Computers are improving in power, speed and affordability by an order of magnitude every five years. Thanks partly to parallel improvements in miniaturisation and ruggedization, the use of this speed and power in C⁴I systems on the battlefield seems set to increase dramatically in the next few years. In spite of received wisdom about children's familiarity with computers, there is no evidence of any equivalent improvement in the ability of recruits to operate these systems. If this lack of ability is not to become a limiting factor on the "Digitized Battlefield", an affordable, dependable and practical training programme for C⁴I systems is urgently needed.

Training for C⁴I systems inevitably involves extensive use of computers as training devices. The widespread use of Computer-Based Training (CBT) and Distributed Training (DT), possibly embedded in operational C⁴I systems, will be essential in future to combat the twin scourges of skill-fade and rapid version upgrades for large, highly distributed user populations. Synthetic Environments (SE), of varying degrees of abstraction, will need to be incorporated within most, if not all, stages of such training. But C⁴I systems, unlike weapon systems and vehicles, tend to be developed using Rapid Applications Development (RAD) techniques. The use of RAD means that "design freeze" may occur after roll-out or may actually never occur at all. The long lead-times usually associated with CBT, DT and SE design and production are inconsistent with such rapidly changing requirements. At the same time, the costs and risks associated with the development of CBT, DT and SE make some form of rapid yet rigorous justification process highly desirable.

Thanks largely to the emerging standardization of computer user interfaces, it is proposed that a generic model of C⁴I systems training is now feasible. By adopting a scaleable default training solution at the outset of any C⁴I project, a strategy of modifying such a model as the main project develops is likely to be more responsive than the current strategy of starting from "scratch". It should also provide a reasonable initial cost estimate for training, a feature missing from most current C⁴I system requirements. Such a model has the added advantage that best practice could be incorporated incrementally, refining it over time. In this way, much of the analysis and design process could be re-used, thereby becoming both faster and more efficient.