Imperial College London

Emeritus ProfessorRobertSpence

Faculty of EngineeringDepartment of Electrical and Electronic Engineering

Senior Research Investigator
 
 
 
//

Contact

 

+44 (0)20 7594 6259r.spence Website

 
 
//

Assistant

 

Mrs Wiesia Hsissen +44 (0)20 7594 6261

 
//

Location

 

912Electrical EngineeringSouth Kensington Campus

//

Summary

 

Overview

Expertise and Research Interests

My early (late ‘50s) research interest was in the design and theory of  electronic circuits, particularly oscillators and the way in which one designs a circuit to take account of component tolerances.  This field of interest continued until the late ‘90s.  In parallel, I became concerned with the way in which computers might help the design process, especially through interactive-graphic interfaces that allow the designer to think, and externalise those thoughts, visually.  The two fields first came together very effectively in my research, begun in 1968, into the interactive-graphic support for the design of circuits.   A working system incorporating many early innovations in  human-computer interaction, and called MINNIE, was developed, and was arguably the first such working system.   Along the way my colleagues implemented the first on-screen calculator! In 1983, following the emergence of affordable high-resolution graphics, powerful computers and inexpensive memory, the commercial relevance of MINNIE was clear.  It was developed to commercial software standards and sold worldwide by a company of which I was chairman and a founding director.

My research since 1968 has developed along two lines which, as in the case of MINNIE, often interacted beneficially.  In the case of circuit theory I collaborated with coauthors to publish Tellegen’s Theorem and Electrical Networks (MIT Press), regarded as something of a classic and soon translated into Russian and Chinese.  Later, I developed algorithms for the efficient computation of the effect of component change, whether intentional or parasitic.  The discovery of ways in which circuits can be redesigned to ameliorate the consequences(e.g., low manufacturing yield) of tolerances led to the publication of Tolerance Design of Electronic Circuits (Addison-Wesley), later translated into Japanese.  Then, in the late ‘80s, my colleagues and I explored the potential of combining the abilities of computers and engineering designers by implementing a system, CoCo supporting the human guidance of automated design.

My work in the field of Human-Computer Interaction (HCI), on which I now focus, began – as already explained – with innovations   associated with interactive-graphic engineering design.  In 1973 my colleagues implemented the first on-screen electronic calculator.  Some innovations were associated with the encoding of data in visual form, a field we now refer to as Information Visualization and which is now my principal research interest.

In 1980 my colleague Mark Apperley and I invented the Bifocal Display, a simple way of allowing a user to see not only detail but context as well: it was the first proposal of what is now known as Focus+Context, presentation, a topic which has assumed considerable importance in the field of HCI.   In 1994 my colleagues and I invented the Attribute Explorer, an interface supporting the choice of one object (e.g., car) from many on the basis of its attributes (e.g., HP, MPG), as well as the interpretation of complex data.  Soon – and again drawing on my interest in engineering design - this concept was extended to the Influence Explorer which allows engineering designers to ‘think visually’ when selecting component values to satisfy the requirements of a customer.

During the last five years my research has focused on the digital equivalent of rapidly riffling the pages of an unfamiliar book to ‘see what’s there”.  This activity, known as Rapid Serial Visual Presentation (RSVP) offers an effective technique for locating an image of interest and poses challenging questions, some of which draw upon fundamental properties of the human visual system.

Other Expertise

My interest in the visual encoding and presentation of data, and the way in which human beings scan and interpret the result, has found an unexpected audience in the art and design community whose problems are a refreshing change from engineering design.

In 1994 I became – albeit temporarily - a film producer.  Attendance at two high-level workshops on the future of engineering design led to disappointment at the lack of any vision.   So, with EPSRC support, I individually interviewed twelve eminent engineering designers from a variety of disciplines and different countries to elicit their visions of engineering design 25 years into the future (2020AD).   Those visions were the ingredients from which I created a film showing a dinner party taking place in the year 2020AD in which leading
designers of that year discuss their approach to design.  The resulting film, in which my Hitchcockian appearance as the butler is suitable muted, is called Translations and is, in fact, less about engineering design than about the way in which human beings and computers can beneficially interact.

Industrial relevance

  • Methods of Tolerance design support the automated redesign of circuits and other products to exhibit higher manufacturing yield.
  • The Bifocal Display concept supports the design of interfaces where smooth continuity between detail and context is crucial
  • The Attribute and Influence Explorers support the effective exploration and interpretation of a wide range of types of data.
  • Developments of the Rapid Serial Visual Presentation of images have wide relevance to the search for a target image, whether it is a page in a report, a painting by Picasso or the illustration of a product in an online catalogue.