Return to People

Bill Seaman’s Paper Proposal

Insight Engine


This research seeks to work toward the digital authorship of a tool to empower insight production, distributed interdisciplinary team-based research, and to potentially enable bisociational processes as discussed by Arthur Koestler in The Act of Creation. The goal of the 1st year of research is to create an interactive system to enable intelligent juxtaposition of relevant media elements via focused interaction, dynamic computational functionality, and intellectual “seeding” of the system.

The system seeks to reverse engineer some of the processes that we use as researchers. It then works toward a human / machine symbiosis where the machine presents an interface to many different researchers work. It does so in a way that enables the human user to playfully explore many different areas of research that may or may not be relevant to their current work in a novel interactive manner. The system presents a “word swirl” in 3D for each different researcher. These can be called up and displayed in the interface or put away. These word swirls have buzz words or titles (that you will supply for your own papers/ or I will add if need be) at the top of the hierarchy (one can also look deeper in the hierarchy with multiple finger touches) and even read an entire paper if it is of interest.

Often new knowledge arises in the space between fields —in interstitial zones of knowledge. If one chooses one “buzz word” or Paper Title from “your” word swirl and one from another researcher that looks to be of interest, the system will seek to find the most relevant examples in the database [by making both a statistical and semantic comparison] (and eventually searching the internet, in the next iteration of the system) and provide those papers for you as a new word swirl of the most relevant juxtapositions.

This puts your research in proximity to someone who otherwise might not find it or know of it. Alternately, if you play with the system it may provide a juxtaposition that is relevant to you and your research in a new way… in an emergent manner… thus the system might provide a historical instance, or newly published paper, etc. that suggests a moment of insight for your research.

The system seeks to be a learning system, where as you use it, you generate new word swirls of papers or media, that we hope to be relevant to you, in a playful iterative manner, or you can throw things away that are not relevant. The goal is in generating “intelligent research juxtapositions” that may arise through the use of the system – either for you or for a person who might learn from your work. This seems to me to be a situation of mutual intellectual gain between differing researchers…


  1. Making visible the invisible, I believe. Michael Hohl had a conference on this theme a couple of years back, and you might find the work presented there interesting.

    I think this is a very interesting challenge. What I always worry about, however, is the implicit assumption that words mean something and that that something is shared by all and remains unchanged over time.

    What I’m not sure about is how, and in what sense, insight enters the picture.

  2. This sounds a little like a frontend to visualise data and make it more accessible for databases from Mendeley or What look like websites that allow academic authors to share their thoughts and papers is actually part of larger research projects to map out knowledge domains and find relations among them.

Leave a Reply