Date: Mon, 26 Jul 93 10:08:09 BST To: re-world@doc.ic.ac.uk From: acwf@doc.ic.ac.uk (Anthony Finkelstein) X-Sender: acwf@gummo.doc.ic.ac.uk Subject: REQUIREMENTS ENGINEERING NEWSLETTER (8) ****************REQUIREMENTS ENGINEERING NEWSLETTER******************** No. 8. 1. Requirements for Complex Systems Design (Gary Reeman) 2. Univ. of Toronto RE reports (Lawrence Chung) 3. Press release on CRIS-94 (Bill Olle) Contents Contributions to: re-list@doc.ic.ac.uk (will be moderated) Subscription or Removal to: re-request@doc.ic.ac.uk Back issues can be obtained via anonymous ftp from ftp-host: dse.doc.ic.ac.uk (IP number: 146.169.2.20). Directory: requirements. Files are called renl1, renl2, etc. If you cannot use ftp then you can get any back issues using email. Send email containing the following to ftpmail@doc.ic.ac.uk open dse.doc.ic.ac.uk cd requirements get quit ********************************************************************** Subject: Requirements for Complex Systems Design From: Gary.Reeman@brunel.ac.uk (Gary Reeman) PRELIMINARY ANNOUNCEMENT REQUIREMENTS FOR COMPLEX SYSTEMS DESIGN Kensington Park Hotel, London Tuesday, November 2nd 1993 * Are you concerned about the failure of information systems to meet the needs of complex work groups? * Do you regard insights into the interactions between users as fundamental to successful CSCW design? * Do you need to design systems that satisfy the usability requirements of groups as well as those of individual users? If your answer to any of these questions is yes you may wish to learn how to apply an approach known as "ethnography". This is a method of participant observation, drawn from the social sciences, that has been exploited to great effect by innovative companies such as Rank Xerox. CRICT, responding to the demand expressed by participants at recent conferences and seminars, has decided to bring together some of the UK's most experienced ethnographers working in the fields of requirements analysis and systems design to teach the basic principles of ethnography. These tutors will work through a series of case studies and practical exercises with the participants. On completing the course the participants will be able to: * Understand the need to analyse the social setting in which work takes place when considering information systems requirements. * Appreciate the advantages of ethnography compared with other techniques for analysing the social character of the workplace (e.g. surveys, laboratory experiments). * Utilise ethnography to expand the description of the working environment obtained from existing techniques. * Determine the implications for requirements based on an analysis of ethnographic data. The event will be held at the prestigious Kensington Park Hotel in London. Target Audience Human factors specialists Requirements engineers CSCW designers Managers responsible for requirements and usability For further details please contact: GARY M. REEMAN CRICT BRUNEL UNIVERSITY UXBRIDGE MIDDX UB8 3PH TEL: 0895 203123 FAX: 0895 203155 EMAIL: GARY.REEMAN@BRUNEL.AC.UK ********************************************************************** From: "K. Lawrence Chung" Subject: Univ. of Toronto RE reports Technical Reports on Research in Data and Knowledge Base Systems ------------------------------------------------------------------------------- Representing and Using Non-Functional Requirements: A Process-Oriented Approach by Lawrence Chung Technical Report DKBS-TR-93-1 (Ph.D. Thesis) June 1993 (xii + 137 pages) ------------------------------------------------------------------------------- Requirements Engineering 1993: Selected Papers by John Mylopoulos, Lawrence Chung, Eric Yu and Brian Nixon Technical Report DKBS-TR-93-2 July 1993 (86 pages) ------------------------------------------------------------------------------- Available from: DKBS Technical Report Secretary veronica@ai.toronto.edu 6 King's College Road, Room 283 Facsimile (1-416) 978-1455 Department of Computer Science University of Toronto Toronto, Ontario, Canada. M5S 1A4 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- Abstracts --------- Representing and Using Non-Functional Requirements: A Process-Oriented Approach by Lawrence Chung (e-mail: chung@ai.toronto.edu) Technical Report DKBS-TR-93-1 (Ph.D. Thesis) June 1993 (xii + 137 pages) The usefulness of a software system is determined partly by its functionality -- i.e., what the system does -- and partly by quality characteristics on its development or operational costs, accuracy, security, performance, reliability, maintainability, and the like. Although these quality characteristics depend on the process that leads from abstract non-functional, or quality, requirements of a software system to its design, there has been no systematic approach to dealing with such requirements during software development. This thesis proposes a comprehensive framework (referred in the thesis as the Non-Functional Requirements framework, or NFR framework) for representing and using non-functional requirements during the development process. The NFR framework consists of five basic components which provide for the representation of non-functional requirements in terms of interrelated goals. Such goals can be refined through methods, while goal synergy or conflict can be detected through correlation rules. Goals can also be evaluated in order to determine the degree to which a set of non-functional requirements is supported by a particular design. Evidence for the power of the proposed framework is provided firstly through its application to accuracy and security requirements. These applications demonstrate, in particular, that the framework meets the need for capturing various types of available development knowledge in dealing with and analyzing accuracy and security requirements. Additional evidence is provided through the implementation of a tool, the NFR assistant