Steve Harrison, Scott Minneman, Anne Balsamo
XFR: eXperiments in the Future of Reading is an interactive museum installation designed and constructed by the Research in Experimental Documents (RED) group at Xerox Palo Alto Research Center (PARC). XFR was installed at the Tech Museum of Innovation in San Jose (referred to in this article as the Tech Museum) from March 1 through September 5, 2000. It included 10 interactive exhibits that focused on new reading experiences, as well as a working, state-of-the-art Book Artist's Studio. The show was a culmination of 20 months of research, design, and development by the eight members of RED, who were supported by consultants in graphic design, computer programming, exhibit fabrication, and large-scale printing. By the end of the six-month run, 300,000 visitors had seen the show. The show is described in more detail in the accompanying article, "The What of XFR."
The design and creation of a museum show is not a typical PARC research project. Collectively, as well as individually, the members of RED are charged with the challenge to design and to develop prototypes of new media genres and new document forms. This charter provides the broad rationale for RED's research in new modes of interactivity and innovative research methods. The focus on new media genres resonates with Xerox's interest in digital documents and new modes of knowledge sharing. RED's use of the notion of "genre" comes from literary theory, cultural studies, and semiotics, from which RED borrows some terminology and methods. This focus on genre directed RED's research toward investigating how meaning is constructed within a particular discursive and institutional situation. This situation is structured not only by the properties of texts (the traditional emphasis of a literary approach to genre research), but also by technological factors (e.g., screen resolution), social categories (e.g., the identity of the reader), and cultural practices (e.g., traditions of storytelling). In that regard, it also influences the means by which we design and create. This work is an ever-emerging process. This article describes how we explore as we create and offers reflections that may be useful for others exploring this territory.
Although the RED group began with a broadly defined interest in genre research, it was unclear how this interest would translate into a set of design strategies. But strategies gradually emerged throughout the project and drew on the creative practices and traditions from our backgrounds in art, science, engineering, and design.
On Seeing Genre
Genre is both the product of our inquiry and an analytic framework. By explicitly adopting a "genre perspective," we sought to illuminate the conventions for constructing interactive exhibits in both museum and entertainment settings. These conventions are simultaneously "encoded" by exhibit designers as well as "decoded" by exhibit visitors. Borrowing methods of genre analysis from literary studies and semiotics, we looked at how meaning is built into and conveyed by exhibit form, signs, and layout and how the visitor constructs meaning around the interactive experience.
On Working with Conventions
As we learned to identify the genre conventions that influence the meaning of interactive exhibits, a design strategy emerged and we began to design consciously against those conventions. Our goal was not to be unconventional just to be unconventional, but to explicitly counter conventional wisdom in the aim of producing a new genre of museum exhibit. This "designing against" conventional wisdom operated at different levels throughout the project, whether it was to test the conventions of interactive museums, to reuse an existing technology under development at PARC, or to rethink the practices of user-centered design.
On Creating New Genre
Drawing on our backgrounds in art and design, we understood the power of combining pre-existing forms to create hybrid modes (e.g., rock opera, or wearable computing). This method of idea generation fit quite easily into our enterprise of creating new genre. The combinatorial approach brings forward specific characteristics for the new designnot only the formal aspects of the media and criteria for the creation of content, but also social conventions and narrative structures.
A Group Project Emerges
In the summer of 1998, PARC was invited by the Tech Museum of Innovation in San Jose to contribute a short-run exhibition on art and technology. One of the galleries in the then-unopened museum was to be dedicated to showcasing innovation-in-progress from Silicon Valley research labs. Our first inclination was to build a show around projects from PARC's Artist in Residence (PAIR) program. As that became logistically unworkable, RED proposed creating an exhibit based on current PARC technology.
A Topic Emerges
After devoting several brainstorming sessions to finding a topic for the exhibit that would draw on the research interests of every member of the group and that would showcase current PARC technology, Rich Gold, manager of RED, decided that the topic of the show would focus on reading. This was one of few instances in which he managerially mandated an aspect of a RED project. More typical for the group was the use of the "manifesto" as a focusing tool and a mechanism for developing a common language. As the group developed various lists of possible exhibits or technologies, Rich would spontaneously create a new manifesto to rationalize the organization of the exhibits in a specific way: either as covering a range of characteristics, for example, or telling a particular story about PARC. These manifestos would appear as e-mail messages with titles like "Monday Morning Musings" and would often include a proposal for a new name for the show. Always spontaneous and unscheduled, these manifestos appeared at least a half a dozen times during the development of the show.
