Urban Environments and Interactive Technologies
Four Versions of the Digital City
Friday, Sept. 25, 1998
Speakers: Theresa Duncan, Shigeru Miyagawa, Glorianna Davenport,
Kurt Fendt and Ellen Crocker
[These are edited summaries, not complete transcripts.]
Theresa Duncan: Zero Zero is a CD-ROM adventure distributed by Rhinestone Publishing which I created with Jeremy Blake. The story is set in Paris on the eve of the 20th century in 1899. There is a brief linear opening that sets up the story about the main character, a little girl named Pinkée, who goes through the city asking what the future will be like.
The main interface is the rooftops of Paris, which Pinkée climbs over to travel through the city. This contrasts with many games for little girls that are often about small enclosed places and don't have girl characters that move through space. I wanted to show a girl who is adventurous and curious, so the interface to the city is about discovery and surprise. Within each shop and place, there is a story and a world.
Pinkée can go dozens of places where she meets characters with their own perspectives on her questions about the future. For example, she meets a woman in the greenhouse who loves plants and trees, so she says that they are going to enclose Paris in a great glass dome, so it will be spring and flowers will grow all year round.
I like stories that are more complicated than the current kids' books that don't have a lot of plot, so there are a lot of places where characters are referred to or briefly introduced that are later introduced in more depth. I wanted it to be a rich literary experience, so I tried to use words and ideas that children might not otherwise experience. So for example, Pinkée can go to the Bakery, where she meets Phillipe, the baker. He loves philosophy and creates cakes shaped like the heads of famous philosophers which produce philosophical ruminations when clicked.
Shigeru Miyagawa: StarNet is conceived as a web based hub for international education. It is based on theStar Festival CD-Rom which was completed here at MIT a few years ago. Star Festival represents a journey through both time and place. It is really the story of my own return to my native city of Hiratsuka in Japan, which took place during the week leading up to the Star Festival, an important local event. I left Japan when I was 10 and lived in Alabama, where there were no Asians except for my family. The only Asian role model I had when I was growing up was George Takei, who played Sulu on Star Trek. I think it is ironic that he plays my character in Star Festival.
A time capsule metaphor is one good way to think of how Star Festival digitally represents a small town in Japan. The interface is a fictional PDA lying in the street, which contains a lost record of the professor's trip, which is really my trip. Like any good mystery, the story begins at the end. The professor has disappeared, but no narrator holds our hand to explain what is happening. The premise is that on his way to his old home, the professor collected and stored whatever he found. Thus, the city of Hiratsuka can be explored through these stored artifacts, like bread crumbs dropped by Hansel in the forest.
Space is represented in a variety of ways. Places are captured through video journeys. There are 3-D virtual representations of spaces in which students can walk through and see video representations of the spot. There are point and click graphical representations of spaces, such as the professor's home. In addition, there is also a map representing the stops the professor visits on his way home through his old neighborhood.
The city is represented in time as well as space. As we follow the professor, he begins to realize that he could not know himself without knowing his past and his heritage. Similarly, the contemporary Japan that is represented cannot escape the past when looking at itself. This is captured through a collection of small videos of people that the professor remembers from childhood. In a striking example, the professor interviews his mother as she stands on a spot where her high school once stood and recalls the bombing of the city.
People are a critical element in representing an urban space, and must be represented as much as the city that they built. The CD contains roughly 300 original photographs that were kept by my family and myself. It is interesting that kids ultimately come back to these. Why are these pictures of people important? It is because we learn through empathy. It is not enough to create a travelogue by simply recording the town. Learners really come to understand the city through the eyes of professor, who they befriend. They care about what happens to him and relate to his journey across cultures, so they begin to care about the place and culture he came from.
StarNet is being turned from a CD-Rom into an internet television series available over the web on a weekly basis. Originally the CD was designed for college and high school students. This past year we conducted a series of beta tests at Boston Public Schools. We are quite surprised that even kindergartners took to the show. One of the teachers commented that StarNet is like putting your whole family in a car and going for a drive in Japan, and every passenger notices something different depending on their age and interest. It is particularly critical that the students themselves control their experience, because putting them in the drivers seat fosters a sense of independence and exploration.
