Research Agenda

    My areas of research interest include:

    • Augmentation
    • Document Processing
    • Interactive system design, particularly patterns
    • Standards and the standardization process
    • Visualization and virtual information spaces

    Students looking for ideas for research projects, thesis and dissertation topics, and papers should look at the ideas for research projects.

    Current research efforts, most of which are connected to the CASCADE project are described below. CASCADE is a research testbed designed to provide a space within which graduate students can conduct research projects on a variety of topics in a functional context. The background section of the CASCADE web site provides more details on the research questions that CASCADE is addressing. (The CASCADE web site link above will spawn a new viewer, which you may simply close when done.) The interested reader might want some background on how this eclectic set of research interests came into being. Some history is provided at the end of this panel.

    • Research on techniques and methods to develop docubases. Docubases are DBMS like systems optimized for structured documents. This means that they address the issues of concurrency, locking, replication and granularity but in a way that is optimized for how people work together on documents. While DBMS systems work well for records, the usage patterns for document components are different enough that simply putting the document elements in a DBMS may involve more overhead than is needed and less than optimal structuring of the data to conform to relational models.
    • Research on techniques and methods for visualizing document spaces to aid navigation and other document related tasks. Visual navigation of data spaces involves finding appropriate mappings of abstract data to the sensory domains. A couple of the papers under the Papers panel address work here. Specifically, the paper on "Multilevel Navigation of Document Spaces" and the paper on "Virtual Reality and Abstract Data: Virtualizing Information".
    • Research on techniques and methods for agent design and integration into the work process. In particular, we are looking at ways to integrate social awareness agents into software systems. This would enable us to provide a sense of the social periphery in collaborative software systems.
    • Research on interface design, particularly several efforts to define methods to "engineer" interfaces. In essence, I believe that we should be able to predict, with a high degree of certainty, that an interface will work. In order to do so, we need to be able to identify the questions that need to be asked during the analysis and design and to have stable and reliable principles that guide the building of the system. One of our efforts to identify design patterns is included in the Papers panel -- "Patterns for Human Computer Interaction"
    • Research on information technology standards and the standardization process is also of great interest. There are several papers under the papers panel that address different topics here. I also offer one of the few courses at the graduate level that addresses both the theory and practice of standards development as well as providing an overview of the state of standards development in the information technology area.
    • Last, but surely not least, I have an avid interest in where we are going with electronic documents. I am firmly convinced that we are in the midst of a document processing revolution that makes what happened post the introduction of moveable type pale in comparison. This interest is, at this point, less a matter of empirical data gathering, and more one of understanding and projecting the trends observed in the marketplace. The paper with Jeff Campbell on "The Document Processing Revolution" provides an indication of where my thinking is today.

    History

    The eclectic nature of my research agenda may be explained in part by some history. When I came to the department in 1986, my interests were in the area of large scale document processing, particularly, the creation and formatting of large documents that might be produced on demand. I was in the final stages of the Planet Earth project, a project funded by the Corporation for Public Broadcasting and the Annenberg Fund, to produce custom textbooks on demand. We had, in the mid 1980's, successfully managed to reduce the time to produce a fully featured textbook from six months to 2 hours! While the cost per textbook was about $2,000, it was clear to us that technology would reduce cost dramatically within a decade or so.

    I was also engaged in the production of a series of academic journals with the help of the Xerox Corporation. This project highlighted the need to convert documents between systems. More directly, with support from Xerox, I undertook to develop an intelligent system to convert documents produced in simple text processing systems to more formal structured markup -- i.e. SGML. These projects led to an exploration of the standards for the development of documents. This in turn led with time to the exploration of standards for interconnection, interoperability and human computer interaction. Ultimately, this interest in standards took on a life of its own as I became involved with the national standards organizations. This led to a number of projects that tried to make sense out of developments in the way standards were developed. Because standards are complex group authored documents, it was only natural that some efforts would eventually be directed to how to assist in the development of standards.

    Just as documents had led to standards, so documents led to interfaces -- in two ways. First, documents themselves are interfaces to information, so it was possible to ask why computer interfaces couldn't be like document interfaces. More directly, the 1980's were a period of rapid evolution in the area of interface design. People were exploring patterns languages and this seemed to be an area worthy of examination as we were working to build good interfaces to document systems, because they are by their very nature intensely interactive system. Finally, we come to the issue of exactly what the nature of the interaction is between humans and machines. I see this most broadly as the augmentation issue -- how does the electronic system augment the human system. A review of all the early work by Doug Engelbart led to a conviction that while much had been said, little had been done since his pioneering work to build truly augmented systems.

    Accesses since Oct/26/2001: