Large-scale software development is an evolutionary process. In an evolving specification, multiple development participants often hold multiple, inconsistent views on the system being developed, and considerable effort is spent handling recurrent inconsistencies. Detecting and resolving inconsistencies is only part of the problem: a resolved inconsistency might not stay resolved as a specification evolves. Frameworks in which inconsistency is tolerated help by allowing resolution to be delayed. However, the evolution of a specification may affect both resolved and unresolved inconsistencies. We present and elaborate a framework in which software development knowledge is partitioned into multiple views called "ViewPoints". Inconsistencies between ViewPoints are managed by explicitly representing relationships between them, and recording both resolved and unresolved inconsistencies. We assume that ViewPoints will often be inconsistent with one another, and we ensure that a complete wor...

This paper describes efforts to develop a transformation-based software environment which support s the acquisition and validation of software requirement s specifications. These requirements may be stated in-formally at first, and then gradually formalized an d elaborated. Support is provided for groups of requirements analysts working together, focusing on differen t analysis tasks and areas of concern. The environment assists in the validation of formalized requirements b y translating them into natural language and graphica l diagrams, and testing them against a running simulation of the system to be built. Requirements defined in terms of domain concepts are transformed int o constraints on system components. The advantage s of this approach are that specifications can be trace d back to requirements and domain concepts, which in turn have been precisely defined. 1 Introductio n We have built a demonstration requirements/specification environment called ARIES I which requirements analysts may use in evaluating system requirements and codifying them in formal specifications. This work helps to address several roadblocks in providing knowledge-based automated assistance to th e process of developing formal specifications. One of the principal roadblocks is that formal specificatio n languages are difficult to use in requirements acquisition, particularly by people who are not experts i n logic. ARIES provides tools for the gradual evolution of acquired requirements, expressed in hypertext an d graphical diagrams, into formal specifications. The