12
mixed reality game ジオジオスタンプラリー niryuu [email protected] 20091028日水曜日

Nottingam Gms

Embed Size (px)

Citation preview

Page 1: Nottingam Gms

mixed reality gameと

ジオジオスタンプラリーniryuu

[email protected]

2009年10月28日水曜日

Page 2: Nottingam Gms

mixed reality game• 自己紹介:モバイラー、社会学(埼玉大学大学院、エスノメソドロジー、ヒューマンインタラクション、コンピュータ支援共同作業)

• Mixed Reality Game:位置情報などを使い、現実世界の情報と、仮想世界が重ねてある(混合現実感)ゲーム

• 今ではAR(拡張現実)と呼ばれているが、その辺はいろいろある

• ノッティンガム大学Mixed Reality Labと、メディアアーティスト集団Blast Theoryが、2001年頃から共同で作品を作っている

• 会話、ログなどを利用して、ゲーム内で人々がどう最新の技術を理解して、ゲームを行っているかの社会学的な研究も

2009年10月28日水曜日

Page 3: Nottingam Gms

Can You See Me Now?• ネットと現実をつなげた「鬼ごっこ」

• GPSとネットでつながった「ランナー」が、オンラインの3D空間を動く一般人を捕まえる(南極からの参加者も)

• 世界中で、5日間ほどの期間ずつ行われた

the other players and the three runners.The game labels avatars with players’names and highlights the runners witha red sphere that makes them highlyvisible, even from a distance. Onlineplayers can also select a zoomed-outmap view, which shows more distantplayers and runners and text labelswith key location names. Finally, theycan view and enter text messages andhear the runners’ audio. Figure 1ashows an example online player’s inter-face, with the player’s avatar in theforeground and a runner close by in thebackground. Figure 1b shows the inter-face in map mode.

The runners’ experienceThe runners move through the streets

and can see online players’ and other run-ners’ positions on a handheld map, see theplayers’ text messages, and communicatewith one another using walkie-talkies.

The game delivers the runners’ inter-face on an HP Jornada handheld com-puter from a server in a nearby buildingover an 802.11b wireless local area net-work. A GPS receiver plugged into the

computer’s serial port registers a runner’sposition, which is sent back to the serverover the wireless network. We built thisequipment into a robust outer jacket (seeFigure 2).

Given the iPAQ’s small screen size,the runners’ map lets them zoombetween a global view and a close-uplocal view centered on their currentposition (see Figure 3). The runnersused walkie-talkies with earpieces anda head-mounted microphone. They car-ried digital cameras so that they couldphotograph the physical location wherethey caught each player. These picturesappeared on CYSMN’s archive Web sitesafter the event.

Deploying CYSMN required supportfrom a behind-the-scenes technical crew,housed in a central building in the gamezone (along with six public terminals forlocal online players). This three-persontechnical crew in the control room ranand managed the online server and sup-ported the runners. This team used, forexample, a game space overview andinterfaces for managing queuing players,monitoring the wireless network, dis-

playing the runners’ status (includingconnection status) and GPS status, andplaying the game. The control room andrunners also used a walkie-talkie chan-nel separate from the 802.11b networkfor communication.

Causes of uncertaintyCYSMN’s most significant source of

uncertainty was the GPS. In Sheffield,we used standard GPS with GarmineTrex receivers, and the game zonespanned a mixture of open urbanspaces with a few narrow, built-up sidestreets. Our system log analysis showedthat reported GPS error ranged from 4m to 106 m with a mean of 12.4 m anda standard deviation of 5.8 m. In Rot-terdam, we upgraded to differentialGPS and used Trimble Lassen LP re-ceivers with Sarantel antennas. Thegame zone contained a similar mix ofopen spaces, several of which lookedout over open water (with a good viewof the sky to one side) and narrowerbuilt-up streets toward the game zone’scenter. Log analysis showed that in thiscase, reported error ranged from 1 m to384 m, but with a lower average errorof 4.4 m and a standard deviation of4.9 m. To improve accuracy, we con-figured the receivers to ignore satellitesthat were low in the sky (below 15de-grees), although this sometimes madegetting a GPS fix more difficult. In bothenvironments, we had black spots,where multipath reflections led to par-ticularly high errors and, thus, largejumps in the reported position.