which realizes the NFR framework and assists a developer in using the framework. This tool is actually used to develop portions of both a hypothetical research expense management system and a real credit card system. Third evidence is provided through application of the NFR framework to small portions of three different kinds of information systems, namely, credit card, health insurance, and government document systems. In particular, these applications re-confirm the usefulness of the framework with respect to the analysis and detection of faults, such as ambiguity, omission, synergy, and conflict. ============================================================================== Requirements Engineering 1993: Selected Papers by John Mylopoulos, Lawrence Chung, Eric Yu and Brian Nixon (e-mail: {jm,chung,eric,nixon}@ai.toronto.edu) Technical Report DKBS-TR-93-2 July 1993 (86 pages) For any computer system to be fit-to-use, it is essential to have a clear understanding of its expected functionality and quality as well as the objectives and structures of its environment in which it is to perform its functions. In our view, such an understanding is a vital component of a recently-emerging area of software engineering research and practice, namely _Requirements Engineering_. Requirements Engineering activities range from elicitation of ill-defined descriptions to formal representation and utilization of knowledge about the system and the overall organization. In this light, this report collects some recent work on Requirements Engineering from the University of Toronto. The papers, some of which appear here in slightly revised form, are grouped into two sections: 1. Non-Functional Requirements (3 papers) 2. Organization Modelling (3 papers) The first section deals with Non-Functional Requirements (NFRs). The usefulness of a software system is determined partly by its functionality -- i.e., what the system does -- and partly by global requirements on its development or operational costs, accuracy, performance, security, reliability, maintainability, portability, robustness and the like. These _non-functional_ (or _quality_) _requirements_ play a crucial role during system development. The papers in this section present a framework for dealing with non-functional requirements and its application to performance and security requirements engineering. The first paper proposes a comprehensive framework for representing and using non-functional requirements during the development process. This framework provides for the representation of non-functional requirements in terms of interrelated goals. Such goals can be refined through refinement methods and can be evaluated in order to determine the degree to which a set of non-functional requirements is supported by a particular design. Evidence for the power of the framework is provided through the study of accuracy and performance requirements for information systems, as applied to an expense management system and a credit card system. The next paper focusses on the application of the NFR framework to performance requirements for information systems. This involves representing a variety of implementation alternatives, expressing performance goals in terms of performance and information system concepts, and organizing the process into layers corresponding to subsets of the language to be implemented. This is illustrated by attempting to meet a set of goals such as ``achieve good time performance for authorizing credit card sales'' while mapping the conceptual design of an information system to its implementation. By treating security requirements as a class of NFRs, the third paper further applies the NFR framework to developing secure information systems. The paper identifies a variety of generic development knowledge and goal interactions, and shows how the application allows for the systematic capture and reuse of generic development knowledge. The paper also shows that security requirements serve as a class of criteria for selecting among design decisions, and justify the overall design. This paper also describes a prototype development tool, and illustrates it using a credit card system example. The second section deals with Organization Modelling. Most information systems operate in some kind of organizational environment. Understanding the needs of an organization is thus an important part of information system requirements engineering. The papers in this section present elements of a framework for representing and using knowledge about organizations in requirements engineering. The framework provides modelling features to represent the structure of an organization (including information systems as elements), as well as issues and arguments that are used to analyse current structures and to design new ones. The first paper in this section proposes to model organizational structure as a network of dependencies among organizational actors. Actors are taken to depend on each other for goals to be achieved, tasks to be performed, and resources to be furnished. These dependencies are formalized in terms of intentional operators such as belief, goal, ability, and commitment. An intentional model of organization structure goes beyond capturing _what_ actors do, by also _why_ they do them. Examples are drawn from the domain of engineering project organizations. The second paper elaborates on the Actor Dependency Model, with a more detailed informal description of the types of dependencies and how they can be used to distinguish among different organizational configurations. The ability of the model to answer ``Why?'' and ``What-If?'' questions are illustrated with examples from business process reengineering, in the domains of goods acquisition and insurance application processing. In the third paper, the Issue Argumentation Model is introduced to represent the issues and concerns that organizational