Presented as a Tire-Tracks Diagram
Ben Shneiderman, University of Maryland
May 15, 2017, firstname.lastname@example.org
This paper describes processes by which Human-Computer Interaction (HCI) has become a widely successful research discipline and User interface/Experience Design (UXD) has become an influential profession. It presents visual evidence of disciplinary growth by way of simple charts and a tire-tracks diagram.
Within the HCI and UXD communities there is a great sense of pride of what has been accomplished in the past 35 years since the historically significant 1982 Gaithersburg conference on “Human Factors in Computing Systems”, which many see as the starting point for these topics. There were a few workshops earlier, but this conference drew a surprising crowd of 906 attendees, signaling strong interest in the topic and helping launch the ACM’s Special Interest Group in Computer-Human Interaction” (SIGCHI) (Shneiderman, 1986). The conference series was then taken over by SIGCHI, which has grown to be a large professional group, which in addition to its main annual conference, sponsors 35 smaller conferences on specialized topics. HCI research has also become an international activity with professional societies, academic programs, and industrial impact worldwide.
The trajectory of research begins with carefully controlled studies of individual users carrying out key tasks requiring perceptual, cognitive, and motor skills. The growth of interest in communication tasks required studying pairs and groups of users, which expanded to research on online communities, which transformed into social media experiences for billions of users. The older reductionist models of research, which remain valid and useful, might be characterized as micro-HCI, which focuses on narrowly-defined tasks, which can be studied in 1–3 hours. By contrast, macro-HCI focuses on richer aspects of user experiences, such as relationship-building, empathy, and trust, which require study over weeks and months. Admirably some HCI researchers see their scope as including economic development, community safety, peace studies, environmental sustainability, and energy conservation (Shneiderman et al., 2016).
There are thoughtful reviews of HCI’s history by Myers (1998) and Grudin (2017). There are also careful bibliometric and scientometric analyses by Barkhuus and Rode (2007), Bartneck and Hu (2009), and Henry et al. (2007). These are valuable for their insights, but I was eager to have more global measures of growth that might indicate broader impact.
A natural way to begin documenting the growth of a discipline is to look at the papers published. Fortunately, the heroic efforts of Gary Perlman give a clear picture of the dramatic growth of papers in conferences and journals (Fig 1). His HCI Bibliography website (www.hcibib.org) constitutes a definition of what is in the discipline and what is not. His definitions have been tested by time, through challenges by journal editors, questions by conference journal editors, and feedback from users. The data show remarkable growth from only 74 entries (papers in conferences and journals) published in 1980 to an astounding 11165 entries in 2014. This 150-fold growth shows the vibrant activity in an unquestionable way. By now there are approximately 25 journals and 100 conferences focusing on HCI research, while many other conferences and journals regularly include special issues, sections, or individual papers on HCI research.
Figure 1: Growth in the HCI Bibliography abstracts from 74 to 11165 entries during 1980 to 2014, mainly from conference and journal papers. (http://www.hcibib.org/pubdates.html)
Another way to track the growth of a discipline is through the use of the free publicly-accessible Google Ngram Viewer. This web-based tool allows users to submit words and phrases and get a graph that shows the frequency of those terms in English-language published books. A simple search on “usability, user interface, user experience, human-computer interaction”, shows the dramatic growth since 1982 (Fig 2). Since books are written for wider audiences than research articles, it is natural that user interface and usability are much more frequently used than human-computer interaction. Similarly, the in-discipline promoters of user experience as a newer term in the past decade have yet to produce the impact they wish in published books.
Figure 2: Growth in usage of terms (user interface, usability, user experience, and human-computer interaction) in Google’s Ngram Viewer that cover almost 20M English-language books during 1975 to 2005. User interface and usability have grown substantially, while user experience and human-computer interaction are less frequent in this broad collection of books. https://books.google.com/ngrams
A third way to assess the growth of HCI is through a broad news source such as the New York Times. Their frequency of mentions shows growth in documents that refer to the component terms (Fig. 3). While not as dramatic as the other sources, it still shows substantial increase.
Figure 3: Growth in usage of terms (user interface, usability, user experience, and human-computer interaction) in The New York Times from 1980 to 2012. Information provided from internal New York Times resources. The data for the past few years was incomplete and was removed.
Tire-Tracks Diagram for HCI
For more than 20 years, policy oriented information technology advocates in Washington, DC have been eagerly showing the “tire-tracks diagram” to Congressional staffers, funding agency directors, business leaders, journalists, and anyone who would listen (NRC, 1995, 2003, 2012). This chart shows tracks for university projects, industry research & development, and products that grow into billion and ten billion dollar markets. Arrows criss-cross the tracks showing the vital connections between these three tracks, such as a PhD student going to work in industry, a research paper leading to new products, or an industry visit to an academic research lab.
Sets of these tracks for Personal Computing, Internet & Web, Enterprise Systems, and other market segments give the impression of tire tracks. The take away message is that federal funding of early stage university projects, which are closely linked with industry research is the catalyst for much of the success of information technologies. The tire-tracks diagrams are a convincing way of showing the impact of federal funding and the efficacy of academic-industry connections in propelling new technologies to market success.