Our second major uncertainty arosefrom the use of 802.11b networking. Wehad only partial coverage even thoughwe invested considerable effort in de-

36 PERVASIVEcomputing http://computer.org/pervasive

D E A L I N G W I T H U N C E R T A I N T Y

Figure 2. A runner with game equipmentbuilt into his jacket.

ploying 802.11b in both game zones.(We deployed an 8 m mast on a rooftopsupplemented by two omni antennas inSheffield. In Rotterdam, we deployed anetwork of seven wireless access points:four on buildings and one each on alamppost, van, and ship.) Consequently,runners would move in and out of con-nectivity, frequently leaving and rejoin-ing the game. For Rotterdam, system loganalysis revealed three broad categoriesof packet loss intervals:

• Periods of short loss (less than 5 sec-onds) that account for 90.6 percent ofloss intervals and were probably dueto communication errors

• Two-hundred seventy-eight moderateperiods of loss (between five secondsand 10 minutes) that were probablydue to detours out of connectivity orinterference

• Two loss periods of about 15 minutesand one of about 40, probably result-ing from major equipment failures

The runners’ separate walkie-talkiechannel provided broader coverageacross the game zone than the 802.11bnetwork, although it sometimes had in-terference from other walkie-talkie andradio users.

A third source of uncertainty arosefrom frequent technical failures such ascables working loose and damaged con-nectors (our runners were really run-

ning; consequently, their equipment suf-fered a battering) and soft failures suchas batteries running out of charge. Theseproblems added to GPS and connectionproblems.

Delay was our fourth source ofuncertainty, arising from Internet and802.11b network delays and gameserver processing delays. Although vari-able, we saw a typical delay of six sec-onds or more between one participantacting and another participant seeingthat action.

Experiencing uncertainty inCYSMN

In Sheffield, CYSMN ran for six hoursover two days and received over 200online plays; in Rotterdam, it ran for20 hours over five days and receivedover 1,000 online plays. For our analy-sis, we drew on three data sources:ethnographic observations of players,runners, and technical crew—includingvideo data capture and transcription;system log analysis such as GPS, net-work traffic, and over 3,000 onlineplayer chat messages; and player feed-back email. We provide a summary ofthe highlights here.

The playersOverall, CYSMN was well received, and

the players seemed to feel genuine en-gagement and even tension, especiallywhen the game was working smoothly.Players would hear their names over theaudio channel and then be chased. Theyhad to tune into their environment,including awareness of the runners(hearing them as they negotiated trafficand breathed heavily while running) andother ambient sound (including a mobiledisco at one point). In Rotterdam, someplayers even saw runners pass by a win-dow as they came to get them. As oneplayer put it in an email:

The start of me becoming totallyengaged was when I met up with mypartner who was playing in the sameroom, and through fits and starts wefound each other and then ran“hand in hand” in desperate flightacross the city. I then had this realfeeling of the need to protect herfrom being caught, and we couldwork cooperatively in keeping an eyeout for incoming runners.

However, the game did not always runsmoothly, and the uncertainties some-times had an apparent effect. Playersnoticed—especially when they were

JULY–SEPTEMBER 2003 PERVASIVEcomputing 37

Figure 3. The runner’s interface: (a) closeup and (b) map view. Blue arrows showrunners, red arrows online players, andthe bottom of the screen shows the mostrecent player text messages. Theinformation at the top in green showsthe current estimated GPS error as provided by (left) the GPS receiver, (middle) the network connectionstrength, and (right) the number ofonline players in the game.

(a) (b)

ences. We’ve seen evidence for our suc-cess in positive responses from the pub-lic, press, and commissioning bodies;further European bookings for CYSMN;and receipt of the 2003 Prix Ars Elec-tronica Golden Nica award for interac-tive art. As a research vehicle, we’vehoped to test our technology in the mostrealistic and stressful situations we can,moving from the lab to the field and giv-ing access to numerous users. This buildson the approach of staging public per-formances as a research method thatwe’ve developed since 1996.4

The concept for CYSMN is a chasegame. Three runners, who are profes-sional performers, run through actualcity streets equipped with handheld com-puters, wireless network connections(using 802.11b), and Global PositioningSystem receivers. They chase up to 15online players through a virtual modelof a city.

The online players’ experienceOnline players can move through the

model with a fixed maximum speed,access a city map view, see other players’and runners’ positions, and exchangetext messages with them.