The manifestos created the occasion for informal and unstructured group discussion through e-mail postings, name games, alternative exhibits, and alternative organizing principles for the entire show. Subsequent group meetings often were shaped as a response to the current manifesto. Although the discussion of exhibits seemed to dominate, we also used the manifesto to evaluate the coherence of the entire show. This enabled the group to collectively internalize the commonalities of the message we intended to deliver.
This focus on reading provided both opportunities and constraints. The topic was easy to explain and justify to PARC management, and later to Xerox marketing division, because of its relevance to the core technology of Xerox. When asked the question "Why Reading, Why Xerox?" Gold answers, "We [Xerox] make things [printers] that make things [documents] that people read." There was certainly a strong appreciation among RED members of the social and cultural value of the topic. Of equal importance was that the topic had strong connections to previous research by various RED members. This meant that the group project would also be able to incorporate the research contributions of individual RED members.
A Stance Emerges
One of the constraints imposed by the topic was that it often provoked a less-than-enthusiastic initial impression from the museum professionals we were working with both at the museum and as consultants. A technological show about reading didn't have the powerful attraction that an exhibit about, say, dinosaurs, or robots, had. We were told repeatedly by various museum professionals that no one reads in a museum, and worse, that no one would purposely visit an exhibit with "reading" in the title. The Tech Museum staff had done an informal focus group of one of the early names for the show and indeed found a notable lack of interest. But we persevered in believing that not only was the topic appropriate to the museum's context but also was an interesting way to frame a story about the significance of current PARC technology. Continually having to reaffirm this belief in the face of skeptical collaborators encouraged us to amplify our unconventional approach to the process and practice of museum and exhibit design.
We certainly challenged dominant paradigms of user testing. Although we did conduct a series of informal user studies by bringing in children to use exhibit prototypes at different times during our development process, we did not evaluate user experience in the conventional way that either the human-computer interaction (HCI) community or the museum exhibit design community might recognize. We did not, for example, systematically analyze user performance with a particular interface. Nor did we conduct any beta tests of individual exhibits before the show opened in the actual context of the museum. Although we learned many things from the way in which various groups of children interacted with the prototypes, RED found it more productive to spend the time grappling with the fundamental questions involved in creating new genre: What is "usable?" Is it better to design for the majority of visitors or for the ones who really get the message? Who gets to decide what is usable? How does an exhibit communicate that it is the message and not a medium for explaining something else?
A Research Perspective Emerges
Initially, we understood the notion of a new media genre in a rather narrow sense: as relating to new document forms. Throughout the project we returned to the observations we made of different museums and entertainment complexes, actually treating them as genres. By sharpening our genre perceptions, we could identify more clearly the elements of that existing genre; there were key differences in setting, in the identity of the audiences, and in the cultural purpose of exhibits. We came to see the many similarities in the terms of our analysis between new media genres (as part of our research charter) and the analysis of science and technology museums that we had done as background research on the show.
How does an exhibit communicate that it is the message and not a medium for explaining something else?
We were tracing a chain of meaning-making by looking at a wide range of things that actually provide meaning for the museum visitor. Whereas understandings about the symbolic purpose of museumsthat they reproduce cultural stories, for instanceis not a new insight; analyzing museums in terms of genre conventions lends insight into the process whereby meaning is created in the museum encounter.
Different kinds of museums have different conventions of display, narrative, explication, and modes of interactivity. In this sense, there are several subgenres of museums. The science center might be considered one subgenre, the hands-on technology museum another. More specifically, we came to appreciate the importance of exhibit signs as a highly conventionalized form of documentation within the subgenre of technology museum. Through the use of specific rhetorical conventions, exhibit signs address, and in many senses, create the identity of both the museum visitor and the museum exhibit "creator." Use of direct address (through statements such as "when you push the button you will see...") rhetorically connects the reader with the immediate experience, reinforcing the "hands-on" quality of technology-based exhibits. But we also became familiar with the other elements that contribute meaning to a museum visit that don't look like documents at all, such as the organization of exhibits, the relations among collections of exhibits (the taxonomy of exhibits), the materials of the built pieces, and the structure of the physical space itself.