actors (or stakeholders) have about organizational configurations, and which lead them to choose one alternative over another, e.g., how an information system is to be embedded in the network of organizational relationships. This model is based on the NFR framework that is the topic of section one of this technical report. A third modelling component, the Agents-Roles-Positions Model, adds flexibility in modelling the make-up of organizational actors. This paper highlights the ability of this organizational modelling framework to accommodate a multi-perspective approach to information system development. In particular, it allows issues and concerns from technical as well as social science perspectives to be taken into account within one systematic framework for analysis and design. The modelling features of the framework are explained using examples from a hospital setting. A more detailed example, based on a published ethnographic study of an actual organization, is used to show how the separate components of the framework work together. ********************************************************************** From: Bill Olle <100010.3176@CompuServe.COM> Subject: Press release on CRIS-94 PRESS RELEASE IFIP WORKING GROUP 8.1 CONFERENCE IN THE CRIS SERIES METHODS & ASSOCIATED TOOLS FOR THE INFORMATION SYSTEMS LIFE CYCLE 26 - 28 September 1994 Maastricht, Netherlands CALL FOR PAPERS Themes Following on the four successful conferences held in 1982, 1983, 1986 and 1988, CRIS 94 will also concentrate on the features and applications of methods (now regarded as a more correct term than methodologies). Methods have grown in acceptance since the earlier CRIS conferences and there has been considerable conver- gence in the basic techniques used by practitioners. CASE tools have continued to proliferate and there has been an emergence of tools which generate an operational system. There are two impor- tant areas in which both methods and CASE tools are lacking, namely the area of distributed systems and the area of access control. It is the aim of CRIS 94 to provide a platform for exploring the state of the art, the state of practice, the state of research and the state of harmonization work. With this in mind, the following major themes are identified within the overall area of methods and tools. Papers are solicited which can be keyed to one of these themes. Papers in the general field of method and tools which cannot be keyed to one of these theme may also be submitted. Theme A: Comparative evaluation based on methods and associated tools. What is the latest thinking on evaluating a method? What are the current criteria for acceptability by practitioners? To what extent is the acceptability of a method predicated on the way in which the associated tool is engineered? Are certain methods more suitable for specific application areas than for others? How can an application area be categorized in terms of its requirements for a method? Is there such a thing as a universally applicable method? Theme B: Standards and harmonization of methods and tools. Would a standard way of preparing the diagrams associated with certain widely used techniques be useful? Will the US Federal standard for a data modelling technique (IDEF1X) be accepted in the rest of the world? Is Euromethod the right approach? Are some ways of using Eurome- thod to be preferred? Will CDIF really help enterprises using several different tools? How does the ISO IRDS approach help? Which standards are needed for IRDS content modules? What are the advantages and disadvantages of standardizing an information system life cycle? How will the ISO SC7 project to define a standard model of software engineering help users? Theme C: The shortcomings in existing methods. What needs to be added to existing methods to handle the distri- bution aspect? What techniques are available for determining whether or not a system should be distributed? What technical and economic factors should a user of a method take into account when deciding whether to adopt a distributed processing system or a distributed database system? How can access control be factored in to existing methods? What techniques are available for analysing requirements for access control? Are the access control facilities in ISO SQL92 adequate for system designers? How can the "build or buy" decision be incorporated into existing methods? What techniques are avail- able for handling the later stages in the information systems life cycle such as testing, installation, evolution, migration and phase out? How do these integrate with earlier stages? Conference Organization: SERC (Software Engineering Research Centre) is the local sponsor for the conference. The University of Limburg is the local organizer. This conference is held under the aegis of IFIP Working Group 8.1, the title of which is "Design and Evaluation of Information Systems". The conference is the fifth in the series of CRIS conferences (CRIS was originally an acronym of Comparative Review of Information Systems Methodologies). The proceedings of CRIS 94 will be published by Elsevier - North Holland. Time table Letter of intent due: 30 Sept 1993 Paper due: 15 Jan 1994 Notification of acceptance: 15 Mar 1994 Camera ready copy due: 15 April 1994 General Conference Chair: Alex Verrijn-Stuart, University of Leiden, Netherlands Program Committee chair: Bill Olle, T.William Olle Associates, Ltd, Walton on Thames, Surrey, England Organizing Committee Chair: Jan Dietz, University of Limburg, Maastricht, Netherlands Conference Secretariat: All enquiries should be addressed to the Conference Secretariat mentioning CRIS 94. CRIS 94 Conference Secretariat Mrs. Mieke van Zutphen University of Limburg P.O. Box 616, 6200 AD Maastricht Netherlands Phone: +31-43-883656 Fax: +31-43-258495 Email: cris94@be.rulimburg.nl **********************************************************************