"If an engineer misses details and won't spend time looking for a more optimal solution on a white board, I wouldn't expect them to do it in the day-to-day work of their career either."
What a foolish generalization. This is exactly the kind of attitude that I don't like. Just because someone starts with a sub-optimal solution doesn't mean he can't/won't improve it. Also my approach was to start with a sub-optimal solution so that we have a foundation to build on, which I think is much better then spending your entire interviewing time thinking up a dynamic programming algorithm and leave a blank board at the end.
Optimizing a solution is an art, and can take varying amount of time, and sometimes even your sub-optimal solution might be better than a more complicated optimal solution.