A Set of Technologies Emerges
We were also involved in evaluating which technologies at PARC would be appropriate for this setting and this show. We focused our search on emergent technologies (both those at PARC and those that we were interested in exploring in our own research) as the basis for the exhibit. PARC technologies are always being demonstrated; it is one of the traditional ways that research is communicated to the business divisions of Xerox and to other research audiences. It was easy to cull a list of "mature" technologies at PARC, those that were then the subject of successful demonstrations. But a successful demonstration in a PARC setting wasn't sufficient to ensure a successful exhibit for the museum show. Among other things, the technology demonstration had to be accessible to a general audience; it had to allow some mode of interactivity and provide evidence of that engagement, and it had to be robust. It is one thing for a researcher to demonstrate his or her technological prototype in the context of the research lab. It is quite another matter to use that same technology to create an exhibit that is sturdy enough to withstand six months of daily wear and tear by a range of museum visitors (some of whom are intent on seeing if the technology can be broken).
Besides availability and robustness, the topic of reading limited our selection of technologies. It was difficult to imagine how modular robotics or MEMS (micro-electro-mechanical systems) technology could, in its current stage of development, serve as a basis for an exhibit on reading. Although extensive group discussion gradually refined the topic to a more specific focus on the relationship between reading and new technologies, it was critical that the broad topic was decided so early in the process.
Throughout the two years of project development, we found ourselves asking genre-centered questions about all levels of the project.
Seeing Genre in the Details
The analytic inquiry led us to think about the particulars of existing document genres. We looked at all sorts of books, Web sites, printed ephemera, and text in the environment. We asked, for example, what made a "coffee table book" a coffee table book, that is, what were the defining characteristics of the "coffee table book" genre? This sort of analysis proved useful as input for brainstorming sessions on the formulation of new genres and for evaluating the genre of our speculative designs. In particular, we paid attention to the form of books and printed materials and the relationship of form to the content. One of the recurring metaphors in our explanation of our endeavor was the "elephant book"children's book in the shape of an elephantthat has the property of a highly synchronized form and content.
We also studied the conventional document forms of the science and technology museum. We wanted to find out how visitors get the meaning of an individual exhibit. Do they read signs? How much explanation is desirable? What are the typical modes of address? In some cases, this analysis focused explicitly on the generic conventions of a particular type of museum document (e.g., the genre conventions of timelines). In other areas we borrowed other people's analysis of genre conventions. For example, because we included comic and graphic storytelling in several exhibits, we used Scott McCloud's book Understanding Comics. Each proposed exhibit was designed to be interactive in some way. Thus we were also interested in an analysis of conventions of hands-on museum interactivity, such as interface design, input devices, and output modes. This included an assessment of the visitor's experience. How long do visitors spend with individual exhibits? Will they perform sequential steps? What state will they leave an exhibit in?
Another source of useful information came from examining the textual elements of individual exhibits. Docents' instructions often include information about the important messages embodied in an exhibit. The comparison of those messages and the experience of an exhibit were invaluable in helping us identify salient genre conventions. Technical manuals and occasional reverse-engineering helped us assess the technical sophistication employed by an exhibit. They were also a source of statements about standards, requirements, and pragmatic details relating to the built form of exhibits (i.e., Americans with Disabilities Access Act requirements and the point size for signs read at various distances). Although the museum professionals we worked with could describe the grade level of vocabulary that should be used in signs, our rhetorical analysis revealed other conventions such as the frequent use of a disembodied voice of authority and the unsigned author. Reading closely and critically, we also noticed how the particular point of view of corporate sponsors showed up in various exhibits.
Seeing Genre in the Context
We studied exhibitions at the Tech and other museums using direct observation, video recording, and photographs. These were augmented with interviews with museum professionals. Visits to the Exploratorium, to the Metreon (the new Sony entertainment complex in downtown San Francisco), and to the California Academy of Arts and Sciences helped us discover how large interactive installations tell many stories at both at the manifest and latent levels. Most of our early understanding of specific genre conventions arose through our interviews and meetings with museum people. To verify these conventions and how they manifest across exhibits, we analyzed our videos, notes, and photographs. In particular cases, the data we gathered departed from the accounts given by the professionals we interviewed. We were told that visitors seldom finish sequential exhibits, but in fact, we observed several successful and entertaining exhibits that provided a highly structured, step-by-step visitor experience.
