Software requirements: Are they really a problem?

Author(s): Bell, T. E., and T. A. Thayer
Venue: International Conference on Software Engineering, Proceedings of the 2nd international conference on Software engineering, San Francisco, California
Date: 1976

Type of Experiement: Case Study
Class/Experience Level: Graduate Student
Participant Selection: class participation


The paper discusses about the analysis of the data on two different types of software requirements to determine what kinds of problems occur and whether the problems are important. The empirical studies concentrated on documents with major emphasis on software requirements defined in the paper. The paper analyzes two cases, a small system and a large system, for comparison. The paper concludes that software requirements are important and “their problems are surprisingly similar across projects.”

Process Outline

  1. Students met informally between class sessions to generate the requirements and design
  2. Formal meetings occur between the two teams during design to resolve questions about the requirements
  3. Members of both teams record problems detected in the requirements but no controls were imposed to ensure complete documentation

Software requirements are important and requirement problems are frequent and important. Early improvement would reduce the cost of designing to requirements that are subsequently changed – with the result that the design work must be repeated.