Stories

back Stories Overview

SAP User Experience

Projects

ASUG

Conference Reports

Future Scope

 

INTERACT 2005 – Communicating Naturally Through Computers?

By Gerd Waloszek, SAP User Experience, SAP AG – October 12, 2005

INTERACT 2005, the 10th IFIP TC13 International Conference on Human-Computer Interaction, is over. It's time to look back and ask: "Was it worth going to Rome?" Surely, the trip to Rome was worthwhile, but was this conference? The answer is as always: It depends what you make out of your conference visit, whether you find the time to reflect on it, look into the conference proceedings, or even write a report about it. It’s different from the CHI conference in a number of ways: INTERACT is smaller, much more academic, and there are only a few students looking for jobs.

Interact 2005 logo

Figure 1 : Logo of the INTERACT 2005 conference

In this report, I cannot cover the whole breadth of the conference – a look at the topics shows that most up-to-date issues were dealt with in one way or another. Instead, I will focus on the plenaries and panels and add a few remarks with respect to the conference motto, "Communicating Naturally through Computers."

The conference featured three guest speakers, Bill Buxton, Flavia Sparacino, and Steven Pemberton. Bill Buxton (Buxton Design) promoted the use of sketching as a tool for early design and for exploring as many design ideas as possible. "Many people believe that you cannot waste and throw away good ideas," Buxton remarked, "but good ideas are not precious." If you believe that they are precious you tend to grow attached to them. Buxton contrasted sketching with prototyping. According to him, prototypes are too specific, already require some investment of resources, and are thus much more difficult to throw away than sketches. Prototypes are already concrete answers to design problems, while sketches are only questions or proposals. Buxton also rejected the spiral metaphor for design, according to which designers gradually narrow their focus to a single solution. For him, a tree with many branches, indicating exploration alternatives, is the better metaphor for design. In the course of the design process, designers cut most branches – that is, reject alternatives – in order to find the right solution or set of alternative solutions. Buxton also asked the "simple" question: "Where do we spend our time?"

  • Getting the design right (Buxton stated that 90% of development time is allocated here – this is where usability is achieved) or;
  • Getting the right design

Buxton surely wants to promote the second option. But let me ask who will be appreciated more by his or her manager – the designer who muses about the right design and draws sketches, or the designer who hacks prototypes and produces redesigns? In most companies, there is still a climate where one is expected to sit in front of a computer and produce something. I do not see this changing in the near – or distant – future.

Bill Buxton   Flavia Sparacino   Steven Pemberton

Figure 2 : The invited speakers: Bill Buxton, Flavia Sparacino, and Steven Pemberton

In her talk, Flavia Sparacino presented a "new architecture for the information age." She wants to redesign spaces for these needs by integrating augmented and mixed reality (or ambient intelligence) into classical environmental spaces. Examples of such spaces include: museums, cities, performance spaces, homes, work spaces, entertainment spaces, and retail spaces. In the future we will have smart places, smart rooms, smart clothes, and more smart things in our environment, with the prospective side-effect – as Norbert Streitz (IPSI Fraunhofer) stated in the “Clippy” panel – that “smart spaces make people smarter.” The human body acts as an input device to operate (or communicate with) these smart environments through movements, gestures, and possibly mimicry and physiological data (see panel on context and emotion-aware computing). Sparacino, who owns a company that markets her ideas (Sensing Places, an MIT offspring), gave several demonstrations of such systems, most of them tailored to museums. She also presented examples of her vision of a city of the future under the motto “The city as canvas,” including large light installations and interactive shop windows and billboards. While you will still be able to switch off your TV and radio, in the city of the future all these insistently flickering canvasses will be harder to ignore...

The third guest speaker, Steven Pemberton, who was and still is deeply involved in the development of the Web, started his talk with Moore's law (according to which computing power doubles every 18 months) and asked what the extra power we gain from year to year is used for. Answering his own question, he complained that it is mostly used for pixel-pushing and that most computers are idle during most of their active life. Pemberton asked, "So why aren't we using that extra power to make life better?" As far as I remember, he did not provide an answer. Instead, he praised the power of declarative, high-level programming and demonstrated how this programming model can dramatically simplify coding – he probably thought that this was an answer. Pemberton showed that, using the declarative approach, you can easily create Web applications using the new XForms standard and open APIs, such as those offered in Google Maps (Jared Spool calls the new trend of assembling Web applications using open APIs "Web 2.0").

