INTERACT 2009 – Research & Practice?
By Gerd Waloszek, SAP User Experience, SAP AG – December 10, 2009
INTERACT 2009, the 12th IFIP TC13 International Conference on Human-Computer Interaction, took place in Uppsala, Sweden. Uppsala is home to Sweden's oldest university, and its 200,000-strong population makes it the country's fourth largest city. This conference visit was special for me: For the first time, I not only attended a conference but also – together with my SAP colleague Ulrich Kreichgauer – submitted a short paper that Ulrich presented.
The INTERACT conference has a more European "feel" (including British English as the conference language) than, for example, the CHI conference, and it is one of the smaller international HCI conferences. This year, about 450 people attended. The largest contingent – not surprisingly – came from Sweden, followed by participants from Germany (49) and the United Kingdom (45). As I already wrote in my INTERACT 2005 report, this conference has a chiefly academic touch. I encountered this bias when our short paper was judged by some reviewers as "not scientific enough", and I found it confirmed when the audience members were addressed as "researchers" on several occasions. The conference hosts do, however, seem to have identified this bias as an issue and have attempted to counteract it: The conference motto was "Research & Practice", and an "industry day" was introduced as a new ingredient. But was that enough to justify the motto? I will return to this question at the end of my report. For the time being, I will leave a question mark after the motto (as in the title of my report).
Figure 1: Logo of the INTERACT 2009 conference
As in my previous conference reports, I will not attempt to cover the whole breadth of the conference. There were up to seven sessions in parallel, so that it looked as if choosing a conference thread to attend would be very painful. To my surprise, however, the bigger pain for me was trying to find any paper sessions that interested me as a practitioner. Once again, I was confronted with the academic orientation of the conference. As there were only two panels at the conference, I could not resort to my usual conference strategy of mostly attending panels. Luckily, the "industry day" came to my rescue on the second conference day. I attended all the presentations in this special thread, except for the promising closing keynote, because it clashed with the presentation of our paper in a parallel paper session.
Keynotes
The INTERACT 2009 conference featured four keynote speakers. Kristina Höök, also known as Kia, (University of Stockholm, Sweden, and Mobile Life Center) held the opening plenary keynote presentation, "Mobile Life – Body and Interaction"; Nicklas Lundblad (Deputy CEO of the Stockholm Chamber of Commerce, Sweden; formerly Google) held the Thursday plenary keynote presentation, "Lies, Damn Lies and Privacy", which kicked off the second conference day; and Liam Bannon (University of Limerick, Ireland) presented the closing plenary keynote, "Towards Human-Centred Design". Sofia Svanteson (Ocean Observations, Stockholm, Sweden) closed the industry day with a keynote about design excellence, entitled "Excellent Design – It's not About the Process, But the People, the Culture and the Leadership", which I regrettably had to miss, because our short paper presentation took place at the same time.
 |
 |
 |
 |
Figure 2: The invited speakers: Kristina Höök (opening plenary keynote), Nicklas Lundblad (Thursday plenary keynote), Liam Bannon (closing plenary keynote), and Sofia Svanteson (industry day closing keynote)
In her keynote, Kristina Höök emphasized that mobile technologies, which she has been exploring for over a decade in the lab, "have now moved out into the world – into the wild" and she asked what this change means for the HCI field. Höök also identified a second trend, namely the changing role of consumers, who are increasingly creating content on their own. In a third vein, she presented her vision of a "ludic society," in which "work mixes with leisure, private with public – a society where enjoyment, experience and play are adopted into all aspects of life."
Nicklas Lundblad, who recently left Google, where he studied privacy technologies, looked at IT privacy issues and how they are currently dealt with, namely as an add-on that is designed into systems. He came up with a surprising solution to preserving privacy: lying. He demonstrated that people lie to protect their privacy – not only on partnership Websites – and proposed to further pursue and investigate this approach: to move from privacy enhancing technologies to "lying-enhancing technologies" – in short from PETs to LETs.
