My favorite KH-11/Hubble story is that during the design of the Hubble Space Telescope, NASA originally wanted a 3 meter mirror. However when they began to design and budget the satellite they were quietly told they should reconsider and use a 2.4 meter mirror, as this would be significantly cheaper since they would be able to leverage existed machinery capable of created mirrors of that size and precision. That infrastructure and manufacturing process had already been created a decade earlier for Keyhole.
30 years later with Hubble still in service, I think I definitely would have taken the 3m mirror(which is >50% bigger than a 2.4m mirror). Couple that with the fact that despite all the cost savings, Hubble's mirror was famously flawed for the first 3 years of its life and required special missions and attachments to fix the problem.
That's easy to say in hindsight. It could just as easily have gotten Hubble cancelled.
> I think I definitely would have taken the 3m mirror(which is >50% bigger than a 2.4m mirror)
Opportunity cost. That budget might have been passed over, and we might not have gotten Hubble at all. We're lucky we got the discount.
Thinking ahead, it's not a bad idea to mix in many smaller budgeted projects, such as the Discovery and New Frontiers programs seek to accomplish.