The ultimate goal of StarNet is to send students back to where they come from, to get them to look into their own life, and to get them to think about exploring and expressing their own life and heritage. In order to do that, they have to get off the net, shut down the computer, go outside and look around. Thus, the ultimate goal is to extend learning beyond cyberspace into the home and neighborhood.
Glorianna Davenport: I'm going to give you a historical tour of about 15 years of projects that have used the city, or some representation of the physical city as part of a experiment in interactive experience. Some of the projects are about learning and some of them are about travel. I will emphasize how the projects each relate to the city in these ways:
- The city space as physical paths, nodes, boundaries and landmarks.
- The city as place representation, such as a map combined with time, where you can understand change, and combined with people.
- The city as story, which is this continuity of interaction and change over time.
- The city as memory and metaphor, and it is what we are finally seeing in some of the representations of cyberspace.
One reason for taking this historical tour is to understand what we were able to do as we moved away from an analog video image into a digital world where we could have a computational system respond to an explorer in cyberspace. I hope that seeing them all together in sequence will be interesting and challenging for thinking about the future of the digital city.
In the "Aspen Project," developed at MIT between 1978-1980, we focused on exploring a city through surrogate travel. We shot scenes every ten feet along roads in Aspen, because we wanted to know if you could come to know the city by navigating through a set of frames. We were also trying to figure out how to build an evolving city, based on representations of the city changing over time. Of course, we liked people, so we put some people in it. For example, you could go into the police station and listen to a policeman. We were on the edge of thinking how people actually change cities, but given the technology of the time, we didn't put in many people.
In the "New Orleans in Transition Project," we shot before, during and after the World's Fair with a particularly focus on development along the river front. The idea was to get many different points of view, and so there were 50 characters represented. We built an interface that tried to automatically display the next most likely things that users might want to look at, such as a map, a history of the river front, some people, or some biographies. This was also used as an experiment in constructivist learning. We connected videodisks to a networked environment, and we built an editing system so students could edit pieces of video together and create their own vision of the material.
During the "Elastic Charles Project," I looked at how to get a whole class of twenty people to make a story of a city from many perspectives. This project focused on the idea that people are very powerful in conveying the feeling of a city, and they are part of the city experience. In a hypermedia journal about the Charles River, students could create their own links and make their own story connections. We were playing with a construcivist idea, but it was not fully formed because students couldn't add new video or text material. This was also starting to play with some of the things that you would like a story teller system to do.
We continued to work on the problem of describing material to the computer in a way that is not so artificial and intensive, so that anyone can submit pieces. For about two years, we filmed the Boston Artery Project, but it turned out to be just too hard to do with just my students. We are now launching a project in which we think we can build a system where people can add elements on a daily basis, and have the interface recompute itself to know what elements are in the database. This should allow somebody to submit something over the network, and have their contribution added to the network automatically.
In the newest project, Flavia Sparacino who is doing her Ph.D. work in thinking of the city as a memory palace. She is building something called the city of news, which is currently a one person 3D browser. Given a map, you can store information that you are interested in certain places. This January she will be putting it in a CAVE environment to explore the possibility of multiple people navigating through memory space, where they stash memories and ideas.
A lot of us in this business have had a lot of ideas that are about people and a dream about learning, about seeing, about remembering, about trying to understand how the world works. We shift our dream relative to the technological potential for supporting the communication process. My vision of the future of digital cities is not about one or another environment, but about the fact that we are going to have an enormous amount of variety of things to do, places to go, media to be involved with in a constructivist way through creation and dialogue.
I think we are in an evolutionary process of going to a different way of learning and a different way of communicating with each other. I think that part of that has to do with stories, which have always been about the sort of social learning that was needed at a particular point of time as humans evolved in their ability to cope in a complex society. We are now in a very complicated period of time with much more even distribution of responsibility and potential for learning, but there is much less control over learning. We need tools to help us make decisions about what we look at and how we put things together.
