WHAT ASPECT OF THE PROBLEM WAS MOST DIFFICULT?
Minimizing the merit function given the required detector 
size and mass constraint. What was interesting is that 
the 70% relative illumination at full-field fell right out 
of the design; that is, I didn't have to specifically set 
up the problem to achieve it.

WHAT INSIGHTS (IF ANY) DID YOU GAIN FROM WORKING ON THE PROBLEM? 
No real insights, but I now have a greater appreciation for 
air-spaced systems!

ADDITIONAL COMMENTS: 
I initially found it hard to interpret how the merit function 
was going to be evaluated (i.e., at 0.7, 0.8, and 1/2 full-field). 
I was initially confused by the mix in the decimal and fractional 
notation.