Cooper and others mention how self referential design will lead you down the wrong road, how it does not lead to solutions that solve the problem for the real users. There are few things to consider here:
- Designers, by nature, focus on different things than most people. We are tinkerers, we love to figure out how things work and why they were done the way they were. Users who take things at face vaule or are afraid to find out how and why will have a very different point of view.
- Designers or not, there are different ways of doing things. There are personal preferences, ergonomics and life experiences that alter the way we do things.
I agree with this, but I feel it is a bit too black and white. It is a true statement when considering the extreme cases, but there has to be some balance between a designers' personal knowledge of life and target user findings. We can do exhaustive user studies, but we cannot study or test every aspect of a product. Nor can we study things that do not yet exists. These things come from a designers gut and personal experiences.
Without self-referential design we loose vision for new and exciting products. We can also loose focus on the products' direction. User research can help validate and align the concept, but value has to be given to the initial idea that came from one person that ran into a problem. The need to solve the problem for themselves is self-referential design at it's best and worst.
Cooper does no advocate against having a vision, but the problem is that these concepts get distorted into arguments against any designer decision. We need to have discussions around the value of pure ideas. There is a trick to pushing a vision through the process without getting it white-washed in countless meetings. At some point, the spark is lost and you have a cheap copy of your own concept. A recent article in Wired about Apple evil/genius ways of doing business. It talks about how the most companies try to level off the playing field, allowing everyone to chip in. At Apple it's more about one person dictating where the product needs to be and when it is done. So you have Steve Job's vision being executed by the worlds best peons, which is not a bad thing. It's not that they are not contributing, it's that everyone s driving towards a single goal that may only be clear to one person.
Does this mean that having a vision is self-referential design? It depends on the level that you take each definition, but in a sense, yes. A vision will drive you against user research findings. It puts the designer's gut feeling on a pedestal. And since backing up those feelings is harder to do than using user research, force is sometimes the vehicle that will get you there as is the case with Apple.