Online players start at the CYSMN Webpage, where they can explore back-ground information about the game,including instructions on how to play.They enter a name for themselves andthen join the queue to play (we restrictedthe number of simultaneous players).When it’s their turn, they’re dropped intoa highly abstract 3D model of the host-ing city. The model shows the streets’layout and outline models of key build-ings but doesn’t feature textures ordetails of dynamic objects such as carsor, of course, most of the population.Additionally, the Rotterdam map in-cluded wire-frame representations oftwo buildings still under construction.Online players use the arrow keys to run

around this model. They cannot entersolid buildings, move out of the gamezone, or cross several fences. They mustavoid the runners. If a runner gets withinfive virtual meters of an online player,

the player is seen and out of the game.Scoring is based on the time elapsed sincejoining the game.

Online players see themselves repre-sented as running avatars, similar to

JULY–SEPTEMBER 2003 PERVASIVEcomputing 35

Useful URLs! Blast Theory: www.blasttheory.co.uk! CYSMN Rotterdam Archive: www.canyouseemenow.v2.nl! CYSMN Sheffield Archive: www.canyouseemenow.co.uk! Its Alive!: www.itsalive.com! Mixed Reality Laboratory: www.mrl.nott.ac.uk! Uncle Roy All Around You: www.uncleroyallaroundyou.co.uk! Unwiredfactory: www.unwiredfactory.com

Figure 1. Online player’s interface: (a)close up view and (b) map view.

(b)

(a)

2009年10月28日水曜日

Page 4: Nottingam Gms

GPSの精度↓悪?

技術的な問題?2009年10月28日水曜日

Page 5: Nottingam Gms

そんなことはない• ランナーの会話などを追っていったら、トランシーバーでどこの位置の電波が悪くて、どこに移動して解決したなどの情報を共有していた

• それどころか、一般人を電波のいい場所におびき寄せ、つかまえる戦略も用いていた→GPSの精度は、ゲームを構成する要素になっていて、みんなちゃんと理解して遊んでいた

2009年10月28日水曜日

Page 6: Nottingam Gms

Uncle Roy All Around You

• Uncle Royという、架空の人物の居場所を探すゲーム

• 現実感を破壊する様々な試み

• 実世界のプレイヤーと、オンラインのプレイヤーが共同で探す

• 実世界のプレイヤーは、地図を見て位置を自己申告する→その場所に応じたあいまいな手がかりが届く

In this paper, we describe a third performance called Uncle Roy All Around You. This also takes the form of a game that mixes online players with street players. However, this time the members of the public are on the streets using mobile devices as well as online, with actors only appearing at key moments. Furthermore, the game has a more complex and nuanced structure which replaces the frenetic chase with a mysterious journey through the city.

THE DESIGN OF UNCLE ROY ALL AROUND YOU Uncle Roy All Around You mixes street players who journey through a city in search of an elusive character called Uncle Roy, with online players who journey through a parallel 3D model of the same city, are able to follow the progress of street players, can communicate with them and can choose to help or hinder them. The core artistic theme of the work is trust in strangers – be they remote players, Uncle Roy or passersby. An mpeg video of Uncle Roy is available at www.crg.cs.nott.ac.uk/~sdb/videos

A street player’s experience Street players purchase a ticket for an experience that will last for a maximum of one hour. On arrival at the venue they hand over all of their personal possessions including bags, wallets, mobile phones and keys, in exchange for a handheld computer, a ritual that is intended to increase their sense of anticipation, vulnerability, dependence on Uncle Roy and isolation and disconnection from the everyday experience of the city. An actor briefs them that their mission is to rendezvous with Uncle Roy and explains how to use the handheld computer. They then head out into the city, cross a busy road and enter a nearby park. Their first task is to find a red marker on the PDA map, to get to the physical location that this indicates, and then declare their position to Uncle Roy. Street players declare their position by using the stylus to drag the ‘me’ icon on their PDA map to their current location and then pressing the ‘I am here’ button (figure 1).

Figure 1: street player’s map, zoomed out and in Whenever they do this, they receive a short text message back from Uncle Roy that provides them with a clue as to where to go next. In this way the street players undertake a journey through the city, following a trail of clues that lead

them through the park and into the narrow city streets in search of their eventual goal – Uncle Roy’s office.

Figure 2: following clues in search of Uncle Roy’s office The clues are preprogrammed (Uncle Roy is therefore the automated voice of the game, not a live actor) and are attached to different zones of the game map. A street player gets an initial clue the first time that they declare themselves to be in a region and a second and different clue on subsequent declarations. A key feature of the game is that Uncle Roy’s clues are deliberately designed to be ambiguous – some are relatively direct and useful, while others are misleading to the point of being mischievous, encouraging players to follow diversions, drawing on the history of the local environment, implicating passersby in the game, heightening the sense of being watched and also casting doubt on the intent and personality of Uncle Roy, especially the extent to which he can be trusted. Clues also constantly remind players that they are on limited time and that the clock is ticking. Examples of clues include: Good. I want you to walk towards the Mall. Watch a tourist cross the road and follow them. There are some hidden steps among the buildings. You have NN minutes remaining.

