A multistage clustering methodology to partition large developing societies into a collection of communities with distinct characteristics is introduced. The methodology is based on multiple attributes expressing the socioeconomic status. In the context of our approach, no index of development is needed. An excellent application of the developed methodology lies in investigating the telecommunication needs for rural development in Egypt. The merits of our approach are reflected in overcoming the data inavailability on the village level and providing for an appropriate size of communities for field experimentation and analysis. An additional benefit is that the intricate sampling problem is indirectly solved.

The bandwidth limitations of the practical analog-computer units lead to dynamic errors in the computer solution. These errors can be represented either as a slight shift in the root positions of the system characteristic equation or as perturbations in the constant coefficients of the differential equation under solution. In this note, the authors report a unified approach to combine these two representations of dynamic errors and also show that these errors can be compensated for by setting the system after initial perturbations in its constant coefficients and in its driving functions have occurred. The approach is generalized for a system of simultaneous linear differential equations with constant coefficients. This technique will be of value in achieving a reasonable degree of accuracy with relatively inexpensive analog-computer components, and in extending the range of operating frequencies for a certain specified accuracy.

Large-scale scientific computers are characterized by having a high ratio of computing requirements to the input-output requirements. Thus the speed of arithmetic operations should be of prime importance in this group of digital computers. In this note a mathematical formula is derived to determine the multiplication time obtained by each of the known high-speed multiplication techniques. Then a comparative analysis is performed to show the amount of speed improvement resulting from each technique and the influence of the different factors affecting speed.