The problem is that "developing each to the point that the merits and flaws of its underlying design concept is evident" is software engineering. It takes a substantial amount of work to get to that point, including usually having actual customers bang on the product. If you develop 3 independent solutions to the point where the flaws on each of them are apparent, by definition you will throw away 2/3 of your work.
I don't think that's necessarily a bad situation - hell, the bulk of my work is prototyping, and I throw away probably 90% of it. I just don't think this is unique to design, and I think that software engineering differs in that you usually need to proceed farther to see the flaws in your design. Science is a more apt metaphor - the scientists I know easily throw away 99% of their lifetime output, because the point of science is to prove yourself wrong as quickly as possible.
I don't think that's necessarily a bad situation - hell, the bulk of my work is prototyping, and I throw away probably 90% of it. I just don't think this is unique to design, and I think that software engineering differs in that you usually need to proceed farther to see the flaws in your design. Science is a more apt metaphor - the scientists I know easily throw away 99% of their lifetime output, because the point of science is to prove yourself wrong as quickly as possible.