You are doing well. Drift towards Buckingham Palace. After a couple of minutes click I am here. You have NN minutes remaining.

It’s not a shock. The borders are policed. It’s always been so. Look for a road leading of the square – one you have not been down before. Walk down it. You have NN minutes remaining.

And when they finally reach the zone containing Uncle Roy’s office: Go to number 12 Waterloo Place and ring the buzzer marked Roy.

The first performance was staged in a area of central London of roughly 1600 meters East-West by 1200 meters North-South, centred on the Institute of Contemporary Arts (the hosting venue), with St James’ Park in the south and the narrow streets of Westminster in the North (figure 3). There were 49 distinct regions in the game, ranging in size from roughly 150 by 150 meters in the open park area down to roughly 10 by 10 meters in the narrow city streets. The clues were defined by colouring the game map (each colour being mapped onto a distinct pair of clues), allowing

In this paper, we describe a third performance called Uncle Roy All Around You. This also takes the form of a game that mixes online players with street players. However, this time the members of the public are on the streets using mobile devices as well as online, with actors only appearing at key moments. Furthermore, the game has a more complex and nuanced structure which replaces the frenetic chase with a mysterious journey through the city.

THE DESIGN OF UNCLE ROY ALL AROUND YOU Uncle Roy All Around You mixes street players who journey through a city in search of an elusive character called Uncle Roy, with online players who journey through a parallel 3D model of the same city, are able to follow the progress of street players, can communicate with them and can choose to help or hinder them. The core artistic theme of the work is trust in strangers – be they remote players, Uncle Roy or passersby. An mpeg video of Uncle Roy is available at www.crg.cs.nott.ac.uk/~sdb/videos

A street player’s experience Street players purchase a ticket for an experience that will last for a maximum of one hour. On arrival at the venue they hand over all of their personal possessions including bags, wallets, mobile phones and keys, in exchange for a handheld computer, a ritual that is intended to increase their sense of anticipation, vulnerability, dependence on Uncle Roy and isolation and disconnection from the everyday experience of the city. An actor briefs them that their mission is to rendezvous with Uncle Roy and explains how to use the handheld computer. They then head out into the city, cross a busy road and enter a nearby park. Their first task is to find a red marker on the PDA map, to get to the physical location that this indicates, and then declare their position to Uncle Roy. Street players declare their position by using the stylus to drag the ‘me’ icon on their PDA map to their current location and then pressing the ‘I am here’ button (figure 1).

Figure 1: street player’s map, zoomed out and in Whenever they do this, they receive a short text message back from Uncle Roy that provides them with a clue as to where to go next. In this way the street players undertake a journey through the city, following a trail of clues that lead

them through the park and into the narrow city streets in search of their eventual goal – Uncle Roy’s office.

Figure 2: following clues in search of Uncle Roy’s office The clues are preprogrammed (Uncle Roy is therefore the automated voice of the game, not a live actor) and are attached to different zones of the game map. A street player gets an initial clue the first time that they declare themselves to be in a region and a second and different clue on subsequent declarations. A key feature of the game is that Uncle Roy’s clues are deliberately designed to be ambiguous – some are relatively direct and useful, while others are misleading to the point of being mischievous, encouraging players to follow diversions, drawing on the history of the local environment, implicating passersby in the game, heightening the sense of being watched and also casting doubt on the intent and personality of Uncle Roy, especially the extent to which he can be trusted. Clues also constantly remind players that they are on limited time and that the clock is ticking. Examples of clues include: Good. I want you to walk towards the Mall. Watch a tourist cross the road and follow them. There are some hidden steps among the buildings. You have NN minutes remaining.

You are doing well. Drift towards Buckingham Palace. After a couple of minutes click I am here. You have NN minutes remaining.

It’s not a shock. The borders are policed. It’s always been so. Look for a road leading of the square – one you have not been down before. Walk down it. You have NN minutes remaining.

And when they finally reach the zone containing Uncle Roy’s office: Go to number 12 Waterloo Place and ring the buzzer marked Roy.

