This paper summarizes the results of a study which compared the efficiency of two methods of measuring program similarity in the context of novice programmers trying to reach identical objectives. Both methods look for similarity by comparing 'program profiles'. Such profiles are created by feature extraction routines which map each program onto a tuple <f 1 , f 2 ,..., f n > where each f i is a count of an occurrence of a particular feature. A comparison routine is then invoked which detects similarities between tuples. The results showed that in this environment the comparison routine based on the Halstead metric failed to perform as well as a conceptually simpler alternative.

Sociological theory poses the issue of whether ideational or coercive factors provide the central basls of social order. The former view maintains (and requires) that a value consensus exist in the empirical world; the latter assumes the existence of a ruling class which dominates the ideational institutions of the society. A preliminary scrutiny of available evidence suggests that: (1) a value consensus does not exist, and (2) there is a high concentration of wealth and economic control which appears to extend into vital political and ideological areas, thereby implying the existence of a dominant or ruling class. Based upon these conclusions, the concept of ideological hegemony is introduced as a significant component of a Marxist view of social order. The hegemonic process is described and available evidence is examined in the areas of political socialization and mass media. A pattern in which debate and discussion are circumscribed while alternative values and world views are ignored or suppressed was found to characterize these two areas. This pattern was seen to provide further support for the Marxist view that an ideological hegemony imposed by the dominant class is the ideational manifestation of a social order based upon coercion.

No abstract

Database audh trails record each morMication to data values, and thus assist management in controlling 6nattcird losses atu%r legal vulnerability. Most commercial databasea lack integrated audh mail mechanaims. While tiggers ptuvide a means for a database administrator to implement an audit trail, they impose a specific spsn of coverage, and generally require progrdng to effect a change in audit trail policies. The emergence of deductive databases, such as LDL, makes possible the &sign of mechanisms for audit trail administration that are both more powerful and more elegant, as ilhtsmated herein.

No abstract