Sunday, March 31, 2013

Book note: Can Robots Commit Crimes?

I read today in the Chronicle of Higher Education about a forthcoming book "When Robots Kill: Artificial Intelligence Under Criminal Law" (forthcoming from Northeastern University Press), by Gabriel Hallevy. It is fascinating to see the growing debate about the 'nature' of robots and intelligent systems and to what extent they should be considered to have agency of some kind and therefore also responsibility for their actions. According the the Chronicle, Hallevy makes the case that we already hold other non-human entities responsible for their actions, for instance corporations even though they have no spirit, soul or physical body, so why not robots?

As someone who read all the works by Isaac Asimov (a long time ago) these questions are not new. All Asimov followers know his Three Laws of Robotics. Asimov envisioned a future society where robots had intelligence and agency in way that we are still far from. In his writings he explored what the relationship between humans and robots would or could look like, and what we humans could do to protect ourselves from robots. We are of course still far from the world that Asimov wrote about, but maybe we are getting closer. It seems as if Hallevy's new book is a sign of that.

Tuesday, March 26, 2013

The Evil of Design

Even though I am happy to see the wonderful push for design thinking and a designerly approach today in academia and in business, it is also a bit disturbing to see the lack of critical thinking about design. Design as an approach is today by many seen as the silver bullet to almost any kind of problem. A design approach is considered to be able to deal with any kind of situation. I do agree that design as an approach is powerful, maybe more so than many believe even among those who advocate it. I do agree that many issues today should be approached in a designerly way. But it is also crucial to remember that design as an approach is not inherently good.

Almost all things that scare us and make our lives difficult and dangerous are designed. Some of the most wonderful examples of great design are also considered to be manifestations of evil. Humans design wars, genoside, weapons of destruction, and maybe even more extraordinary but less obvious designs aimed at suppressing people (such as political, governing or business structures), sometimes even in combination with wonderful 'user' experiences that makes people appreciate being oppressed (what a great design!).

In this new emerging era of design thinking and designerly approaches it is important to remember that design is only an approach--it is a process. It is a powerful approach, but there is no guarantor that the outcome will be good design and there is no guarantee that the design process will not lead to evil designs. At the end of the day, it all comes down to the character of the designer. Values and beliefs guides and shapes the judgements made by the designer. Developing design competence therefore means developing once personal character as well as once design thinking ability. 

Anyone engaging in design in a serious way therefore has to be constantly aware of and reflect on the 'nature' of design and also be highly critical of any simple versions of design that promises processes that will lead to good results in some 'automatic' prescriptive guaranteed way. Critical thinking is as important in design as in science.

[In the book "The Design Way" these issues are discussed in the chapters "The Evil of Design" and "The Guarantor-of-Design (g.o.d.)"]

Tuesday, March 19, 2013

Design thinking revisited by Norman

Don Norman has in some earlier writings been quite critical to the notion of "design thinking", however now he argues that he has changed his mind. I agree with the new Norman. He makes the case that if we see design thinking more as a tool or method then he is ok with it. And he is also careful with stating that it is not the case that all designers are engaged with design thinking, or that all design schools do teach it. I could not agree more. Working with designers all over the world I know first hand that design thinking is not necessarily understood everywhere where the label design is used. Even worse is that a deeper understanding of design as a human universal approach (much broader and critical than design thinking) is almost completely lacking still. But hopefully this is changing in the years to come.

I have one problem with Norman's new position though. He writes "Two powerful tools of design thinking summarize the approach: the British Design Council's "Double-Diamond, Diverge-Converge Model of Design"; and the iterative process of Observation, Ideation, Prototype, and Test called "Human-Centered Design." I think it is dangerous to substitute a way of thinking with a couple of formalized and standardized methods with their own structure and process. Design thinking to me means a critical stance that makes it possible to engage with a situation in a way that is suitable to the context and circumstances. I am sure there are many design situations that would not be suitable to approach by the Human Centered Design process that Norman describes. I have seen numerous design process carried out in an excellent way that in no shape or form resembles that process. So, design thinking, is about thinking, not about particular tools or methods, it is the thinking itself that is the tool!

Sunday, March 10, 2013

Interface tile designs--evolution or fashion

I guess you have already realized that one of the most popular interface design styles today is tile design. The examples are many, Pinterest (maybe where it started), Windows 8, USA today, etc. Even "older" designs are changing their design to become more tile like. Ebay is trying to look like Pinterest.

The tile design has an immediate appeal. It is clear, structured, appears to be highly organized and of course, is in many cases visually appealing. We have over the years seen many "styles" come and go, usually they have been seen as evolutions of the interface--new paradigms. GUI is seen as superior to command based interaction for instance and not as a style or fashion. Touch and gesture is seen as developments away from traditional keyboard interactions--not as a fashion. But is tile interfaces the next step in the evolution of interfaces or is it only a style, a fashion? And if so, does it matter?

My own very personal and unscientific analysis of tile interfaces has led me to the conclusion that the positive aspects I mentioned above does not come without problems. The dominating visual aspect of tiles seems to make it difficult to create distinctions between different types of content and also between differences in importance (apart from size). It also seems to be difficult to create a supporting browsing structure that does not "force" the user to inspect all tiles in the same way and with the same effort. I have to admit that I have not done enough examinations of this to be able to argue for any of these observations, but I am sure that they could use some study.

If tile design is less an overall evolution and improvement of the interface and more of a fashion, it does not mean that is necessary bad or less valuable. Styles and fashion are common in every area where it is difficult to make serious developments on the fundamental qualities of an artifact, for instance, clothes. New styles and fashions can be seen as ways to keep things from becoming boring and repetitive. And that is maybe what the tile design paradigm is all about.

I would like to hear what others think about this. Maybe interaction design, at least some aspects of it like traditional screen design, is at  a stage where style and fashion will become even more important and truly new interaction paradigms will be less common.

Personally, I am already a bit tired of tile designed interfaces. They do not work well for me. USA Today is a good example where I do not see any positive improvements when it comes to the overall user experience. But, I may be wrong.

Thursday, March 07, 2013

Interfaces of not?

During the last few years we have seen a drastic change in what is usually seen as the interface of interactive artifacts and systems. There are many who examine this change, such as this article about "Why The Human Body Will Be The Next Computer Interface".

I am excited to see speculations about this since I am working with Lars-Erik Janlert on a new article about this topic. The article is to some extent based on our previous article "Complex Interaction" even though it hopefully breaks some new ground when it comes to the big questions of "what is an interface"  and the future oriented question of "what is the future of interfaces".

There are a number of interesting questions that we try to investigate int the article, for instance, what is the most useful definition of interfaces, does the Kinect have an interface, when interfaces are everywhere how can we think about them, etc. These questions are important since if answered they may help and support designers who on an everyday basis make decisions about the manifestations and materializations of interfaces. Questions like "should we try to make the interface smaller", "should we try to remove the interface" or "are gestures a good form of interaction in this case". The consequences of all these decisions are not obvious even though for a particular device they may make be. When we look at the landscape of interfaces that we all live within, every decision regarding one interface will influence the use and usefulness of all the others. There is an 'ecosystem' or landscape effect of all design decisions. Anyway, in our article we deal with all these issues and also relate them to the history of interfaces.