Amberlight's Hierarchy HCI Briefings Library Amberlight Whitepapers

The History of Human-Computer Interaction

Human-Computer Interaction (HCI) developed as a field in the late 1970s and early 1980s in order to address the problems faced by the users of increasingly complex and powerful computer systems. HCI drew on research from communication theory, graphic and industrial design, linguistics, social sciences and cognitive psychology in order to gain insights about the goals, motivations, skills and abilities of system users. This knowledge allows designers to create systems that are technologically effective, but can also be operated by their intended users.

This represented a huge leap forward in the quality and usability of user interfaces. Previously, interface development had tended to be an afterthought, tagged on to the end of systems projects. The development of these interfaces was generally left in the hands of programmers who, while expert at operating the byzantine systems of the day, had little or no idea about the abilities of their users.

The inclusion of research knowledge from the cognitive sciences allowed some insights about human abilities to creep into design, which in turn led to the establishment of HCI as a discipline that could help inform development.

1940s - Ergonomics

Prior to the emergence of HCI, it had already been long acknowledged that issues of human psychology lie hidden at the heart of many design problems. During the Second World War, British psychologists conducted research into the attentional resources of radar operators. The researchers were able to make recommendations on how to optimise the design of radar monitoring stations in order to keep the operators focussed and allow them to distinguish between signals even when tired or preoccupied.

By the end of the 1940s, a considerable body of research knowledge about human performance, and in particular the psychological aspects of work, had been compiled. This led to the establishment of a formal discipline for the engineering of human work systems: ergonomics. Ergonomists are concerned with the ‘fit’ between the human and the job. The better the design of the fit, the safer and more efficient the work carried out will be.

1950-60s – Birth of computing

Within a decade, computing technology became available which would transform human work. Although large, expensive and relatively rare, by the end of the 1950s computers had begun to revolutionise the way which people worked. In 1960, the Sabre airlines reservation system, the first computerized transaction-processing system was launched. In the same year, Jim Slagel at MIT produced a program which could “get an A on an MIT calculus test.”

The opportunities created by the new computing power seemed endless. Unfortunately, systems like Sabre were enormous, complicated affairs that most people could not begin to comprehend. The failures of computers to ‘fit’ into people’s work caused J.R. Licklider (1960) to postulate the concept of “man-computer symbiosis.” A coupling of human brains and computing machines to revolutionize information handling. This eventually became known within the ergonomics field as Man-Machine Interface (MMI) and the term ‘MMI design’ described applying known ergonomics principles to interface design to achieve the best fit between man and computer.

During the 1960s, the seeds were also sown for the personal computing and internet revolutions of the 1980s and 90s. The growing power of computers led researchers to theorise about future ways of interacting with them. Several computing concepts that are common today were born: Doug Englbart (1962) invented the mouse, and Ted Nelson (1960) coined the term ‘Hypertext’. In 1969 the US Department of Defence commissioned four leading universities to conduct research into networking. This was ARPANET, the precursor to today’s Internet.

1970s – Rise of the PC

During the 1970s computing power increased and became less expensive. For the first time, computers began to appear on people’s desks. With the advent of the personal computer, IBM surprised many people in the software industry by ‘unbundling’ software – removing free, proprietary software from their hardware and allowing non-proprietary programs to work on their PCs.

While software companies enjoyed the benefit of unbundling, they also started to fall behind in terms of software support. Smaller companies began to learn that the cost of supplying support for their own systems was extremely high. The challenge for the software companies was to design interfaces that allowed customers to use their systems efficiently, without having to contact their technical support departments.

1980s - GUI

By the late 70s, with personal computing playing an ever larger role in day-to-day work, many researchers were exploring alternatives to the command-line interface. The Xerox Star, launched in 1981, included the rudiments of what Ben Shneiderman (1982) was soon to term ‘direct manipulation’. Graphically-based interaction offered several advantages over command-line interfaces: incremental action and rapid feedback from the interface, the ability to reverse actions, and the use of actions instead of ‘learning’ a new language.

