The life of a prototype can vary. On one hand the time before a prototype expires and is thrown away can be very short like in wizard of oz paper prototyping. In this case the prototype also usually requires a great deal of hand holding and filling in the blanks. It requires someone to be present to breath life into it for the user to experience it during evaluation. In some way, the presence of the evaluator and short life makes such an evaluation more sterile or artificial.

On another hand, prototypes can also live longer on their own and be self-sustaining. These types of projects are visible in such things as labs prototypes which over the last few years have become quite popular. Such prototypes are very much close to finished products, yet the people behind them still see them as experiments worthy of being thrown away. More so, they allow the user to experience the prototype during a longer time frame making the evaluation of them closer to reality.

Very much related to the posts on alternatives, here is a quick sketch about one snapshot of what I believe constitutes good design. The sketch underlines the importance of stretching the design space by means of exploring alternatives. The wider the design space, the richer the pool of ideas from which the best can be selected by means of evaluation. (This sketch has been inspired by 90percent of everything, Bill Buxton and evolutionary design.)