We assessed how "the museum" in general serves as a defining context for exhibits and galleries, to ask the question, for example, what domain of exhibit belongs to the "technology" museum, as opposed to the "science" museum? We also tried to assess how visitor expectations are shaped: how much do they want to be shown and told, how much do they want to discover? What experiences will the visitors to our show bring with them from their tour of the rest of the museum? What conventions does the Tech Museum use to tell an overarching story about its exhibits? As a group we started attending to the peculiarities of large traveling exhibits, including the two that preceded XFR. We were especially interested in issues of scale, density, flow, and space organization.
Creating the Details of New Genres
Although there was no one typical path of development for each exhibitas it moved from an abstract idea to a built piece on the gallery floorcommon phases or stages marked our process of exhibit development. Exhibits were generated based on one or both of these provocations: (1) a technology that we wanted to have in the show, (2) a reading experience or an argument about reading that we wanted to include. A second stage included a thorough examination of the mode and mechanics of interactivity for each exhibit. Another stage involved a detailed assessment of the narrative and graphic content. See the accompanying sidebar, "How We Organized Our Work," for more detail on the kinds of activities we took part in.
Although we were strongly influenced by and respected the design conventions of both the HCI and the museum exhibit design communities, we also departed from these conventions in significant ways. For example, our mode and process of exhibit production differed greatly from the accepted process of museum exhibit designers. Exhibit design is often divided into discrete stages, and each stage is handled by a different group of professionals; the design of the exhibit precedes the hand-off to fabrication experts. Our exhibit design process also included research activities about new reading technologies and new reading experiences; consequently, our design, fabrication, and reengineering phases were intertwined and ongoing. Similarly, museum exhibit designers often assert that exhibits should include only enough interactivity to get the point (of the exhibit) across to the visitor. This conventional understanding about the desirable range of interactivity made no sense in the context of our exhibit. Each of the individual exhibits we designed was about the exploration of a new form of interactivity (reading). Our exhibits were designed to encourage visitors to make repeated use of the device as a way of learning to use it to read a new and unusual document. Designing exhibits to encourage repetitive use resulted in longer "visit" times. Whereas the typical visitor spends nine to 12 seconds at other exhibits at the Tech Museum, visitors to the XFR show spent an average of 25 to 27 seconds per exhibit. (These visit latencies were determined by an independent reviewer hired by the Tech Museum to evaluate the success of the XFR exhibit.) In learning about visitor latencies, we learned that within the genre of museum exhibit design an exhibit that requires a visitor to spend too much time is considered a failed exhibit. This piece of conventional design wisdom reflects other genre considerations, such as the desired rate of throughput (i.e., the number of visitors that can flow through an exhibit within a specific period of time) and the number of visitors served by a particular exhibit. Although clearly we shared the ultimate objective of professional museum exhibit designers, namely, to build robust, interesting exhibits that could be used by a wide range of visitors, in basing our design process and methods on a "genre perspective" we departed significantly from a typical design trajectory.
A similar analysis emerges for the typical design and evaluation process promoted by the HCI community. Whereas that community often defines good design by productivity and efficiency, where the design objective is to facilitate the "use" of an application or interface by an end "user," our design objectives were slightly different. Where we departed most significantly from an HCI approach to digital design was in the conception of the end user of our design work. Rather than consider our designs and exhibits as involving a "user," we used the concept of "visitor" to identify the intended recipient of our design work. This insight was the result of using a genre perspective as an organizing framework for our design work. The concept of "visitor" is a genre convention of museum designers. It is used to identify members of the public who will interact with the museum's collection (of exhibits, programs, displays, and so on). Using the term "visitor" to describe the intended recipient of our design work, rather than "user" or even "audience," imposed (or strongly suggested) certain design considerations. For example, visitors are guests of the exhibit in some sense. Thus, even though they may pay an admission fee, they are to be treated with courtesy and respect. In keeping with this understanding, the purpose of our exhibits was not to confront them or ignore them. Rather, our aim was to understand their expectations and provide meaningful clear experiences, and to do so with respect and consideration. Pragmatically this suggested that we pay close attention to vocabulary, mode of address, and content. We scrutinized the exhibit in terms of representations and language use. Our intent was to create a "balanced" exhibit that would be of interest to a range of reading publics: from adults to small children and across gender, race, and nationality. Whereas an individual exhibit may not have encompassed a full spectrum of representations (for example, many exhibits used English as the only language), the exhibit taken as a whole was designed to reflect this spectrum, and indeed included pieces that were multilingual, multigendered, and multiracial.
