Thursday, October 27, 2005

Design and Theories

[This post is a bit too long! On the DRS mailing list there is an ongoing discussion on the role of theories and the practice of design. This is a version of my post in that debate. Be aware that here it is outside the context of the list. By the way, that list is now and then really interesting. You can search for PHD-DESIGN@JISCMAIL.AC.UK]

I have with interest read the thread on theory in design. One of the interesting aspects is what constitutes a theory that has the power to influence practice. Some valuable comments have been made on this issue. Another aspect that has been mentioned has to do with the possibility of knowing if a design is an actual result of a theory. I just want to make a few comments on these two issues.

First, there is a basic paradox involved when it comes to the idea of theories influencing design. If we accept a very simple definition of design as a process aimed at producing something new, something unexpected, something not already known or existing, then any "tool" we are able to create and develop (such as a theory) will to some extent lead to a situation where a larger part of the final result will be dependent or a consequence of the used "theory".

So, in a world where we have really, really good theories, we will have a design process that with certainty will lead to the "best" solution. However, at the same time the design process will not be a design process anymore, at least not in the sense we defined it above. So, our ambition to come up with design theories that will "help" designers is (in the very extreme case) an attempt to reduce the very aspects of the design process that we usually see as the core, character, and sign of design.

This is of course no problem in itself. If we can create theories that actually fully reduces uncertainty in design it might be a good thing. However, I am pretty sure that if we get closer to that situation the very foundation from where we judge design results will change, maybe not because we are not happy with our "perfect" designs, but because we might be bored and seek variation. This paradox is not today a real problem, neither to design practice or design research, but I think it is important to reflect on the paradox while we pursue our intentions (both in practice and research).

My second comment is on the idea that we can "measure" the success of a theory. The idea that design is to produce something new makes it almost impossible to measure the influence of theory in practice. This is also a kind of paradox. If we assume that a specific design is a result of a specific theory we have to assume that there is a logical causation between the ideas of the theory and the final outcome. This is something that might be the case all the time without us being able to detect such logical connections in the extreme complexity that is manifested in every design situation. In order to make the case that a theory is (to some extent) actually "responsible" for the design outcome also means that we have to reduce the influence and importance of the designer (and especially designer qualities such as experience, imagination, intuition, intelligence,...). This is again not necessarily a problem, but it is a "paradox" that we need to be aware of in our attempts to understand the role and place of theories in design.

In both the cases I have discussed above there is a paradox that we have to acknowledge. These paradoxes are not present in the world of science. The paradoxes does not constitute anything that necessarily lead to any answers when it comes to how to approach or develop theories. Some people will probably argue it is a question of balance or that the paradoxes are only important at the extreme ends of a continuum, and are not anything relevant in the "real" world. This might be true, and maybe that kind of realization is valuable as a reality check when it comes to our hopes and aspiration on what role theories can play.

So, I am not sure what I actually want to say with this posting. Maybe it is only to make the case that the role and place for theory is a question of definition and will, but not of the idea of "theory". Instead it has to do with how we define design. There are some aspects of what theories are, that are consequences of the way we define design and maybe also want design to be. We can re-define design in a way that actually removes the paradoxes. However, at the same moment we will remove some of the very qualities of design that we love and are drawn to.

Some of the postings have mentioned other ways to understand theories in design. The idea that theories can "inform" designers opens up for a way of understanding and working with theories that changes the preconditions for the paradoxes. However, this position is not well developed. It is a position that should benefit from more deliberate reflection from our community, both practitioners and researchers. Even if there are many advocates for this position it is not clear, for instance, how such a position should/could be implemented in design education.

Well, I really have to stop......

Erik Stolterman

Saturday, October 15, 2005

I love Robots

An interesting thing today on TV. It is an ad where an automatic vacuum cleaner is advertised. The whole ad builds on the idea that the cleaner is a robot. And the ad ends with people looking into the camera saying "I love Robots". Maybe we are entering the era of robots, finally! Isaac Asimov wrote about our relations and interactions with robots in his famous novels. And in some movies we have seen ads with very similar message as the vacuum cleaner ad, that is, we love robots. I think the idea of robots as a way to interact with artifacts will be more common. The whole idea of artifacts that behave, move and interact with humans is such a powerful idea and image, proven to work in all kinds of fiction. That attractive quality will probably pursuade us to willingly move into the Age of Robots!

Sunday, October 09, 2005

Results-oriented UI

Recently Microsoft presented their new User Interface for the new coming version of Office. The basic idea is now "results-oriented". It means that the user should only focus on the desired results and not on how to get there. See for instance Jacob Nielsens comments. All attempts to increase the ordinary users possibility to work in an easy way is welcomed and so is this one. It is however difficult to see the great "philosophical" step that this is supposed to manifest. The distinction between what is an operation and what is a result is delicate. The common way of pointing to a place on a page where you want the page number can be seen as an operation or as results-oriented depending on the chosen level of abstraction. Either you see it as if you command the number to be set at a specific place, or you see it as an act of desire, i.e. you point to a place where yuo want the number to be placed. Of course, the more of actions done by chosing instead of defining increases the number of commands while decreases the number of possible actions. It seems though that the new Office will make it possible to have both styles. So, while this might well be a good evolution of the Office product line, it does not really seem to be a radical change or some breakthrough in basic principles. But, it seems to be a healthy development of one of the most used user interfaces. Let's hope for the best.

Saturday, October 08, 2005

Consequence of transformation


Just read this piece on phone services that are free on the net. It seems that there is still a long way to go before we understand the new dynamics of what a long time ago was called the "new economy". It is obvious with companies like Skype and Vonage that we are still only at the beginning of the transformation that everyone talked about in the late 90s. It is also obvious that all those who after the "bubble" made fun of the "new economy" and the "whole thing" maybe laughed too early. I think we all should be humble in front of the changes that in so many ways constitute manifestations of our transforming grounds.

Thursday, October 06, 2005

i-Conference, part 2

Well, back from the i-Conference. A growing number of departments and schools, deans and faculty, in the field of information/informatics/computer sciences seem to be recognizing that somthing new is developing. It was a really interesting conference, and I think something that will be remembered many years from now as the first conference, the starting point.

The School of Informatics where I am at is in the middle of this new development.