Kurt Fendt: Berliner Sehen allows students to gain insight into German language and culture. It portrays people's life stories in Berlin, and allows students to explore how those life stories are connected to the community, to the place, and the changing city. This interface allows students to follow the characters through different settings in their authentic lives to see how they are connected through the people they know, their day-to-day actions, and the history of the neighborhood.
The right side of the screen features eight key characters, which represent a range of people from east and west. There is also a list containing broad notions such as Myself, Others, Relationships, Public Sphere, Private Sphere or How People Refer to the Neighborhood. Based on which character and notion the student selects, the periphery of the viewing space fills up with pictures that represent video clips about the character relative to the selected notion. Then students can pull any of the pictures from the periphery into the center area to view the relevant short video clips. There are a total of 680 clips representing about thirteen hours of conversations. When a video clip is in the center, any other notions represented by the video clip also light up. If a student selects one of the other notions related to the clip, then the video in the middle remains the same, but the pictures representing clips around the periphery change. This allows students to shift to a whole new angle for exploring the material.
The students can currently access the material through the personal life stories, the notions, or through a combination of notions and people, but when people talk about their neighborhood, they refer to places. In a new phase of the project, we are creating a way for students to access to the material through space as well as stories. The students now can click on places referred to in the conversations. The students can also call up different maps over time and see how the history of the neighborhood also forms part of the people's life stories. We wanted to provide an actual walk through the neighborhood, so we are including Quicktime VR panoramas so students can follow along the streets and explore what it looks like today. They can also enter buildings, and see that a lot of the locations that people refer to as they tell their stories. The students can explore how Berlin is changing considerable, and how the two versions of the German culture have been clashing. We did the original filming in 1995, but we just went back and did some more photos and panoramas. The students can tell that it is not resolved at this point, and it is quite interesting to observe the process.
Ellen Crocker: In language teaching, we emphasize that students should revisit material quite often, because as they come back, they pick up on information that they may not have picked up before. The students first try to get a sense of what is out there. They try to find out what the people are up to and what are they talking about. When they listen to the material the first time, they do hear some things they understand, but there is also a lot they don't understand. Then, through their observations, they begin to build connections between stories, between people, and between locations. They begin to make connections on a very subtle level about who the characters know and where they live. We find that students develop relationships with the people that they are studying and get excited about what the characters are trying to do, so they begin working with this material in a very personal way. Since learning takes place when students manipulate the material, they can also create their own links to the video they have seen. This can be used for an assignment, or just to help them keep track of how they are exploring the neighborhood and these spaces.
Reponse to the Demos: Audience Discussion
David Thorburn: I think that one feature of a number of the videos was that there were so many ways to branch off and to go, did it also mean, in a related manner, that there was no direction? Was it just a branching experience and not a coherent one?
Glorianna Davenport: First of all whenever you have a presentation in front of you, of course it is very coherent because the speaker puts it together, he knows what's there and so on. But I think there was a lot of coherence. It was designed for that, it was designed in such a way that as you travel through space, you can try different things, but there is a look that holds it together, if nothing else. I would characterize them as sort of coherent databases, semi-coherent databases, and non-coherent databases.
Shigeru Miyagawa: Just briefly to add to Glorianna's response, it is true that there is a great deal of material and we need to organize it in some fashion. It is sort of a contradiction: on one hand you want a lot of material and you want the users to walk through as freely as possible, on the other hand, you need something to organize the material. So what we came up with, and with a limited degree of success, was that we put in a narrative backbone, flexible enough so that a user, depending on age and interest, could go through and recreate that narrative for him- or herself.
I'll tell you one true story that happened to us at BPS that really surprised us. I already mentioned how surprised we were that kids from kindergarten to high school really took to this. And what happened every time, was that a kindergartner when told to go into the show would look for me when I was in kindergarten and then begin to construct a story from that point. And a fifth grader would go in and look for me as a fifth grader and begin to construct a story from that point. It was really shocking to me that anyone would think of doing that. This is why it turned out that the kids were spending more time with the still photographs then anything else in the story.