Liam Bannon from the University of Limerick, Ireland, held the closing keynote. He seemed to hold it mostly for himself, running around and not looking at the audience for long periods of time. As a blogger stated, he also moved freely around between various topics. However, in retrospect, his talk was not as chaotic as it may have seemed at the time. All in all, his keynote was a plea for human agency to fight concepts in which all-knowing machines take control, such as ambient intelligence (as exemplified by UbiComp (ubiquitous computing) and the "Internet of Things"), focus on human values, and to re-think the human-machine/people-technology relationship as well as the role of HCI.
See photos in the appendix.
Industry Day
On the second conference day, the "industry day" was held as a new ingredient in INTERACT. It started right after the keynote as a parallel conference thread, and I stayed there until the closing keynote for the conference day started, as I already mentioned above. The presenters came mostly from Scandinavian design and technology companies, though some were independent consultants. Their topics ranged from showing what a company does in the UX area, to ISO standardization, Agile projects, and other case studies. With respect to Agile, once again, both commonalities and differences emerged, showing UX people often holding the position of a "project manager" who takes care of the overall plan. A sometimes overcrowded room revealed that the conference attendees were very interested in sharing experiences from the industry.
See photos in the appendix.
Panels
There were only two panels at INTERACT 2009, and I attended both of them.
Responsive Interfaces: Mobility, Emotion and Universality
 |
 . |
 |
 |
 |
Figure 3: Participants on the Responsive Interfaces panel – Mark Chignell (moderator), Naotsune Hosono, Deborah Fels, Danielle Lottridge, and John Waterworth
This panel set out to derive a list of ten guidelines for designing future collaborative systems. The starting point was the insight that the current model of desktop interaction will be supplanted by a different style of interaction in the future, which takes care of mobility, emotionality, universality, as well as changing perceptual and cognitive requirements. The five speakers presented different perspectives: Mark Chignell argued that future interaction will be dominantly mobile ("changing the game"), while his Ph.D. student Danielle Lottridge talked about emotional interfaces. Naotsune Hosono is involved in universal design at Oki and talked about telework systems that he had designed for disabled people in social isolation, while Deborah Fels discussed the principles of universal design in general and demonstrated, among other things, how they can be applied to sign language-based interfaces. Finally, John Waterford argued that the future will bring a blending of physical and mediated reality.
This was a rather lively panel, perhaps because it took place in a smaller room. Not surprisingly, it could not achieve its goal in the limited amount of time. One reason may have been that an extended discussion began after Mark Chignell mentioned the use of mobile phones in cars as an example of "things to avoid.". To the surprise of most of the audience, Kristina (Kia) Höök interrupted him and insisted that drivers in Sweden are allowed to use mobile phones in cars and that Swedish people are educated enough to know when they should not do so. Once again, Kia had managed to provoke (which is she likes to do, as she mentioned more than once during the conference)...
See photos in the appendix.
Demarcating User Experience
 |
 |
 |
 |
 |
 |
Figure 4: Participants on the User Experience panel – Virpi Roto (moderator), Nigel Bevan, Jettie Hoonhout, Ilpo Koskinen, Gitte Lindgard, and Kristina Höök
The participants on this panel, introduced by Virpi Roto from Nokia, set off to define, or "demarcate", the term "user experience" or – as Virpi Roto also put it – to narrow down the perspectives on user experience. Luckily for Virpi Roto and Nigel Bevan, they could more or less reuse their presentations from the industry day...
