How to Engineer Software

A guide to the application of the theory and practice of computing to develop and maintain software that economically solves real-world problem How to Engineer Software is a practical, how-to guide that explores the concepts and techniques of model-based software engineering using the Unified Modeling Language. The author--a noted expert on the topic--demonstrates how software can be developed and maintained under a true engineering discipline. He describes the relevant software engineering practices that are grounded in Computer Science and Discrete Mathematics. Model-based software engineering uses semantic modeling to reveal as many precise requirements as possible. This approach separates business complexities from technology complexities, and gives developers the most freedom in finding optimal designs and code. The book promotes development scalability through domain partitioning and subdomain partitioning. It also explores software documentation that specifically and intentionally adds value for development and maintenance. This important book: * Contains many illustrative examples of model-based software engineering, from semantic model all the way to executable code * Explains how to derive verification (acceptance) test cases from a semantic model * Describes project estimation, along with alternative software development and maintenance processes * Shows how to develop and maintain cost-effective software that solves real-world problems Written for graduate and undergraduate students in software engineering and professionals in the field, How to Engineer Software offers an introduction to applying the theory of computing with practice and judgment in order to economically develop and maintain software.

STEVE TOCKEY is Principal Consultant at Construx. Steve is an expert in software engineering practices including object-oriented development, distributed object computing, software project management, and software quality. He earned a Masters of Software Engineering degree from Seattle University and has served as an adjunct faculty member for the program. Prior to joining Construx, Tockey worked at Lawrence Livermore National Laboratory, Boeing, and Rockwell Collins.

Foreword xi Preface xvii Acknowledgments xxv Online Resources xxvii Part I Introduction and Foundations 1 1 Introduction 3 2 The Nature of Code 39 3 Fundamental Principles 67 4 Functional and Nonfunctional Requirements 91 5 UML Overview 115 6 Partitioning Systems into Domains 125 Part II Semantic Modeling: Model-based Functional Requirements 151 7 Use Case Diagrams: Scope and Context 153 8 Class Models: Policies to Enforce 183 9 Interaction Diagrams: Process at a Mid-Level 237 10 State Models: Process at a Fine-Grained Level 261 11 Partitioning Domains into Subdomains 305 12 Wrapping Up Semantic Modeling 323 Part III Model-based Design and Code 369 13 Introduction to Design and Code 371 14 Designing Interfaces: Specifying Real-World Interaction 379 15 High-Level Design: Classes and Operations 407 16 High-Level Design: Contracts and Signatures 447 17 Detailed Design and Code 503 18 Formal Disciplines of Design and Code 539 19 Optimization 583 20 Model Compilation 633 21 Advanced Open Model Compilation 675 22 Wrapping Up Model-Based Design and Code 705 Part IV Related Topics 723 23 Estimation 725 24 Development and Maintenance Processes 759 25 Economics of Error Handling 787 26 Arguments Against Model-Based Software Engineering 815 Part V Summary 827 27 Closing Remarks 829 Part VI Appendices 843 Appendix A: Documentation Principles 845 Appendix B: WebBooks 2.0 Background 849 Appendix C: WebBooks 2.0 Domains 853 Appendix D: Semantic Model for Order fulfillment 857 Appendix E: (Pro Forma) Order fulfillment Design 885 Appendix F: Semantic Model for Payment 905 Appendix G: (Pro Forma) Payment Design 927 Appendix H: Semantic Model for Scalability 943 Appendix I: (Pro Forma) Scalability Design 969 Appendix J: Semantic Model for High availability 985 Appendix K: (Pro Forma) High availability Design 1001 Appendix L: Semantics of Semantic Modeling 1011 Appendix M: Sample Production Rules 1049 Appendix N: Software Structural Complexity Metrics 1061 References 1081 Index 1091