Creating Wholly New Genres of Exhibitions
Because the topic of reading is so broad, we had to limit the scope of our research and design of new reading experiences. This was partly due to the necessity imposed by limited resources, but also other exhibits at the Tech Museum influenced us. Unlike a science museum, which has a mission to include exhibits about basic scientific principles, the Tech Museum focuses almost exclusively on technologies that affect daily life. Although we know that science and technology are inextricably linked, the Tech Museum focuses its collection of exhibits on a delimited range of contemporary science and technology phenomena. For example, the exhibit on laser eye surgery doesn't address the underlying biology of visual perception; rather, it focuses attention on the apparatus used in eye surgery, and through a hands-on demonstration allows a visitor to experience the skillfulness required to use such a tool. Even as the collection of exhibits at the Tech Museum were delimited to a particular range of science and technology topics, there is also a sense in which the museum takes pains to cover the topic of contemporary technology.
We found it useful to think of our installation in the same way: to make sure that we were "covering" the domain of reading in some way. To help us rationalize our selection of exhibits to include in XFR, we used a modified semiotic analysis to organize and code the range of proposed exhibits. We began by identifying the main dimensions of the project as it emerged. These included (1) a focus on public versus private reading, (2) a social or collective versus individual reading experience, and (3) examples of machine versus human reading. Identifying the main characteristics of each exhibit helped to determine the overall coherence of the show in that we were able to see how we were "covering" a range of possible reading situations and technologies. This also helped us identify the dimensions of reading that we chose not to address in the show, for instance, the study of the cognitive aspects of reading or the sensory aspects of reading.
This technique of identifying the primary messages of each exhibit influenced our combinatorial creative strategy. (The accompanying article describes the design of Hyperbolic Reader: Henry's World and Fluid Fiction: Harry the Ape, which resulted from this method.) The exhibit called "Walk-In Comix: Slip to Text" crosses wall reading with graphic novels. We found that placing an existing genre on a new medium was easier to understand and imagine than creating new genres on old media.
We also used strategies from engineering such as techniques for factoring problems. This worked in the few cases in which we had a couple of principles we were trying to demonstrate. We had a cluster of PARC technologies that we believed to be a good basis for exhibits on reading, such as text-to-speech software, summarization programs, word spotting, and automatic translation. Our attempt to create an exhibit that demonstrated some or all of these technologies required us to rework the various ideas to address the main characteristics identified in the semiotic analysis. Many attempts did not provide satisfactory results. In the case of "RED, the reading eye dog" we reduced the number of main characteristics that we wanted to demonstrate and combined them into a single exhibit.1
This generative genre exploration was taken to its logical conclusion when the name of the show became "Dispatches from the Gutenberg Galaxy." We called the new genre a "muzine"a neologism that connotes both "museum and magazine." Working with the concept of a "muzine," we struggled to find a form for the exhibits that expressed printing, such as housing each exhibit in an oversized book or document. Very soon, the "muzine" hybrid metaphor became stifling and cumbersome. It also created a closed kind of language about the exhibit that contrasted with the ever-emergent form that typified the rest of the process. Where there was initial group enthusiasm for the large-scale new genre, subsequent discussions of planned exhibits could not be easily framed in this particular language. Given the choice between a strong metaphor and strong exhibit, the strong exhibits won.