Nigel Bevan pointed to the fact that ISO 9241-11 more or less reflects a company or organizational perspective by emphasizing the business benefits of workers who are effective, efficient and satisfied. The first two are objective criteria, while the latter is a subjective one. In this context, satisfaction is seen as "freedom from discomfort and a positive attitude towards the use of a product." Bevan showed that in the new ISO/IEC 25010 standard, satisfaction comprises all aspects of user experience:
- Likeability (pragmatic user goals) -> includes the user's perception of effectiveness and efficiency, in contrast with organizational goals
- Pleasure (hedonic goals; following Hassenzahl) -> includes stimulation, identification evocation (self-maintenance, (pleasant) memories)
- Trust (satisfaction with security)
- Comfort (physical satisfaction)
Jettie Hoonhout from Phillips represented an "industry" point of view. She showed examples of bad and good experience, and remarked that all of them were experiences. She added: "If you love them, you'll forgive them, even though they are hard to understand." The goal is not always to make things easier: Sometimes the goal is to make a product look cooler. While she started her presentation with the motto "Life is experience!", Kristina Höök (Kia) asked: "User experience: a problem to be solved or the meaning of life?" She emphasized the role of aesthetic experience and the interactive nature of experience, and she listed the following qualities of experience:
- Co-created
- Aesthetics
- Zillions of possible experiences
- Nice is boring
Designer Ilpo Koskinen from Helsinki, Finland, focused on his new concept of "co-experience," which is actually not too far away from Kristina Höök's point of view. Finally, Gitte Lindgaard, coming from a university, revealed a fairly familiar view of user experience: "In UX, there is an emphasis on the person/people interacting with the product or service, but without compromising the traditional utilitarian or more recent usability values." I do not think that Kia would subscribe to this definition.
In the end, Virpi Roto was left fairly alone with her attempts at demarcating user experience, with Nigel Bevan coming closest to her attempts. The panelists did at least agree that the "ease-of-use" concept behind usability is too narrow and that user experience is a broader concept that includes subjective criteria. Often, however, commercial success seems to be more important than a good user experience.
See photos in the appendix.
Further Remarks
Papers
As I mentioned earlier, my SAP colleague Ulrich Kreichgauer presented a short paper about our perceived performance (PeP) project at the conference (for details about the PeP project, see The PeP Project: Evaluating the Responsiveness of SAP Applications from a User-Centered Perspective).
Figure 5: Ulrich Kreichgauer presenting the PeP short paper about perceived performance
Branding and usability. In a paper session in the "user experience" track, I was confronted with two conflicting papers: One paper claimed that perceived brand value has an effect on the quality of the user experience, while a second paper stated an influence in the opposite direction, namely that a good or bad user experience (on the Web) has an impact on perceived brand value. After the first paper, I considered switching to the marketing department, but after the second I calmed down again...
Improving window switching interfaces. In a session about novel user interfaces and interaction techniques, I came across a paper dealing with window switching interfaces. Despite window menus, the Windows task bar, and the Apple dock, there still seems to be a demand for more efficient solutions to dealing with multiple windows. I found out that there was a lot of interest in these issues at SAP recently. The authors stated that "switching between windows on a computer is a frequent activity, but current switching mechanisms make it difficult to find the correct window." In order to determine actual window switching needs, they observed 9 users for periods of between 2 weeks and 9 months while they were using logging software. From their observations, they derived design principles for window switching interfaces, two of which they investigated in this study in two experiments: spatial stability and size morphing. They found that a spatially stable arrangement of windows fared much better than a random assignment, because a random assignment does not support expertise development (recency assignment performed better than random assignment, but also worse than a stable assignment). They also found that although (gradual) size morphing affected stability, it fared no worse than stability.
Posters
I would also like to briefly mention two posters. Students from the Technical University of Dresden and colleagues from SAP Research CEC Dresden (Germany) presented an approach designed to help users exploit the "Internet of Services" by creating complex interactive service-based applications. Users can annotate Internet services and thus add meta-information to existing service descriptions related to user interface generation, service dependencies, and service composition. Services can be composed visually. Users need not write any code – it is generated completely from the description. Students from the University of Århus, Denmark, presented a new way of creating an overview of administrative procedures using timelines. Users manipulate the information given on the timeline by interacting directly with it. As an example application, the students developed and evaluated a prototype supporting administrative procedures related to parental leave.
Presentations
At INTERACT 2009, I learned that "hip" presentations have to have a dark brown background with white type. At least three of the keynote and panel presentations adopted this color scheme (Kristina Höök, Liam Bannon, Ilpo Koskinen). But I also learned that you can be even more "hip" by using "prezis," as Nicklas Lundblad did in his keynote presentation. Prezis are Web-based presentations (they require a yearly fee) based on one large canvas, on which you can define a path that is followed during the presentation (see the prezi Website and Nicklas Lundblad's INTERACT 2009 prezi).
