tangible interfaces to computation
As interactive bitmapped graphics became viable in the 1980s, the human/computer interface began transforming from text-based interfaces to GUIs. Today we are engaged in another transformation. As computation becomes more ubiquitous and available in small, low power, networked and distributed form factors, computers are entering our shared social space.
We carry them with us and use them to verify entry to places, to prepare us for our next meeting, to send each other messages during lunch. A shift is occurring in when and why we access information and in how we can use computation.
This talk looks at the tangible interface to such ubiquitous computation, covering the development of interface design strategies and applications of computation that exploit the tangible and physical aspects of these new devices. In addition to the ‘privatized mobility’ we have come to expect from shrinking processors and increasing power efficiencies, some of the more interesting characteristics of this realm of design are the exploitation of peripheral attention (for monitoring); tacit skills in manipulating physical objects (intuitiveness); multiple information retrieval, management and interpretive strategies from a lifetime of manipulating physical devices (priority; persistence, particularity) and the capacity for impedance matching of displays not just with single users but with diverse groups – for supporting many person interactions within the rich social contexts to which we are accustomed.
This talk will review several projects that will serve to demonstrate the possibilities of this area and to introduce methodologies for evaluating and understanding the complex interactions involved in these tangible and social uses of computation. We will review the problems inherent in generalizing this approach and will discuss the use of tangible media as a proxy for social interaction, experimental methods, and directions for ongoing and future research.
This work examines the use of physical interfaces (tangible media) as a design strategy, and the implication of a physically based design approach for the development of information products.
The focus will be on examining characteristics of the physical presence of (information containing) objects that can be used to design information products. These characteristics include persistence, peripheral presence, priority and particularity.
Physical objects - in contrast to digital objects – are persistent; they do not ‘disappear’ when they are not in use. An aspect of such persistence is peripheral accessibility; physical objects continue to be where you left them and are therefore readily accessible for just-in-time reference. In addition, their visibility does not map completely to the attention given them, unlike digital counterparts. They can be thought to have an independence from their users’ attention scheme because they do not appear and disappear according to whether they are in use. They are therefore attention-directing devices. As such, physical devices exploit peripheral information cues - sound and movement - to indicate state change, thereby unobtrusively redirecting attention when needed.
The priority given to an object with which one shares the physical environment, i.e. why people print out important email messages, has to do with ‘planning’ environmental interactions that are unpredictable. The objects involved are placed as obstacles to come across or to interact with in conscious or non-deliberate ways. Tracing this strategy of putting things in ones paths and arranging the physical environment as an information space informs the design of information environments.
The particularity of physical objects and their contrast to textual representation, provides a difference to be productively negotiated and reveals textcentric assumptions. However is it the non-abstractable specificity of physical objects that is the main focus of this characterization, and is exploitable in the design of information.
These aspects of physical presence will be examined with reference to their effects on knowledge production; learning; and monitoring activities, and the subtle difference between these types of attention. Furthermore, the aspects of physical ‘things’ under investigation will be applied to ‘entities’ categorized as soft, hard and live.
The investigation are undertaken using two primary strategies:
Firstly, the empirical methodology is the analysis of the footage generated in longitudinal video observation studies of interactive products in situ. This material is collected with a series of custom, low-power, sensor-triggered video systems that I have already designed, tested and in some instances deployed. The sensors trigger on interaction events between the product and the user, collecting video data on the seconds of interaction. This allows observations over months of product use making new aspects of interaction observable. This study differs from other systematic studies of interactive products because of the longitudinal nature of the observation. By changing the way we observe interaction with products, the novel representations may point to opportunities to change the design of interactive information products – in particular with respect to use over the product life.
The second strategy is the design and implementation of tangible media products. The group of designs implemented will explore the parameters discussed above, examining the pragmatic design issues involved in the implementation of these design directions, and facilitating the observation of their efficacy in actual work and use situations. These design projects:
a) OneTree and OneTree sensor system: a biological/computational instrument with a hybrid network of low-power sensors for deployment throughout the Bay Area to facilitate longitudinal (50years) observation of cloned trees, coordination of a distributed web based data collection system, and the comparison with an A-life tree system model;
b) ITS video system (discussed above for data collection): the design and implementation explores issues of coupling to a hybrid network of physical inputs, sensors and triggers in a jini-implemented network, and as such addresses the issues of networked tangible devices;
c) handheld – based learning tool: the implementation and testing of a learning support tool that implements a recurrent autonomously timed system using a Poisson pacing in support of formal training syllabus; (implements the persistence parameter of physical presence sans other aspects)
The implementation of these designs will populate examples of specific tangible design strategies and the issues involved in generalizing this approach. The theoretical implications will be examined in the light of actor network theory and other relational approaches.