Having a name for the show was not enough. Abandoning "muzine," we returned to finding the meanings already present in the agglomeration of exhibits we were creating and in our own process. We were unable to turn over much of the execution of individual exhibits to outside fabricators, content producers, or programmers because of the deep interaction between form, content, and meaning. Likewise, the creation of the new genre of exhibition became simultaneously a naming process, a design process, and a production process. The message that this exhibit was to represent research was then conveyed by using materials that we, as researchers, could use in experiments and forms that would be read as experimentaland most important, that we actually did build. We chose a design metaphor, mostly expressed through lighting, of "the research lab at night." We based this on photos of PARC labs at night. Exhibits and signs were built of aluminum extrusions and connecting hardware used in lab benches.
So, in the words of Rich Gold, "We were writing all the way down." That is, at all levels of the show we were thinking of how the experience would convey the meaning. The final name of the show, "eXperiments in the Future of Reading," was adopted just months before opening. This name provided the narrative and conceptual coherence we were seeking for the collection of exhibits we had been working on for so many months.
From many accounts, the XFR show was a grand success. As researchers and designers we are enthusiastic about the ideas we uncovered while building the exhibit; ideas that will continue to influence the new media design work of RED. We also learned a lot about how we work and how we don't.
As this article comes to publication, the XFR show is being prepared as a traveling exhibition. Built originally to last only six months and fit exactly into the space at the Tech Museum, the exhibits need to be hardened against aggressive visitors and many years of travel and generalized to fit into wildly different facilities. The exhibits are being further strengthened by a professional exhibit and trade show design and fabrication firm. Our improvisational processes will contrast with those of the "professional" exhibit design world. As we learn how to integrate our world with theirs, we will be creating a form of technology transfer. That should generate more "how" and a slightly different "what."
Steve Harrison, Scott Minneman, Anne Balsamo
Xerox Palo Alto Research Center
3333 Coyote Hill Road
Palo Alto, CA 94304
Methods & Tools Column Editors
Lotus Development Corp.
55 Cambridge Parkway
Cambridge, MA 02142 USA
IT University of Copenhagen
2400 Copenhagen NV
+ 45 3816 8888
fax: + 45 3816 8899
Figure 2. Characteristics of science and technology
museums and the exhibits in them. The genre of interactive
exhibit has lots of variety but is clearly demonstrating
something other than itself and often has large instructional and
descriptive signs. The collection of exhibits, their relative
placement, and kinds of visitors forms a particular genre of
Figure 3. An early prototype of SpeederReader.
Superficially like a video gameand therefore attractive to
visitors who might not approach the topic of readingit
posits a genre of reading against the current convention.
The project began in the late summer of 1998; the show opened in March 2000 and closed the first weekend in September 2000. The main stages of the design and construction process included: 1) brainstorming, 2) message creation, 3) project planning, 4) prototyping, 5) design review, 6) user testing, 7) fabrication, 8) installation, 9) remediation. This was not a sequential process, activities overlapped and individual pieces moved forward at their own rate.
As the project began and then throughout its duration, the group met frequently, evolving its own style of brainstorming and review. The multidisciplinary nature of the group really helped with the brainstorming phase of the project. The group alternated between traditional non-evaluative brainstorming and fleshing out the ideas so generated for several weeks. After an initial list of possible exhibit topics and projects was created, the list was critiqued and evaluated in several ways, including applying a semiotic analysis.
In parallel with generating and refining ideas for exhibits, the message and names of the show and each exhibit were worked and re-worked. Getting a focus on reading was simple compared to the process of figuring out what to say about reading. While RED researched the topic of reading (reading Alberto Manguel's A History of Reading, for example), the exhibition and its constituent genre-exhibits needed to say something about reading.
Of course, reading is not the only theme that focussed the design workthere was a need to communicate about the company and our portion of it. We had a number of meetings with "messaging professionals" representatives of Xerox Marketing and Xerox's consulting PR firm to identify what the show should communicate. Some were of the sort: "Reading will be even more important with the advent of digital technology," others were of the sort "Xerox is a high-tech company."
There are two major senses in which RED tackled project planning: planning what to do and planning how to do it. As can be seen in the companion article, planning what to do was a great success. Planning how to do it was less so.
The "what" planning was carried out by the combination of brainstorming, message creation, and prototyping. A preliminary list of projects was created in the Spring of 1999. A final list of exhibits and all consultants was established in the Summer of 1999.
In contrast, the "how" planning was carried out initially in a variety of semi-formal methods and finally in an ad hoc fashion.
