Amazon calling

Received word back on my interview with amazon.com today...

Amazon has been going through a recruitment blitz in the past few months, so it has seemed to me, to build up their services to cater to appropriate industries. Exciting stuff. And an opportunity to get back in the more thrilling aspects of the software world. A business I could be excited about. Creating software that would be used by many. An opportunity to touch the world.

...the forecast was not as rosy as I'd hoped, and I had my suspicions as I replayed (and re-interpretted) parts of the interviews in my head over the weekend.
My own analysis on my performance:

1. Should have created a full class structure for the electoral voting structure. I designed for the task at hand. Created an area class with a parent ID that would define the type of entity by a class type instance by instance. I'm sure the interviewer wanted Country State Territory and the various classes implementing some crappy interfaces. I went with simplicity, which may or may not be scalable to everything a system that grows needs. Sometimes, elegance is over-rated.
2. Singleton. Ack. How the hell did I not know that when the second interviewer was asking for a pattern for database access classes what he was looking for Singleton. That's just a lapse. There I was looking for a more complex implementation of other parts of whatevers, and I was just left with a stumble.
3. Fibonacci series. My speed in writing a method to kick out a fibonacci number was what made the dinosaurs go missing. Dim, dim, dim.

The unfortunality of the situation was just that. I was not on my game when it mattered for these guys. I did think that I came out looking capable and smart, but definitely not brilliant.