Theresa Duncan: I think that restraint is important because a lot of people will throw in technology for technology's sake, that is, people will throw something in because it's a cool effect, or it looks good. The temptation always is to have more -- more special effects, more sounds, more video -- and that doesn't necessarily make a good story. People aren't as critical of new media as they are of cinema or traditional story-telling, but I think that more should be expected of it.
Glorianna Davenport: I have another thing to add. I am trying to go towards a semi-coherent database and I'm very interested in multiple users submitting material to a project. I understand the complications -- if you have one hundred people submitting their views of the city, how do you make that at all navigable?
I think that one of the big issues that you are touching upon, and about which there is a great deal of sensitivity right now, is control. And if you are an individual author, the understanding from the past has been that you control the story, you control the medium, you control everything about it. And if you have a computational medium, you put an abstraction into the system that allows the audience to have a dialogue in a sense with the system. And those two are rule contradictory. We're building a writing system now and traditional short story writers, whom I think would be perfect to write for this medium, all say, "Not us, not us. We want to control the story from beginning to end. We want the users to feel empathy in a particular way."
Discovery is antithetical to that control. Discovery is where the user has some control. That is the tension in interactive work today. If you are doing something for students and the students then put the material together and republish it, you've designed the system with all the control that a linear author has, but part of your system is that you don't have control over those pieces that get redesigned and put back in. It is a very interesting issue. We are looking at that in another course that I am doing this fall.
Kurt Fendt: In terms of "Berliner sehen," one problem for us was the complexity and the richness of the material -- going through all those life stories, you need to have a critical mass of material so that all the different points of view actually make sense --so that the students really get a rich immersion into those neighborhoods. But at the same time, another problem for us was that this was a project for German cultural studies and language learning, and we didn't really know what the background of the students would be.
We had to provide multiple ways of accessing, and going through the material so that they could follow some of their own hypotheses about the culture-- maybe even their own stereotypes about the German culture -- and then could also find material that contradicts their stereotypes. So we definitely didn't want to control the users' navigation but we wanted to provide a framework within which they could explore freely. So that is why there are 24 images around the periphery that the students can go through and change, but
there is always a framework they can start with.
Theresa Duncan: One of the things that makes multimedia attractive is that the users are guided by their own curiosity. And once they follow whatever they are curious about they are bound to get attracted to other areas. They are not passive, rather they become engaged by the subject matter, it is not just being fed to them.
Glorianna Davenport: The one thing I would just like to throw out is that some of these projects are made specifically for classroom use right now. The question is, is the idea of it extensible to a world where maybe you don't have a specific curriculum idea or a teacher, you have a world view, a city view, where somebody in that city who wants to explore something can do so. That's a question for the future.
Mike Shiffer: It was interesting to see how this morning's presentations loosely centered around how the city can shape how we think about information technology. It seems as if information technology has been a hard thing to grasp -- especially the Internet -- and people seem to find, as Bill pointed out, that the city can provide an effective metaphor for that. Although it appears that when people try to use a city-like metaphor for information technology, they struggle to do so.
But the thing that interested me about this afternoon's demos, and one of the things I wanted to ask each of you, is that these demos seem to portray the opposite, that is, how information technology can shape our perspectives of the city. I wonder if you can tell us what you have learned from observing your relative audiences in terms of how the project may have shaped their perspectives, really changed their perspectives, or enhanced their understanding of the city?
Furthermore, did it shape your own understanding of the city and if so, how did the technology change the way you view the city? That is, how did your journey in creating digital representations of the city change the way you understand these places?
Ellen Crocker: For the filming process of "Berliner sehen" we did quite a lot of footwork ahead of time to actually find out what these neighborhoods were like -- especially the people and their lives, so that we'd end up with conversations that were part of their normal life. And in doing that, one of the things that came out was the fact that a lot of these people in the east and in the west had known of each other and had waved at each other, but when it came time to sit down together, they had never been in close contact with one another.
They had never been as intimate as we were asking them to be, by asking them to sit down together and talk for two or three hours. They were being helped along a little by the filmmaker who was prompting them to talk about their lives, but he was not really asking questions. In other words, they were free to move.
