To invent anything, you have to be removed from the world. In order to have the capacity, the liberty, to imagine something better, you need to step outside of it for awhile.
My advice is to encourage invention and ideas, and then edit. It's about proliferation and promiscuity on the one hand - and then later, rigorous, tough-minded editing. Deam Kamen, the iventor, calls the process "kissing frogs." You might make 100 things and turn one of them into a prince.
What's truly sane about the approach is that the from that you make today doesn't have to be beautiful. There's no need to get hung up on a "good idea." later on, the process of choosing - making sure the good idea doesn't get lost - becomes largely intuitive. I my experience, it has to be.
But product invention isn't just about the product. It's also about the relationship, the flow, the information that surrounds that product. If you say that the actual object is the thing, then you're missing the point of what it means to invent in today's world.
It's true that to think about a new product, you first have to consider it on its own. But not long after, you have to force yourself to do a mental flip and understand that it's really not discrete at all. For example, car manufacturers don't want to sell you a car. They want to sell you 10 cars. And so, they're going to sell you the relationship, the communication, the experience of that car. The car, the product, is part of a bigger flow. The real challenge for an inventor is to understand how it fits into the larger context.
-Bruce Mau, BMD Toronto
Full article <http://www.fastcompany.com/magazine/79/fasttalk.html>