I spoke with developers of the 1995 tire-tracks diagrams who described their personal recollections of technology transfer by students working in industry, industrial researchers visiting academic centers, and key papers that influenced product development. Their informal approach was turned into a polished diagram that made a strong impression on those who studied it. The choice of research topics and markets is a subjective process that is meant to provide a description of how ideas travel and businesses grow. Adding information about unproductive research and failed products would be helpful to students and future entrepreneurs.
I followed their approach and expanded on Brad Myers’ tire-tracks diagram for the HCI field (1998) that emphasized concepts such as direct manipulation, the mouse, and hypertext. My goal was to highlight the ways early research at universities and industry laid the foundation for $1B and later $10B markets (Fig 4). Since no one has recorded the careers for every doctoral student, the visits between researchers and developers, or the readers of papers, we can never have the full story. It is legitimate to question the accuracy of each mark on the tire-tracks diagram and what is missing, but the overall story conveys a useful message about how research can lead to markets and how close contacts with markets can lead to higher-impact research. As more evidence is collected, future tire-tracks diagrams will show more complete and accurate evidence.
The tire-tracks diagram focuses on commercial outcomes, but there are many activities within an emerging discipline that influence these outcomes. Key activities are conferences, journals, degree programs, and funding programs. Then there are necessary changes to an emerging discipline as research methods evolve, as professional societies form, and as the curriculum shifts to include new methods and theories. A separate diagram could focus on the evolution of research methods, principles, guidelines, and theories.
Figure 4: The path to success for HCI and User interface/Experience Design is shown in the movement from university research to industry research that yields products with large markets.
The growth of HCI as a discipline and user interface/experience design as a professional field have been dramatic. Tracking the trajectory of disciplinary success and understanding the underlying components is a difficult job, but one that is worthwhile doing to understand how different fields grow or dissipate. Improved quantitative data would help, as would a clearer theory of what drives success or failure. A truly valuable theory would assess the role of individuals, universities, industry, and professional societies, as well as the alignment with broader societal movements that are tied to the discipline.
Acknowledgements: Thanks to Gary Perlman for help with HCIBIB data, Matt Ericson for obtaining the NY Times data, and Fan Du for preparing the charts. Thanks to Sara Bly, Jack Carroll, Clarisse Sieckenuis de Souza, Steven Feiner, Gerhard Fischer, Jon Froehlich, Joe Konstan, Loren Terveen, and Bernard Rous for constructive comments.
Barkhuus, L and Rode, J. (2007). From Mice to Men — 24 Years of Evaluation Proc. SIGCHI Conference on Human Factors in Computing Systems (CHI 2007), ACM Press, New York, DOI=10.1145/1240624.2180963
Bartneck, C. and Hu, J. (2009). Scientometric analysis of the CHI proceedings, Proc. SIGCHI Conference on Human Factors in Computing Systems (CHI 2009), ACM Press, New York, 699–708.
Grudin, Jonathan (2017), From Tool to Partner: The Evolution of Human-Computer Interaction, Synthesis Lectures on Human-Centered Informatics, Morgan & Claypool Publishers, January 2017, http://www.morganclaypool.com/doi/abs/10.2200/S00745ED1V01Y201612HCI035
Guha, Shion, Steinhardt, Stephanie, Ahmed, Syed Ishtiaque, and Lagoze, Carl (2013). Following bibliometric footprints: the ACM digital library and the evolution of computer science, Proc. 13th ACM/IEEE-CS Joint Conference on Digital Libraries, 139–142.
Henry, Nathalie, Goodell, Howard, Elmqvist, Niklas and Fekete, Jean-Daniel (2007). ’20 Years of Four HCI Conferences: A Visual Exploration’, International Journal of Human-Computer Interaction, 23:3, 239–285, DOI: 10.1080/10447310701702402
Myers, Brad A. (1998), A brief history of Human Computer Interaction technology, ACM interactions 5, 2 (March, 1998), 44–54. http://www.cs.cmu.edu/~amulet/papers/uihistoryinteractions.pdf
National Research Council’s (NRC’s) Computer Science and Telecommunications
Board (CSTB) (1995). Evolving the High Performance Computing and Communications Initiative to Support the Nation’s Information Infrastructure. The National Academies Press, Washington, DC. https://www.nap.edu/catalog/4948/evolving-the-high-performance-computing-and-communications-initiative-to-support-the-nations-information-infrastructure
National Research Council, Computer Science and Telecommunications Board; Division on Engineering and Physical Sciences; Innovation in Information Technology (2003). The National Academies Press, Washington, DC.
National Research Council, Committee on Depicting Innovation in Information Technology; Computer Science and Telecommunications Board; Division on Engineering and Physical Sciences; (2012). Continuing Innovation in Information Technology, The National Academies Press, Washington, DC.
Shneiderman, B., No members, no officers, no dues: A ten-year history of the Software Psychology Society, ACM SIGCHI Bulletin 18, 2 (October 1986). http://dl.acm.org/citation.cfm?doid=15683.15685
Shneiderman, B., The New ABCs of Research: Achieving Breakthrough Collaborations, Oxford University Press (April 2016).
Shneiderman, B., Plaisant, C., Cohen, M., Jacobs, S., and Elmqvist, N., Designing the User Interface: Strategies for Effective Human-Computer Interaction: Sixth Edition, Pearson (May 2016). http://www.cs.umd.edu/hcil/DTUI6
Ben Shneiderman is a Distinguished University Professor of Computer Science at the University of Maryland and a Member of the National Academy of Engineering.