Especially in the east, we had a lot of feedback afterwards from people saying that we brought up issues that haven't been brought up since the fall of the wall. This was very interesting to us in terms of our investigation of what is going on there. They have been very excited to see how this goes on further in the kinds of ways the students are working with them. So that is just an anecdote.
Kurt Fendt: I want to say something about how the students experience the life stories. It always depends of course, on the assignment, how the students interact with the material and what they come up with when they do those constructs. So, for example, one of the assignments was the various ways people talk and experience the neighborhood, Kietz. Some students reported that they had problems with the assignment because it included many contradictory views of what the neighborhood actually is. But dealing with contradictory bits of information is a skill, and these students have had difficulties with the same thing when just writing normal papers. So their problem extended into writing multimedia papers.
This system allowed them to actually bring together and combine those multiple different views so that they could actually come up with a coherent synthesis of what the neighborhood might be in all its varied aspects. So in working with the material, the students learned the skill of dealing with many different view points. I think that turned out to be an added service that we provided with this project.
Glorianna Davenport: I think that one of the things the students got the most out of in the New Orleans project was that they were able to call up characters in the film and find out what they had been doing in the three years since the filming had ended. They then added to the data and felt invested in the database. Their adding to the data effected how they saw it. What you said is right, that it was a fairly complex story that had a lot of different sides and they had difficulty sorting through which view point they wanted to take and how they wanted to look at it. But as soon as they started to be able to contact the characters, they got very involved in the story and what was happening. It argues for a constructivist learning environment. But I would argue for it beyond the classroom; I argue for it in real life.
Shigeru Miyagawa: As Gilberte Furstenberg, who is the pioneer of all this interactive material and from whom I have stolen all my ideas, said, you would never be able to see Japan if you were to go there yourself in the way that you can through StarNet, because you have someone who is there, showing you the city and sharing life experiences with you.
Just recently a publisher approached me having seen StarNet and said, "OK, now write a book so people can really understand what's going on here."
Question: Glorianna, that was interesting when you said you wanted to take constructivism beyond the classroom to a broader context. If I am not mistaken you are saying, "do this in a real city." Would that be a path to, in the future, a digital city and, if you take your view further out into the future, would it lead to more community in the city?
Glorianna Davenport: No, what I tried to point out is that this is an evolutionary process that needs to expand into a non-present communication. There is something about us that still lives "in place." And there are issues like politics and voting, for example, which are handled pretty much as they were three centuries ago in this country. I think that the reason community has become so important and also why ambient tangible interfaces have started to become important is because the digital world has to meet the physical world for the very reason that we are physical beings in a physical space.
But we also have very, very complicated cognitive processes that are much closer to what's happening to large networks. So the answer could be, "yes, maybe." But there are many bad guesses because people just have an idea and they have to evolve the product given the technology that they have. And it's not a bad idea, it's something that gets us a little further along the way to wherever we are going. So go for it.
But, for me, getting the images and being able to record an evolving documentary and make it part of the conversation space is something that takes more than one person being out there wanting to do it. You also need some way of describing the material, searching through the material and having the engine that presents the material in a semi-coherent or even in a coherent fashion, because otherwise you are just looking at that Web in front of you and you really have other things to do. So you can't participate because you don't have the time.
Recalling Linda Stone's point about the high-bandwidth of time that we have --what I would like is for all my time to have a little higher bandwidth, but I would also like the medium, the channel, to help me a little bit so that I can do that better, as opposed to how we have to do it right now with our piles and piles of email in thousands and thousands of folders. Well, that's better than UNIX, where you had some tarred back-up that you have to un-tar before you can get to some message from a year ago. So it's improving.
Bruce Joffe: I was whining before about the huge morass of information that was already overwhelming me. So I was very impressed with Ellen and Kurt's tool. Besides its specific implementation, it seems like a really useful tool for editing, re-categorizing, re-organizing, and allowing others to come in and add new categories and new ways of organizing. Whereas the other three -- and they were all excellent -- looked as if you and you team had to provide all the connections between one thing and another. Can you talk about how your interactors can re-organize the information? For example, Shigeru, your icons -- the photographs, the coins --each lead to a particular story by your design. Whereas in the last demo, one could recombine and re-categorize the elements to create a new story.