At INTERACT 2005, there were four panels, three of which I was able to attend. While I was a bit disappointed with the outcomes of these panels, they were still good food for thought. One panel debated "The Challenge of Personal Information Management (PIM)," which is indeed a challenge for me, too. With the advent of the Web, I tried to organize my information on the computer through linked HTML pages but gave up long ago. Regardless of whether PIM becomes PROM – that is, includes personal role management, as Catherine Plaisant (HIL) suggested – or you simply record everything and hope for technological breakthroughs, as Mary Czerwinski (Microsoft Research) suggested – PIM is not just a technical problem, it's human, as Plaisant stated. Recognizing this, Tom Moran (IBM) focused on the social aspects of PIM and made a case for what he called "social activity management" and mutual information sharing.

Catherine Plaisant   Tom Moran   Mary Czerwinski

Figure 3: Some participants of the PIM panel – Catherine Plaisant, Tom Moran, and Mary Czerwinski

This seamlessly leads us to privacy issues and the other two panels, "Context and Emotion Aware Computing" and "Trust and Incidental Interaction: Would You Let a Talking Paper Clip Run YOUR Home?" I extracted three issues from these panels: control, trust, and privacy. For example, context-aware applications recognize the people and places they interact with. Thus, we have to provide them with information on our identity (for example, through "profiles" that we transmit wirelessly), on the properties of places, and much more. Packed with all this data, will applications keep them secret, or will they stockpile, distribute, or even sell them? Emotion-aware systems will be able to detect our emotional and attentive states. No more sleeping in board meetings! Our data can be used for statistical purposes and condensed into trends and profiles.

Turning to the third panel, Andrew Monk discussed "lifestyle monitoring" – or more accurately "activity monitoring" – of elderly people and asked: "Will I let an activity monitor make suggestions about my care needs?" It's only numbers, but enough of them can tell a lot about you – and be used in ways that you may not agree with you if you are not in control of the data. You may also no longer be in control of computer system interaction due to its implicit (Albrecht Schmidt, University of Munich) or incidental nature (Alan Dix, University of Lancaster). With intentional interaction, the user has the initiative: He or she presses a button, and the lights go on. With implicit but expected interaction, the system's reaction comes at least to our attention: We enter a room, and the lights go on. With incidental (or implicit) interaction, however, the system acts without our intention and attention: We enter a room, and the air conditioning increases (examples from Alan Dix, slightly modified). As one way out of all these control and privacy issues, Albrecht Schmidt answered the panel's topical question with: "I think, I'll let the paper clip run my house, but first I need to know how to cheat it." In the end, the overall issue is trust: How much trust will we have in these systems, and will we even have a chance not to "provide" our data if we do not trust them?

Alan Dix   Andrew Monk   Albrecht Schmidt   Norbert Streitz

Figure 4 : The participants of the "Clippy" panel – Alan Dix, Andrew Monk, Albrecht Schmidt, and Norbert Streitz

Finally, I would like to shortly address the conference motto, "Communicating Naturally through Computers." In the beginning, I had misread the motto and thought that it was "Communicating Naturally with Computers." Even though I still have problems with using the term "communicate" with respect to human-computer interaction, I would at least have understood what this motto means. But when taking a closer look at the conference logo, I found that I was mistaken – it is "through," not "with." In this form, the conference motto does not make much sense to me: It sounds as if the computer were just a medium between people, such as a cell phone or e-mail. (Both depend on computers, but have their own domains of usage despite differing dramatically with respect to their degree of “naturalness.") Finally, I decided to take a look at the conference proceedings for clarification. And what did I find in the editor's preface? This sentence in particular: The theme of INTERACT 2005 is "Communicating Naturally with Computers." OK, I give up...

References

Conference and Rome

Plenaries

Bill Buxton

Flavia Sparacino

Steven Pemberton

Panels

The Challenge of Personal Information Management

Context and Emotion Aware Computing

Trust and Incidental Interaction: Would You Let a Talking Paper Clip Run YOUR Home?

 

To top top