proceedings IntelligentUser Interfaces (IUI) 2000 pp 256–259

13. Tamura H. Real-time interaction in mixed reality space:Entertaining real and virtual worlds. Proceedings ofIMAGINA 2000

14. Tamura H. Emerging technologies: point of departure.SIGGRAPH 2000, New Orleans, LA 2000

15. Ishii H, Wisneski C, Orbanes J, Chun B, Paradiso J.PingPongPlus: Design of an athletic-tangible interface forcomputer-supported cooperative play. Proceedings CHI’99 1999, pp 394–401

16. Mandryk RL, Maranan DS, Inkpen KM. False prophets:Eexploring hybrid board/video games. Extended Ab-stracts of CHI, Conference on Human Factors inComputing Systems. Minneapolis, MN 2002

17. Ehn P, Kyng M. Cardboard computers: mocking-it-up orhands-on the future. In: Greenbaum J, Kyng M (eds).Design at Work. Lawrence Erlbaum, Hillsdale, NJ 1991pp 169–196

18. Brandt E, Gunnet C. Evoking the Future: dma and propsin user centered design. Proceedings PDC 2000, NewYork 2000

457


19. Iacucci G, Kuutti, K, Ranta M. On the move with amagic thing: role playing in concept design of mobileservices and devices. Proceedings DIS2000, DesigningInteractive Systems, New York City. ACM Press 2000,pp 193–202

20. Kreimeier B. The Case For Game Design Patterns 2002http://www.gamasutra.com/features/20020313/kreimeier_03.htm

21. Church D. Formal Abstract Design Tools. http://www.gamasutra.com/features/19990716/design_tools_01.htm 1999

Web sourcesGamePark GP32 http://www.gamepark.co.kr/

Black & White, http://www.bwgame.com

Bluefactory. http://www.bluefactory.com

Botfighters. It’s Alive http://www.itsalive.com/

Can you see me now? http://www.canyouseemenow.co.uk/

Cybiko. http://www.cybiko.com/

Smartits_D4_210102.pdf (verified Feb 11, 2002))

Majestic, Eletronic Arts. http://www.majestic.ea.com/

Nokiagame. http://www.nokiagame.com

Nintendo Gameboy Advance. http://www.nintendo.com/systems/gba/gba_overview.jsp

Picofun. http://www.picofun.com

Correspondence to: Mr S. Bjork, PLAY Studio, InteractiveInstitute, Box 620, SE 40530 Goteborg, Sweden.

458


designing ubiquitous computing games – a report from a ... · designing ubiquitous computing...

Documents