Kurt Fendt: You've noticed that in "Berliner sehen" there are no links. The links are created dynamically between the documents and they are just starting points. You can rearrange them infinitely. But there is of course the problem which Glorianna pointed out: the more students make collections, name their collections and publish their collections, then there is a question about how to organize the collections. So then you have the meta-descriptors.
It runs right now on the internal network, but it could easily be extended to the Internet and then how would you make those new collections accessible? What does the interface need in order to make that accessible? These are questions we are still thinking about and haven't solved yet. The engine is a stand-alone engine which we have designed in such a way so that you can plug in other kinds of materials. And the database that drives the arrangement of the materials is also stand-alone so that at any point, Ellen and I can go in and change something, without getting the programmers involved. We can plug in any new document and it doesn't really effect the engine.
We've had requests from companies to further develop the engine for a knowledge database, because that also allows people to get access to expert knowledge at many different levels. So if you're in a company that is trying to solve a specific problem, you could then look at it from a lot of different angles, whether you are in the marketing department or on the research and development team, and so on. So the system has the potential, as Glorianna mentioned before, to go beyond the classroom. And I think that is one of the important things that we learned while doing this project, that it can go beyond the classroom.
Theresa Duncan: Zero, Zero is a pretty self-contained CD-Rom. You said that you did not like all the information on the Internet, but I find that attractive. If you are doing research and you are looking for a specific subject, it favors accident and fissure. And if you are looking for something, you may wind up finding something else that is fascinating, although it may have nothing to do with what you were originally looking to find. It is like a thought association, a digression -- I may find something that I wasn't looking for and wasn't interested in, but it may send me in a new direction.
Another thing about Zero, Zero, is that a lot of children's stuff is didactic, but this is satirical in the way that Mad Magazine looks at stuff satirically. It's that anarchic sense of fun. And I really wanted to make it a profound aesthetic experience for eight year olds. I didn't want it to be just some toy, and it didn't have to have some lesson to convey, either. But I did want it to be the most beautiful thing an eight year old has ever seen.
Shigeru Miyagawa: We actually began this project as a Japanese language project six years ago and as we explored this approach to education, it became apparent to us that in order to build the kind of project we envisioned we had to really pay close attention to aesthetics. So we started to think about what that meant. And what we came up with was a sense of playfulness that users bring to this kind of environment. So we made an attempt at creating a space where users can have a sense of playfulness as an aesthetic.
Kurt Fendt: There is also the mystery aspect. For example, you get a very fragmented view of the material in the beginning so that you just get a snippet of the story. That leaves the students with the motivation to search for the untold. In that way, they can be drawn in by their own desire to search for more clues which will lead them to discover what happened. It is not the traditional way we perceive games, but there is another kind of playfulness built in that has to do with the material and the way that it has been arranged for students to play with.
Bruce Joffe: Learning and creativity are very often the result of the unintended encounter and putting two discordant ideas together and suddenly you have a new idea. And yet, I think that our culture today is so fragmented by distractions that kids are coming out of school with a 10- to 12-second attention span. Maybe we need to give people the opportunity to focus. Certainly with all the work that you people have done, you never would have completed it if you had just gone off on every little tangent.
Theresa Duncan: But these media are not passive. This isn't a video where you plunk a kid down in front of a television. You have to physically move through it and that is the whole point. You have to engage them in the beginning. And in doing that with multiple points of entry or different areas of exploration, they can go to what intrigues them the most and then try the other places because they enjoy it. So part of it is putting a little candy on top, but there is still a lot of information behind it all. And it is not about fragmenting attention, it's about applying the psychology so that they can have a more profound experience. I think it's effective. Zero, Zero is entertaining, but I have worked on educational CD-Roms and I have seen people sit down and become completely absorbed by it all.