Software entities are more complex fortheir size than perhaps any other human construct, because no

Software entities are more complex fortheir size than perhaps any other human construct, because no two parts arealike (at least above the statement level). If they are, we make the twosimilar parts into one, a（71 ）, open or closed. In this respect software systems differprofoundly from computers,buildings, or automobiles, where repeated elementsabound.Digital computers are themselves morecomplex than most things people build; they have very large numbers of states.This makes conceiving, describing, and testing them hard. Software systems haveorders of magnitude more （72 ）than computers do.Likewise, a scaling-up of a software entityis not merely a repetition of the same elements in larger size; it is necessarily an increase inthe number of different elements. In most cases, the elements interact witheach other in some（ 73 ）fashion,andthe complexity of the whole increases much more than linearly.The complexity of software is a(an)（74 ）property, notan accidental one. Hence descriptions of a software entity that abstract awayits complexity often abstract away its essence.Mathematics and the physicalsciences made great strides for three centuries by constructing simplifiedmodels of complex phenomena, deriving properties from the models, and verifyingthose properties experimentally. This worked because the complexities（75 ）in the modelswere not the essential properties of the phenomena. It does not work when thecomplexities are the essence.Many of the classical problems ofdeveloping software products derive from this essential complexity and its nonlinear increases withsize. Not only technical problems but management problems as well come from the complexity.
A：surface
B：outside
C：exterior
D：essential

(中级)软件设计师     中级     construct     entities     Software     fortheir