House of Mystery
*
A MOO-lesson in the intricacies of the mystery
Final Project
"Virtual Environments", Spring 2000
Supervisor: Susan Warshauer
Lisbeth Klastrup
House of Mystery - a MOO-lesson in the intricacies of the mystery
The House of Mystery that I refer to in this essay, is located
in the StoryMOO (http://dkm.itu.dk:7000)
at the IT University in Copenhagen, Denmark. To go to StoryMOO,
click on the link or type address in your browser, log on as
guest and follow the signs to the House of Mystery.
You may also enter StoryMOO from AppalachiaMOO - go to AppalachiaMOO (http://ariel.clc.wvu.edu:7000)
and from the Welcome Center, go to Plaza, then continue to
Mountaineering Hall and enter the Nordic Connection.
Introduction: intention of the project
Part of my ph.d.-project at the IT University of Copenhagen is to set up and design a MOO in which students, following our university's Masters programmes, have the possibilites of exploring different kinds of narrative structures. By making it possible for the students to "be" in a narrative, rather than just reading it, I hope to make them aware of the possibilites and pitfalls of narrative design, especially in a digital environment. The students in question is primarily thought of as student with little or no knowledge of dramaturgic models, for instance software engineers or computer scientists.
When choosing the subject of my final project of this class, I saw it as a great opportunity to attempt to implement some of the ideas, I have been juggling with before and throughout this class in connection with my overall ph.d. project. That is: how can you design interactive experiences that on the one hand provides the participants with a high degree of freedom and takes advantage of both the on-line presence of real characters and the MOO-environments ability to create a feeling of scenery and geographic space; and on the other hand makes possible the emergence of a narrative, that is both complex, coherent and compelling?
I decided to go ahead and try to construct a version of an interactive narrative performance to be tried out by a number of invited participants in order to see how my ideas turned out to work in practice. In principle, I wanted the narrative in question to be fairly open: that is, I wanted to provide the on-line participants with some story information, but no more than they could make stories of their own too, and interpret each others behaviour differently, depending on the framework of contextual interpretation, they were given.
Furthermore, I wanted to combine this interactive experience with an element of learning. Having played around with the narrative structure, I had constructed; I wanted the participants to be able to find information on how to learn more about this kind of narrative and about how to design this kind of narrative structure themselves.
Partly inspired by my own experiences as a online player on the "Modus Operandi" Roleplaying Game, I decided to work with the mystery as my basic narrative structure. As a genre, the classic mystery is often very formulaic and hence not that difficult to imitate structurally. However, designing a convincing and casually logic mystery plot is not that easy, which I quickly discovered once I set to the task. But, realising that time was short, I decided to stick to my plan of an online "rehearsal" even if all the details were not in place. So eventually this project ended up falling into 3 parts:
Theoretical framework
In Hamlet on the Holodeck, Janet Murray defines the four essential properties of digital environments as being the property of procedurality, participation, spatiality and encyclopedic scope. One could most likely question exactly this branding, but this is not my intention here. I find that they function well as working concepts or as "handles" by which one can try to design an experience, specifically intended for the computer medium. I think of the procedurality as the possibility (or limitations) that lies in the fact that you have to define strict procedures and logic rules for the unfolding of a narrative event if it is to be understood and moderated solely by the computer itself. However, part of this "procedurality" might be the objective registration of the user's movements and current status in the system according to the parameters defined by the given system within which the participants interact. It is my firm conviction that one should take advantage of the property and not dismiss this user-supervision as being too "Big Brother"-like or intruding on the "freedom" of the particpant or system user. I agree with more conservative cybertheorist like Janet Murray and Sven Birkerts in that much of the thrill of reading or experiencing stories arises from the feeling of immersion in "the other world", the text of these make possible - and from the thrill of seduction that lies in not being able to predict to where or what the story will take you after the next "pageturn". - On the other hand, I do not find the nostalgic longing for a Shakespeare of the digital medium very useful when trying to envision how and what a digitial narrative experience should be like or "result" in. On parallel lines to Espen Aarseth, I believe that one should not judge or try to understand digital storytelling solely from the standpoint of a literary theory which primarily is intended to describe what goes on in print fiction. Rather, one must seriously confront the theoretical implications of the procedural and participatory properties as well as one should try to use them creatively to construct experiences that could become truly cybertextual and yet emotionally rewarding from an user's perspective. Hence, the concept of textons and scriptons, that Espen Aarseth coins might be useful when thinking about and describing digital texts that arise from a combination of the static and given units "already there", the scriptons, and the dynamic and ever changing "textons" that arises from the users use of these scriptons. Returning to this specific project, the scriptons would be the stable elements of the MOO; such as the preset MOO rooms programmed and locked by a wizard, generic objects and possibilites of interaction; and the textons the "text" that one reads on the MOO-screen during a specific MOO session and which comes into being the moment, people communicate with each other in MOOspace, moves in and out of rooms and interact with the objects in these.
