Okamura: We had to work on quite a few holidays and although we didn't do any all-nighters, we did work until late at night.

Ohba: I didn't start to hate my work, but there were times when the data became stagnant and I was troubled. There was a period when we thought, “The reflectance won't increase,” and we spent one or two months thinking about nothing but increasing the reflectance.

Ohba: We knew what should work under the ideal situation, but we couldn't achieve that situation and initially nothing would change regardless of what we did. One day the data deteriorated, then the deterioration things changed and suddenly the data began to improve. That's when the ups and downs started, but we didn't know why. Although it would seem that things should improve in increments, such as 90% to 91% to 92%, things just going up and down.

Ohba: I never dreamt about this problem. (Everyone laughs) I never dreamt about this problem. I was thinking about optical ducts all day long at the time, but I never dreamed about them. But I was always thinking about them when I was awake.

Okamura: Ohba said, “None whatsoever,” but that might have been where we found a hint. I guess it's a sense that a researcher develops. When we're asked, “Why did that happen?” it's not something that we can specifically point to.

Okamura: We were always talking. Things such as, “I'm thinking of conducting this kind of experiment. What do you think?” and “What about this method?” Such questions and replies were constantly exchanged and we conducted many different experiments. In a product development environment, I think it is critical to ask, “We've done everything we can think of. What's next?”