Weapon development speed
From Wise Nano
How fast could new weapons be developed?
Technology assumptions
Basic molecular manufacturing: a nanofactory that can manufacture complete, high-performance products from a computerized blueprint in a few hours.
Very powerful computers (though not necessarily super-advanced software).
Design issues
Will it be hard to design and specify the physical and functional aspects of the weapon?
Chris Phoenix, CRN has argued that manufactured products are generally much less complex than software--so when we can use software design methodology (levels of abstraction, rapid build/test) to design them, the design could go quite quickly.
Conversely, it may be argued that any advanced weapon system will contain lots of software, the design of which will be just as hard as it is today; and will have to interact with other weapon systems and with users, which will be hard to specify.
A counter to this: a general-purpose weapon might be designed that did not have to interact much with other weapon systems; and with virtually unlimited computer power, behavior will be easier to specify. Some aspects of the design may be developed quickly by genetic algorithm.
Design Iteration Cycles
Developing new systems usually goes in cycles of problem definition - detailed analysis - design - manufacturing - testing - field use - new problem definition. An individual system will have several design - manufacturing - testing - new design cycles in the course of its development. MNT does not affect all of those stages equally. The big change will be to manufacturing, which will go from the bulk of cycle time to almost as fast as pressing the compile button for software. Problem definition and analysis will be more difficult since the design space and potential threats will be expanding in complexity, but can always be skipped by putting a "rough guess" concept into development. Testing will probably be the new driving factor for development. Making a weapon reliable enough to be used by an army (as opposed to terrorists) requires extensive lab and field testing, which takes lots of people. This gets even more expensive as you pile on the secrecy requirements. At a minimum weeks or months will be needed for testing each design iteration.
A terrorist group (or other NGO) can skip the testing process, since they may be happy with only a 50% chance of the device working as designed. This may have a significant impact on the balance of power between states and other elements.