No group member had all the skills necessary to accomplish all identified tasks. We needed to develop methods of project planning that accommodated our particular division of labor. This was difficult at times because people had to estimate and count on a measure of labor from other group members without knowing exactly how long various sub-tasks would take.
There were two attempts using traditional project planning methods to coordinate the project. The first was a collective attempt to plan the projects that resulted in some general idea about expected participation in projects. This attempt at project planning used a modified delphi method to arrive at a schedule of tasks and skills for each exhibit. Using the preliminary list of projects, tasks were composed on Post-Its, attached to sheets of foam core, and arcs drawn to link and sequence tasks. From this, the number of days of work and specific people were identified. Many participants were uncomfortable with the process since it exposed the enormity of undertaking, the unreliability of the predictions, the formality of the method, the mismatch between needed skills and available resources, and an imbalance between individuals' responsibilities. The foam core project plans were posted in the meeting room for the duration of the project, but never updated or referred to in subsequent meetings.
That was followed by another attempt using a designated project manager and computeriszed methods to track resources and the work. Data collected from the foam core/delphi method was entered into Microsoft Project. Not only did Project prove to be unsuited to the task of projecting tasks and monitoring process, it proved unwieldy to update. It was abandoned after a couple of months.
Project planning then fell to each individual. Culling a long list of brainstormed ideas into a couple dozen projects, project leaders were selected to create prototypes. Responsibilities were divided among members in two ways: some members had responsibility for over-arching tasks: i.e., project management, budgeting, space layout, computers and displays, fabrication, art direction, and each member (designated as a project leader) also had more focused responsibility for at least one individual exhibit. Each project was fully collaborative in that each member of RED participated to some degree in the design and construction of all of the 11 exhibits. However, there was no load balancing or project modification made based upon resource availability.
Prototyping, Design Review, and User Testing
Each project leader was responsible for producing increasingly refined prototypes, many of which eventually became the on-the-floor exhibit. The primary programming environments were Java and Director. Most were built using a bolt-together mechanical prototyping system available from Robert Bosch (in Europe), 80/20, and Item Products which could be assembled by members of RED. The components consist of unfinished aluminum extrusions cut to length and a variety of connecting hardware. Mostly such systems are used to create factory-floor assemblies to support robotics and assembly-lines.
Design reviews were held about every four months where the prototypes would be displayed and critiqued. These reviews were synchronizing events among the many disparate pieces and the synchronizing method for the project.
Closely coupled with the review process was user testing. Coupled in the sense that they together produced the most reflective questions and greatest tension in the group. We did a series of informal user studies bringing children in to use the prototypes.
All of this was accomplished by selecting a design theme and aesthetic that accommodated a prototype look and feel, and a primary method of fabrication that could be tackled by all members of RED. While the fabrication of most exhibits was accomplished by RED; some complex items were created by machinists and cabinet makers.
In mid-February 2000, the unfinished exhibits from the consultants and the ones at PARC were delivered to the Center of the Edge Gallery at the Tech Museum in San Jose. With the exception of some programming, the fabrication process relocated to the gallery as well; in fact, in another break with the traditions of interactive museums, much of the exhibit fabrication was done on-site.
On Monday February 28 at 8:00, the temporary walls around XFR came down, the carpets got cleaned of bits of wire and metal filings, and the opening reception was on.
On Tuesday February 29, we went back to work completing exhibits while an enthusiastic public explored the rest of the exhibition. Most incomplete was Walk-In Comix and the signage system. In short order, a simpler design for the exhibit-by-exhibit signage was worked out and installed. Walk-In required efforts of professional carpenters from the Tech staff, the use of noisy tools and odiferous glue, reprinting of art work, and large floor area for assembly. All of this attenuated coordinating and completing the exhibit. We worked mostly at night since the exhibit was open to visitors, but this meant working aroundor after, evening receptions. After about a month, we were done.
A month after that, we decided there were lots of things left to be done and a few things that needed fixing. In the exhibit design parlance, this was a remediation phase. Exhibit-by-exhibit, priorities were set and worked. Work on remediation continued right up to the closing day in early September.
©2001 ACM 1072-5220/01/0500 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2001 ACM, Inc.