I was asked this question when demoing a research interface at a presentation recently. It’s a good question, and I think one we in the research HCI field tip-toe around a lot.
The answer I gave was that concept interfaces are meant to explore a new idea. A successful concept presents an idea what is compelling enough to warrant future work. Any interface, ever, is going to require a ton of polish. I specifically cited the 80/20 ratio, where 20% of the project appears to take 80% of the work. That final 20% of the project, of the design of your new interface, is going to take so much work that you’re out of the realm of research and into regular UI engineering.
It’s really hard to figure out where to draw the line between research and polish. If you’re trying to describe a new way of adaptively moving buttons around, it needs to be polished enough to be obvious, even if the buttons aren’t their final sizes. However, it hurts a little when people note that the button sizes aren’t quite right when, you know, I’m trying to figure out how to make them move.