Concludingly, with an Aarsethian wording, ideally what I would like to design is an on-line experience for which some scriptons are provided, but that also makes possible an infinity of textons; this experience being simultaneous an experience that is both interesting and immersive and provides the users with a sense of direction, of curiosity and anticipation of unexpectedness.
Assessment of the project:
For this project, I have made a part of my "own" StoryMOO into a House of Mystery. The House of Mystery is intended to be a virtual version of a murder mystery dinner, at which the particpants are supposed to solve a murder that has taken place prior to the commencement of the game. The participants are all provided with scripts that tell them who they are and what their relation to the other characters are. Furthermore I have supplied the characters with a motivation and some code of behaviour (in order to prevent them from lying all the time, for instance)
Following the murder dinner concept, my House of Mystery consists of a dressingroom that contains the characters, people should take on, before they enter the interior of the house. Also (now) this room contains some general information about the background and setting of the murderscene and the rules of the game ("General information for players"). On the day, the play was actually performed, I was online in the room with the players, which had previously been provided with information about their characters via e-mail and I supplied their individual information with a brief introduction to the scene, they were about to enter by "saying" it to them.
The remainder of the House consists of a number of rooms, one could realisticly find in most (mystery novel) houses. The characters can walk in or out of these rooms, look at the room descriptions, notes and object clues to be found in there - also in two rooms (bathroom and kitchen), they can find a bot which they can ask questions related to the murder. One room, "The Locked Room" should ideally be locked until someone carrying the brasskey to be find in one of the other rooms enter it. However, due to a bug in the MOO-code, this lock-function doesn't work as intended, so presently the room is open. During the performance, I had locked the room "with me" (which was possible), meaning that only I could open it. At some point during the performance, I sneaked upstairs and opened the room, and announced the opening of the room by "shouting" it to all characters (the shout-command will show the shout-text in all rooms).
In addition to the fictional setting of the House of Mystery, there are a few "learning" rooms in which the characters can find information about mysteries in general and on how to construct them. They enter these rooms through the OOC-room, a room intended to be a kind of "time-out" place, where players can discuss the mystery out-of-character (and/or ask for my help should any problems arise). Unfortunately, rhese room had no contents on the day of the performance, so I do not know how people will react to them yet, but if and when another performance hopefully takes places, I will be looking forward to seeing if they go there voluntarily after having completed the murder mystery playing. As is, I hope that both the House of Mystery fiction rooms and the fact rooms will endow even a singular visitor with some sense of being in a world of mystery and murder and with a desire to learn more about the workings of it.
The House of Murder performance, Fridag the 29th of April.
On Friday the 29th of April, I had invited 5 people to meet on-line to play a "murder game" from 3 pm - 4pm (it was during working hours, so I did not want to take up too much of their time). Prior to the event, I had instructed them via e-mail with respect to their characters and the basic concepts of MOOing. Also I had asked them not to come online before. The player consisted of an interesting mix of "computernerds" (2 engineers and 1 computer scientist with little or no experience of MOOing but much at ease with medium) and 2 experienced, "literate" MOOers. 4 of them were Danes and 1 Norwegian, so I allowed them to play/interact in Danish and Norwegian in order to make communication more natural and fluently. I recorded the event with a recorder I took from one room to the other (due to lack of time, I did not manage to place recorders in all rooms, but worked well as it where, since most interaction took place in only one room at a time). With me almost silently supervising the event, the players played and stayed in character for approx. 70 mins. and all took their roleplaying very seriously, which was a positive surprise, but also, I surmise, also happening, because I had provided them all with very detailed script of their characters.
Here is some of the things, that I noticed and which might be of more general interest:
Interaction between characters:
The mystery
They tried as hard as they could and found it fun, but alas! my players were not able to spot the murderer. I had provided them with too few clues; and furthermore they did not extract the information intended to be extracted out of bots "Mother" and "Mrs. Sweetpeas". This were mainly my fault: I should have provided them with more physical clues and general information, and I should have programmed the bots better (now they only react to single words, not sentences).