Social Events
Usually, social events at conferences are not reported on, even though they are an important ingredient, particularly when it comes to networking. For example, at the CHI conference, the conference dinner has more or less been transformed into a get-together during the opening of the Commons (the exhibition area). I much preferred the conference dinners at some interesting location. In this respect, the INTERACT conference definitely gets my vote, although I have to admit that it is easier to organize a dinner party for 500 people than for 1500 people or more.
I would also like to mention that there was a music event at the conference dinner, presented by the conference chairs and some colleagues. They presented three pop songs with new, HCI-related texts (see the video at YouTube: http://www.youtube.com/watch?v=byHYlB1CNnw). We were told that this is a very popular Swedish custom, and the audience applauded generously.
Figure 6: HCI street band at the conference dinner
By the way, between the 3rd and the 5th second, I make a brief appearance on the video, probably my first ever YouTube experience:
Figure 7: In the middle of my short appearance...
Trying to Make Sense out of All the Puzzle Pieces...
After the dust had settled, I asked myself about the topics that had floated around the conference and that HCI researchers and practitioners were brooding about. These are my interpretations of what I saw and heard:
- The term "user experience" has come under scrutiny. People want a better understanding of what it means in the current context and what direction it should be headed in. This holds even more true in a changing world where are we moving from the desktop to"wild", that is, mobile environments and to environments with ambient intelligence. The term "user experience" is, however, probably more used in industry than in research...
- The same is true for HCI as a research direction. Researchers want a re-think of the human-machine/people-technology relationship, focus on human values, and support for human agency. They contrast their approach of "technology as a tool" and "augmenting humans" with an environment in which technology is "aware" and sensitive to human desires and needs, and where the system is "all-knowing" and the human is passive.
- "Interaction in the wild" and Web 2.0 (user-generated content) started the trend that allows everybody to become a designer. As a consequence, the role of designers will change ( designers may not be happy about this...): They will increasingly create malleable materials or "design for transformation" (Irene McAra-McWilliam), and people will become their co-designers. Co-experience may also fit in here...
- Surveillance (see the Panopticon, which appeared at the conference at least twice...) or collecting mass data on the Internet, in a mobile environment, or in an environment with ambient intelligence clashes with the human desire and basic right for privacy. This may, however, emerge as a "generation problem" – a lot of younger people no longer mind leaving even intimate traces on the Internet.
- The relationship between HCI researchers and industry practitioners needs to be re-considered (Kristina Höök called for "meeting industry tribes"). Current HCI research may have already lost its relevance (for industry people) because it has little to offer with respect to new societal and technical trends.
Motto
Finally, I would like to briefly address the conference motto, "Research & Practice." Did the motto fit the conference? I would say, "yes and no:" "Yes," because of the introduction of the industry day, and "no" because of most of the remainder of the conference and because the audience members were addressed and treated as researchers. The latter did not make me feel "at home" at the conference, although, compared with other conferences, INTERACT is a "cozy" conference where one would like to feel at "home." All in all, I would still leave the question mark after the motto. But future INTERACT conferences may well change this view.
Conclusion
As I already wrote in my INTERACT 2005 report, the personal value of a conference depends largely on what you make out of it for yourself. Selecting the "right" sessions and topics, making contacts, and putting at least some effort into elaborating the conference afterwards (for example, by writing a report like this one) are key to this. Last but not least, there should also be some items to ponder over:
- What does "user experience" actually mean – in general and to us as HCI practitioners working on business software?
- Does a conference that focuses on a mainly academic audience make sense today, and how can researchers and practitioners come together under one common umbrella?
These questions were raised, but, of course, not answered at this conference. So there is still plenty of food for thought for upcoming INTERACT conferences. Perhaps we'll meet in Lisbon, Portugal, in 2011?
References and More Photos
See Appendix
