30 years of HCI, a Personal Reflection

Alan Dix

Talis, Birmingham, UK
and School of Computer Science, University of Birmingham

Prepared for special 30th anniversary issue of Interfaces (originally BCS HCI Group Newsletter), which unfortunately was never prodcued :-(


This was written as a contribution to the 30th anniversary issue of Interfaces (originally the BCS HCI Group Newsletter).  Sadly the special edition was never published.  In it I reflect on my own 30 years in HCI, not least the importance of the Alvey Programme, which transformed UK computing research, particularly in HCI.

Keywords: History of HCI, Alvey Programme, British HCI

Like Interactions itself, this year is my 30th year in HCI.  Although I wasn't a member at the very beginning, my first ever paper [DR85] was published at the first British HCI conference (although sadly I didn't attend) and the BHCI series was both formative for my own academic and professional growth and the venue for some of my own favourite papers (including the first PIE paper in HCI'85).

I was fortunate to come into HCI on an Alvey funded project at York in 1984.  This was fortunate for three reasons.

First this was the birth of HCI.   Although the early roots of HCI can be traced back to Brian Shackel's " Ergonomics for a computer" [Sh59] published in 1959 (yes the very first HCI paper was British!), it was only in the early 1980s with the growth of PCs and desktop computing that it began to grow as an area.  The first CHI conference was in 1983, the first Interact in 1984 and first BHCI in 1985.  These where definitely the heady days of HCI when everything was new and everything was possible. 

Second York was one of the main centres for HCI in the UK during those years linking computer science and psychology departments.  Key figures at the time included Andrew Monk and Michael Harrison who have recently retired and Harold Thimbleby not that many years after his own PhD at QMC, and the second British HCI Conference was held there in 1985.  In addition many figures in HCI were there in subsequent years as students or postdocs including Peter Wright, Chris Johnson, and Andy Dearden and of course my HCI textbook co-authors Janet Finlay, Gregory Abowd and Russell Beale.

Third this was an Alvey funded project.   It is impossible to overestimate the importance of the Alvey Programme for UK computer science research in general and HCI in particular.   Alvey was the UK response to the Japanese Fifth Generation Computer project, a massive government funded R&D programme focused particularly on AI.  Alvey was more modest but still huge, 350 million pounds over five years, worth about a billion today … and that is pre-FEC!   However it was not just the volume of Alvey spending that was significant, it was the strategic focus.

Partly of this was about the importance of joint academic/industry R&D.  The project I was employed on was on the theoretical/academic end of those funded (formal methods in HCI), but It still had an industrial ‘uncle’ who visited and advised periodically.  Some years after its end, Brian Oakley, the director of the programme, lamented the way government had not capitalised on this impetus to the UK IT industry innovation [Me90].  This said the links forged and expectation of connections between university and industry had a long lasting effect.

The problematic interplay of theory and practice, professional and academic, are still very much alive in HCI, but more critical for the development of HCI in the UK, was the strategic choice of topic areas in Alvey.  Alvey had four strands:  Software Engineering, Intelligent Knowledge Based Systems, Advanced Microelectronics (VLSI) and Man Machine Interaction.

This last strand, which sounds almost quaintly gendered to modern ears, effectively what is now HCI, was there not because it was an emerging area within computing academia, but because industry identified it as important.  If there is one thing academics appreciate more than their intellectual pride, and value more than publications, it is funding.  Although the respect was grudging and partial, it gave HCI a place in computer science departments in the UK that it still struggles to achieve over much of Europe and certainly ahead of the US where for many years HCI researchers in CS effectively had to pretend they were doing something else (software engineering, artificial intelligence, etc.)

The UK is still a world leader in HCI research, and in the case of more theoretical HCI probably the word leader, this is in no small part because of Alvey.

Another joy of these early years was the substantial number of day workshops on all sorts of sub-topics.  Some were organised by the British HCI Group itself, some were IEE Colloquia, and some by the CSCW Programme, which ran a few years after Alvey finished.  These national workshops were particularly important for young researchers, and particularly so for those in institutions without a large HCI presence.  They were also a great way to spread ideas and get feedback.  Indeed, one of my top 20 most cited publications [Di94] was a chapter that came out of a talk at one such event, and another paper written informally as the write-up of a talk at another event was critical in inspiring a lab in Switzerland.

Although less regular in recent years, in those early years I hardly missed a conference, again three of my top 20 papers are in BHCI.  In those days BHCI (and Interact) had a far wider conception of HCI than CHI, so there was a trans-disciplinary vitality that grew much more slowly across the water.  And still, despite all the glitz and high-budget demos at bigger international conferences, there is still little to beat the friendliness, fun and intellectual buzz of BHCI!

So 30 years on we have seen the rise of interactive visualisation (driven by affordable computational power) and CSCW (driven by growing ubiquity of networks) in the early 1990s, the HCI of the web (driven by the web!) in the mid to late 1990s, ubiquitous computing and mobile HCI in the noughties (driven by increasing miniaturisation of sensors and computation).  Now I guess social networks (Facebook), touch (iPad), gesture (Kinect) and most recently AR (Google glass) would be trending.
Although labelling HCI phases by available technologies is a little simplistic, it is not so far from the truth, the lesson from recent bibliographic analysis suggest that if anything the trends are just that, fads that come and go [PM14, LG14].  According to these statistical analysis, and unlike established disciplines, HCI appears to have no core that remains through the years. 

It is not clear whether this is shared across computing, or is an exclusively HCI phenomenon, nor whether this is good or bad: a sign of vitality and innovation or lack of foundations. Personally I do find this worrying and have argued previously that we do need to think about what makes a sound and vigorous discipline.