The first performance was staged in a area of central London of roughly 1600 meters East-West by 1200 meters North-South, centred on the Institute of Contemporary Arts (the hosting venue), with St James’ Park in the south and the narrow streets of Westminster in the North (figure 3). There were 49 distinct regions in the game, ranging in size from roughly 150 by 150 meters in the open park area down to roughly 10 by 10 meters in the narrow city streets. The clues were defined by colouring the game map (each colour being mapped onto a distinct pair of clues), allowing

An online player’s experience An online player, connected to the game over the Internet, journeys through a parallel 3D model of the game space. They move their avatar through this model using the arrow keys on their keyboard, encounter other online players and send them public text messages. Online players also access a set of cards that provide details of the current street players in the game, including their name, gender, a brief description and a photograph that was taken when they registered to play. They can choose to send private text messages to individual street players or listen to their most recently uploaded audio message.

Figure 7: online player’s interface: own avatar (white figure), street player cards (right), street player’s position (red sphere) and text message boxes (bottom). Online players find photo objects as they explore the model. Entering these triggers the display of photographs from the actual city streets at this location, one of which is labeled as Uncle Roy’s office and shows an image of his distinctive office door (figure 8). In this way, online players can find out useful information for street players.

Figure 8: Accessing a photo of the office door Street players’ estimated positions as they move through the city are shown as pulsing red spheres in the 3D model (see below for a discussion of our approach to positioning). Whenever a street player explicitly declares their position to Uncle Roy in order to receive a new clue, this representation is significantly enhanced (figure 9) – red

lines radiate from their position, a large red sphere becomes visible and gradually shrinks down to this position like a deflating balloon, a dramatic sound is played, and the online player also sees the clue from Uncle Roy.

Figure 9: watching a street player declare their position Online players can zoom out to a bird’s eye view that shows the positions of all online and street players and key landmarks marked and labeled (figure 10).

Figure 10: bird’s eye view (one street player is declaring) Finally, whenever a street player enters Uncle Roy’s office, online players are invited to join them. This involves seeing a live webcam view looking into the office which enables them to see the street player in person for the first time (figure 11).

Figure 11: seeing the street player on the office webcam

2009年10月28日水曜日

Page 7: Nottingam Gms

ジオジオスタンプラリー• GPSを用いた宝探しゲーム

• GMS、オープンストリートマップジャパン、ハッカーズカフェ主催

• 5人程度のチームを組んで、iPhone

の「DGRadar」の情報を頼りに宝探し

• (1)立教大学池袋キャンパスにて小規模な宝探し(2)都電沿線のスタンプラリー

• 分析用に撮影しました

2009年10月28日水曜日

Page 8: Nottingam Gms

実際のビデオ(一部)

2009年10月28日水曜日

Page 9: Nottingam Gms

観察• 主にCan You See Me Now?と比較

• GPSの点が何を表すか

• CYSMN:ランナーの位置

• ジオジオ:自分の位置を基にした、アヒルの距離

• どのように情報を伝達するか

• CYSMN:基本的にトランシーバーとテキストメッセージ

• ジオジオ:チームで、目に見える範囲で会話、身振りなどを組み合わせる→指差しなどを詳細に分析すると面白い(未着手)

2009年10月28日水曜日

Page 10: Nottingam Gms

観察• GPSの精度が悪いとき「飛び跳ねてる」と言うのは万国共通

• 一方で、位置の提示の仕方には大きな違い(具体的な場所か、相対的な向きか)

• Can You See Me Now:Runner 2 on walkie-talkie: Runner 2. I’m in pursuit of Dave. He runs along a side-street, consulting the Jornada as he goes, turning left at the end of the street and going down Wilamena before slowing to a walk. Runner 2 on walkie-talkie: Runner 2. I’m heading seawards on Wilamena, waiting for a server update.

• ジオジオスタンプラリー:

06 B:これ(  )まがって、方角的にはどっちなんですか07 A:えっと方角的には::(.)え::と(2.9)

08 A:イースト(2.6)

09 B:イーストって::っと10 A:はい11 B:こうなって

• 専門的な知識は使わず、iPhoneを2台並べるなど特有の解決

2009年10月28日水曜日

Page 11: Nottingam Gms

いろいろ考えてみた• 全部精度300mにして、その中に入ったら音声チャンネルが開く

• 一人で見る用じゃなくて、みんなでわいわいやれるようなインタフェース

• 最近位置情報ゲームが増えてきてるし、その辺も情報収集しないと

2009年10月28日水曜日

Page 12: Nottingam Gms

「ジオジオスタンプラリー2」開催決定!

• 詳細は未定

• アイデア、技術、ネタを求めています

• 最先端の遊び場を作りましょう

2009年10月28日水曜日