The Xerox Star used the desktop metaphor, introducing the now standard WIMP set of tools (Windows, Icons, Menus, Pointers). It was the first PC system based on what would become universal usability engineering principles: prototyping and analysis, testing with users and iterative refinement. Unfortunately, Star was a commercial flop, mainly because the main customers for PCs at the time were business people, and the Star lacked the key functionality of a spreadsheet package. Thus, Star failed one very important principle which usability professionals now accept: that it is not enough for a product to be usable – it must also be useful.

Throughout the decade, the new Graphical User Interface (GUI) paradigm was explored by several companies (Lisa, VisiCalc –1983, Apple Macintosh - 1984, Microsoft Windows, Commodore Amiga -1985). However, the lack of software for many of these, the sheer variety of competing platforms and lack of the computing power to deliver on potential kept users uninterested in the GUI for most the 1980s. It was not until the launch of Microsoft Windows 3.0 (with improved aesthetic design, including clearly ‘clickable’ buttons), allied to increases in home PC capacity, that the GUI found large scale acceptance.

Many of the GUI innovations had their roots in academic research from the 1960s and 70s. By the 1980s, this research had become increasingly distinct from classical ergonomics, generally ignoring the physical worksystem, and instead focussing on the cognitive aspects of interaction between human and computer. The term ‘Human-Computer Interaction’ or HCI was used to describe the field, replacing the outmoded (and politically incorrect) ‘Man-Machine Interface’.

1990s – Internet and collaborative work

By the mid-90s, the proliferation of Windows demanded that software developers adopt the GUI convention. Whilst much of early GUI work was grounded in psychological research, the new interfaces just superficially aped the Windows style. These products often broke standards and allowed the interface to behave in inconsistent and un-intuitive ways.

The growth in Home-PC ownership and the Internet explosion created an enormous demand for HCI skills within development teams. Being able to connect PCs remotely across networks allowed users to work together on projects and communicate via chat rooms and email. The social aspects of computer use had become crucial to successful systems. Designers of products were no longer modelling single user-computer interactions, but the interactions of many humans via computers. Most designers fatally ignored these ‘community elements’, with the result that while e-mail has been a prominent success, other types of groupware are still not as widely used .

With its foundations in psychology, HCI (now often referred to as ‘usability engineering’) allowed experts to model the complex social interactions surrounding the new computing. These models allowed researchers such as Jakob Nielsen (2000) to develop best-practice guidelines for developing online interfaces.

At the same time, HCI was undergoing a revolution of its own. Traditionally grounded in models of processing from cognitive psychology, HCI was now drawing on fields as diverse as ethnography, linguistics and communications theory, as well as arts and humanities subjects, in order to increase insights about users working with computers in the Internet age.

2000s – Mobile computing and beyond

Of course, the Internet boom was not to last. Too many of the people involved felt they knew what their users needed or wanted, and never thought to employ user-centred methods to find out if they were right. Poor business models allowed many companies to throw vast amounts of money at designing products that no-one could use and that no-one wanted.

However, the idea that this signalled the end of services on the Internet was premature, and those companies that were able to provide useful and usable services weathered the storm. Usability and HCI methods had proved their strength as mitigators against risk in design. In addition to developing new Web-based services, HCI approaches are being employed in the design of systems for the new paradigm. Since the late 90s, mobile devices have altered the way in which we interact with computers and each other. Mobile phones, Personal Digital Appliances (PDAs) and wireless networking have given rise to the concept of ubiquitous computing: a world in which technology is everywhere but recedes quietly into the background. People are no longer users of single, isolated devices, but occupants of computationally-rich social environments. It is imperative that systems that have such a fundamental impact on the way in which we operate are not allowed to develop in the ad-hoc manner that has characterised the development of earlier technologies. Human-centric systems require human-centred design, and that is the challenge for HCI in the decade to come.

Further Reading

The standard introductory textbooks for HCI both contain histories of the field:
Preece, J. Rogers, Y. and Sharp, H. (2002) Interaction Design, John Wiley & Sons, NY.

Dix, A. Finlay, J. Abowd, G. and Beale, R. (2003) Human-computer Interaction (3nd Edition). Prentice Hall.

An overview of theory development in HCI, offers a good review of 1980s to present:
Rogers, Y. (2004) New theoretical approaches for HCI.

A general history of computing:
Goldstine, H.H. (1993) The Computer: from Pascal to von Neumann. Princeton, NJ: Priceton University Press.