Hans W. Nissen

Managing multiple requirements perspectives with metamodels

Stakeholder conflicts can be productive in requirements engineering. Capturing, monitoring, and resolving multiple perspectives is difficult and time consuming when done by hand. Our experience with ConceptBase shows that a simple but customizable metamodeling approach, combined with an advanced query facility, produces higher quality requirements documents in less time.

ConceptBase Managing Conceptual Models about Information Systems

ConceptBase is a meta data management system intended to support the cooperative development and evolution of information systems with multiple interacting formalisms. It supports a simple logic-based core language, O-Telos, which integrates deductive and object-oriented features in order to support the syntactical, graphical, and semantic customization of modeling languages as well as analysis in multi-language modeling environments.

Repository support for multi-perspective requirements engineering

Abstract Relationships among different modeling perspectives have been systematically investigated focusing either on given notations (e.g. UML) or on domain reference models (e.g. ARIS/SAP). In contrast, many successful informal methods for business analysis and requirements engineering (e.g. JAD) emphasize team negotiation, goal orientation and flexibility of modeling notations. This paper addresses the question how much formal and computerized support can be provided in such settings without destroying their creative tenor. Our solution is based on a novel modeling language, M-Telos, that integrates the adaptability and analysis advantages of the logic-based meta modeling language Telos with a module concept covering the structuring mechanisms of scalable software architectures. It comprises four components: (1) A modular conceptual modeling formalism organizes individual perspectives and their interrelationships. (2) Perspective schemata are linked to a conceptual meta meta model of shared domain terms, thus giving the architecture a semantic meaning and enabling adaptability and extensibility of the network of perspectives. (3) Inconsistency management across perspectives is handled in a goal-oriented manner, by formalizing analysis goals as meta rules which are automatically customized to perspective schemata. (4) Continuous incremental maintenance of inconsistency information is provided by exploiting recent view maintenance techniques from deductive databases. The approach has been implemented as an extension to the ConceptBase ‡ meta database management system and has been applied in a number of real-world requirements engineering projects.

A decision-based configuration process environment

In the context of the ESPRIT project DAIDA, we have developed an experimental environment intended to achieve consistency-in-the-large in a multi-person setting. Our conceptual model of configuration processes, the CAD° model, centres around decisions that work on configured objects and are subject to structured conversations. The environment, extending the knowledge-based software information system ConceptBase, supports co-operation within development teams by integrating models and tools for argumentation and co-ordination with those for versioning and configuration. Versioning decisions are discussed and decided on within an argument editor, and executed by specialised tools for programming-in-the-small. Tasks are assigned and monitored through a contract tool, and carried out within co-ordinated workspaces under a conflict-tolerant transaction protocol. Consistent configuration and reconfiguration of local results is supported by a logic-based configuration assistant.

Query by class, rule, and concept

The ideal query language for a knowledge base will probably never be found: easy formulation and easy evaluation of queries are two conflicting goals. Easy formulation asks for a flexible, expressive language near to human language or gestures. Easy evaluation of queries requires an effective mapping to machine code, which computes the correct answer in a finite number of steps. This article approaches the problem by a query language with three faces. The first projects queries to concepts of the knowledge representation language KL-One for easy formulation and readability. The second presents queries as rules of a deductive database with fixpoint semantics. The third presents queries as classes whose instances are the materialized answer (view) to the query. The methods for maintaining and updating the views are compiled from their deductive interpretation.

Metamodelling with datalog and classes: conceptbase at the age of 21

ConceptBase is a deductive object-oriented database system intended for the management of metadata. A distinguishing feature of the Telos language underlying ConceptBase is the ability to manage rules and constraints across multiple levels of instantiation in so-called meta formulas, thus offering uniform consistency management across heterogeneous notations or ontologies. Originally developed in the context of model-driven database design in the late 1980s, ConceptBase has been used in several thousand installations all over the world for numerous applications in areas such as requirements engineering, engineering information management, model management, eLearning, cultural information systems, and data warehousing. The internal representation is based on a quadruple object structure, combined with advanced Datalog engines, such that many optimization techniques in ConceptBase have pioneered ideas later pursued in the implementation of XML databases and ontology-based reasoning and data management engines.

Tool Integration in Evolving Information Systems Environments

Evolution is a fact of life in information systems. Not only systems evolve but also their development processes. IS environments must therefore be designed for accommodating and managing change. The management of process meta models in repositories is one important step; we show how process traceability models and process guidance models can be developed and related in a standard repository framework. In addition, the currently available tool integration along the presentation, data, and control perspectives have to be augmented for process integration. In our process-adaptable and interoperable tool concept, tool behavior is directly influenced by the process guidance model and automatically traced according to the traceability model. The approach is demonstrated with a prototype requirements engineering environment developed in ESPRIT project NATURE.

Evolution in Domain Model-Based Requirements Engineering for Control Systems Development

When developing software-based control systems, knowledge and experiences in the relevant domain are of great importance. Small- and medium-sized enterprises (SMEs) that are most active here need to capture requirements under severe time and costs pressures. In previous work we have shown that a domain model based on the requirements formalism i* accelerates the requirements capture. Furthermore, the domain model-based similarity search supports the detection of reusable components from earlier projects. But due to the innovativeness, flexibility, and customer-orientation of control systems development, this domain model is subject to continuous change. Within this paper, we investigate the effects of model evolution on our domain model-based requirements engineering approach. Building on examples from industrial practice, we develop a classification of possible domain model modifications. For each such class, we analyze its impact on the similarity search and derive appropriate counter measures to limit these harmful impacts.

Requirements Engineering Repositories: Formal Support for Informal Teamwork Methods

Relationships among different modeling perspectives have been systematically investigated focusing either on given notations (e.g. OMT) or on domain reference models (e.g. SAP). In contrast, many successful informal methods for business analysis and requirements engineering (e.g. JAD) emphasize team negotiation, goal orientation and flexibility of modeling notations. This paper addresses the question how much formal and computerized support can be provided in such settings without destroying their creative tenor. Our solution comprises four components:

Heterogeneity in Model Management: A Meta Modeling Approach

As models are always abstractions of reality, we often need multiple modeling perspectives for analysis. The interplay of such modeling perspectives can take many forms and plays a role both at the design level, and during the operation of information systems. Examples include viewpoint resolution in requirements management, mapping between conceptual and implementation design in databases, and the integration or interoperation of multiple data and media sources. Starting from early experiences with our now 20-year old ConceptBase implementation of the Telos language, we describe a logic-based conceptual modeling and model management approach to these issues, focusing on recent work which employs a generic meta model to facilitate mappings and transformations between heterogeneous model representations both at the schema and the data level.