However, using the clues provided, the characters did invent stories of their own and also (unexpectedly) they followed some false clues: for instance, when entering the crime scene, the livng room, one character tried to resussitate the dead man (which I hadn't foreseen, but it was a great performance..). Also the scrap of paper found on the corpse, had a phonenumber on it (my excuse for planting information about the wills) and people payed much more attention to this phonenumber than expected: they were discussing trying to reach it from a mobilephone, asking if anyone had a mobile phone on them (sic!) - and they persistently kept returning to the mystic phonenumber throughout the performance. The play concluded very nicely: murderer Ernest came forward and revealed himself as the murderer on my request and I virtually called for the police and had him taken away with the police. While "leaving", he shouted, "I did it for Sudan!" ( a mispelling of Susan, the woman he was involved with), but all had great fun out of this: imagining an altruistic murder, who killed to get money for Sudan!!
Conceptual problems
One major problem of this event was that too much time were spent running in and out of rooms and too little time were spent talking together, interacting with each others as characters and trying to get information out of each other. Though they were motivated for interactive roleplaying, especially the murderer, whom I had scripted to try and "win" the game by escaping conviction, the frantic running around the house, made room for only a little roleplaying and therefore the characters did not find out as much about each others motivations and agendas as I had hoped.
Hence discovering these problems with getting on with solving mystery, I invented new clue on the fly (a note in Locked Room connecting Ernest with Susan) and after making sure that people had read it, I asked people to gather and talk in the living room with shout-command. This worked, though time was running out by then and they were stuck with the few clues, they had. What I would like to point out with this detailed description of my own workings is that much more moderation by a RL person was needed than expected - and my conclusion regarding this is, that one would have to prevent need of moderation by building more restraints into system, by providing the players with more clues and script them more carefully. I have tried to implement some of these suggestions, by placing a general script for the performance procedure in the dressing room ("General information") and by leaving the note from Susan to Ernest behind. I intend to plant more clues in the rooms and in the bots in due time, too.
Another problem related to the clash between navigational exploriation and in-room interaction in this case was, that I had based some of the exposition of the solution on the other characters spotting that Ernest went to bathroom (to sniff cocaine which he was to told do to "in verbs" when in the bathroom); also I had asked Nigel and Susan to try to sneak into office and take a look at the wills. However coming and going in the MOO-system ssimply happens too quickly and noone really payed attention to each others whereabouts (it would only make sense for people to sneak away, if they were explicitly asked to stay in one room and explicitly told to keep an eye on each other; I have tried to do it partly on the "General Information" note).
Concluding remarks:
I have had great fun designing this murder mystery and based on the fun, people seemed to have, when they were performing it, I do feel that I have succeeden in creating a mystery ambience in the House of Mystery. Based on the problems, I encountered, I know think, it would make more sense, if the mystery game was divided into two parts:
part I : where people explores the house and discover clues (not allowed to remove them though, technically some kind of binding the object to the room should be possible). Stealing objects is not allowed until part II, during which people are asked to stay in same room and try to solve murder by talking with each other. They must not leave room unless they have a reason (like asking further questions of bots or the like) and so those trying to sneak out would be spotted. Also I depended on people noticing Ernest's drug addiction by his behaviour and look (and re: the "look" part is was obvious that people was left in the blind, since people's appearence is not explicitly described. However, one could set the description of the various characters in advance and let these descriptions give clues to Ernest's drug, John's trade (he could wear a ugly t-shirt with his band-name) and so forth.
I do feel, that both theorists (the wizards constructing these experiences) and the practicians (the students performing them) can benefit from actually trying out various narrative genres and experiences. Albeit that these experiences only have the character of a never-to-be-repeated performance, judging from my own experience, I do think that one can learn a lot from just one performance, both about interaction and roleplaying and of how people interact with the virtual environment as such. I will definitely keep working on my "House of Mystery", looking forward to the day, when people start going there all by themselves!
Works cited (or used as inspiration for the design of the MOO)
Aarseth, Espen: Cybertext: Perspectives on Ergodic Literature. Johns Hopkins University Press, 1997.
Birkerts, Sven: The Gutenberg Elegies - The Fate of Reading in an Electronic Age, London: Faber and Faber, 1994 (paperback edition 1996)
Cawelti, John. G: Adventure, Mystery, Romance. Chicago, University of Chicago Press, 1976.
Malloy, Judy: "Public Literatur: Narratives and narrative structures in LAMBDAMOO"
(http://www.well.com/user/jmalloy/moopap.html, last accessed May 1th, 2000. (in press for an MIT Press book on the Xerox PARC PAIR Program)
Murray, Janet H.: Hamlet on the Holodeck : The Future of Narrative in Cyberspace. New York: Free Press, 1997
and thanks to Jesper, Jill, Anders, Lasse and Jakob for making the House of Mystery come alive!