Certainly in the early days there was an optimism about building foundational understanding; I think about Norman and Draper’s edited colletion UCSD exploring the underpinnings of direct manipulation was and user-centred design, Card, Moran and Newell’s ‘model human processor’, and more social analysis of Lucy Suchman.  Is HCI always going to be driven by the winds of external technological change, or can we use this to build long-lasting knowledge?

If there are to be such deeper underpinnings, then the British HCI community is just where they are likely to be formed with a long-standing and on-going theoretical strand that has persisted despite it not always being the current fashion.  Here I think of John Long’s intellectual legacy, Thimbleby and Blandford’s work in CHI-MED, Benford and Rodden’s conceptualisations of performance and pervasive experiences, Gaver’s explorations at the nexus of design and technology, the recent Interaction Science spotlight at CHI led by Howes [HC14], and many more past and present.  There are also those who seize upcoming technology, but in ways that use it to push forward our conceptualisation of interaction, Sriram Subramanian ‘s lab at Bristol is one example as is the long-standing work of MRL in Nottingham.

So where are we now and what of the future?

One of the practitioner trends over the last 5-10 years has been a shift from usability to user experience.  This is partly because basic usability is seen as ‘solved’, partly because it is the wow of experience that sells, as Apple demonstrate in its quintessence, yet we are also seeing an increasing number of major products that are slick, beautiful, and exciting but barely usable: scrolling areas inside scrolling areas, inside scrolling areas, a tap and hope for the best style of interaction, and MacOS that become harder to use with each release.  Is this simply the inevitable path of commoditisation, or will there be a kick-back?  Can we have both? Can we both add pleasure and reduce pain – solid, dependable, interfaces, that are still a joy to use?  In principle, the relative close proximity of diverse groups in the UK should make this the perfect place to answer this kind of question.

Social, political and developmental aspects of technology are also becoming increasingly important, and again an area where we have significant UK leaders, for example Andy Dearden’s long-term work in ICT4D or Matt Jones’ group in Swansea.  The people focus in HCI makes us, in a way, the conscience of computing, maybe tis is a mantel we need to grasp more actively.

Finally, there is the growth of maker-culture with its promise of democratising production and indefinite flexibility.  The biannual Tiree Tech Wave that I organise is partly about a different kind of ‘meeting’, but also partly an active exploration of this area.  HCI has always had a strand of research through doing, and maker culture is all about the close connections of people and technology, but most makers would not recognise the things they do as in anyway connected to HCI.

To some extent this does not matter, I’m happy to see a diverse range of designers, hackers, techies and artists gather, do things together, and perhaps also learn a little of the difference of technology at the edge of the grid.  However, there are substantial HCI issues that will arise as this moves from hobbyist to mainstream, many shared with those of end-user programming.  If everyone becomes a designer, what are the interfaces that enable them to design, and how do we ‘build in’ usability into the outputs without everyone doing a 3 year HCI degree? 

In summary, British HCI has a long and successful past dating back to the late 1950s, and is the home of many world leading research groups today.  The broadening of areas, cheap international flights and general business means we are not so much a ‘community’ now as we were 30 years ago.  So maybe we need to look to ways to strengthen nationally as well as grow internationally, especially as there seem to be several areas of growing importance that British HCI is well placed to lead.

... and if you would like to discuss any of these issues further, I’d love to see you out on Tiree at a Tech Wave.

References

  1. [DR85]  A. J. Dix and C. Runciman (1985).  Abstract models of interactive systems. People and Computers: Designing the Interface, Ed. P. J. &. S. Cook. Cambridge University Press. pp. 13-22. http://alandix.com/academic/papers/PIE85/PIE-paper.html

  2. [Di94]  A. Dix (1994).  Computer-supported cooperative work - a framework.  In Design Issues in CSCW Eds. D. Rosenburg and C. Hutchison. Springer Verlag. 23-37.  http://alandix.com/academic/papers/cscwframework94/

  3. [HC14] A. Howes, B. Cowan, S. Payne, P. Cairns, C. Janssen, A. Cox, A. Hornof and P. Pirolli (2014). Interaction Science Spotlight. CHI 2014, ACM.  http://chi2014.acm.org/spotlights/CHI-Science-proposal-final.pdf

  4. [LG14]  Liu, Y., Goncalves, J., Ferreira, D., Xiao, B., Hosio, S., & Kostakos, V. (2014). CHI 1994-2013: Mapping two decades of intellectual progress through co-word analysis. Proc. CHI, Toronto, Canada, 3553-3562. DOI: 10.1145/2556288.2556969

  5. [Me90]  Angeli Mehta.  Ailing after Alvey: The Alvey programme was Britain's big chance to compete in information technology - Brian Oakley, a former director of Alvey, reflects on what went wrong. New Scientist 1724 (7th July 1990) https://www.newscientist.com/article/mg12717242-300-ailing-after-alvey-the-alvey-programme-was-britains-big-chance-to-compete-in-information-technology-brian-oakley-a-former-director-of-alvey-reflects-on-what-went-wrong/

  6. [PM14] Stefano Padilla, Thomas S. Methven, David W. Corne, and Mike J. Chantler. 2014. Hot topics in CHI: trend maps for visualising research. In CHI '14 Extended Abstracts on Human Factors in Computing Systems (CHI EA '14). ACM, New York, NY, USA, 815-824. DOI=10.1145/2559206.2578867

  7. [Sh59]  B. Shackel, Ergonomics for a computer, Design 120 (1959) 36–39. http://www.hcibook.com/e3-images/online/not-all-about-technology/shackel-1959.pdf

 

 

 


1983 IT film on Alvey programme
(from Chilton Computing)

 


http://www.hcibook.com/alan/papers/thirty-years-of-HCI-2014/

Alan Dix 2/4/2016