UML 2.0 in a Nutshell: A Desktop Quick Reference (In a Nutshell (O'Reilly)) [Englisch] [Taschenbuch]

This is the book that UML In a Nutshell should have been.

Several years ago I picked up the original UML In a Nutshell with high hopes; I didn't bother reading much of it in the bookstore because (after all) the book was an O'Reilly. It had an animal on the cover; quality was assured. I snapped it up and went home. It turned out to be the one of the worst computer books I'd ever bought, and many of the Amazon reviews agreed with me. I wrote my own scathing (but rather funny) review, and to be honest I don't know what happened to the book itself, I no longer cared.

A few weeks ago I received email from an editor at O'Reilly asking if I was the person who had posted that review, and would I be interested in a copy of their re-written book on UML 2.0? The book arrived a few days ago, and I've spent a couple hours going through it. (In the interest of disclosure, please note that I did not pay for my copy).

To put it mildly, UML 2.0 In a Nutshell is a vast improvement. I don't know how to emphasize this: It's like waking up from a bad nightmare of Throgzaks-are-after-you (and of course, you can't run) to realize that everything is okay and it was just the cat sleeping on your face. It is a huge relief that O'Reilly recognized their error and decided to fix it.

This book is smaller, more succinct and to the point. The authors dive into meaty subject material right away, starting with the stuff that most engineers are likely to use. The writing is pleasantly conversational, targetted to a technical (rather than a managerial) audience, and the subject matter is well organized. A challenge in grokking the UML is that it is a "wad" of interrelated concepts, and the book has sufficient forward references ("You'll learn more about X in chapter 4") that I felt comfortable just forging ahead. The diagrams are clear and meaningful, and there is (gasp) actual humor from time to time.

I get the impression that O'Reilly's first UML book was published in a hurry just to "get something out there." I have the impression that they took their time getting this one right. There is almost no comparison between the two; this is the book to get. Recommended.

I approached this book with some trepidation. I did not want to get into a sales pitch about the merits of one modeling tool over others. It quickly became obvious that this book is not about tools. In fact, the opposite it true. This book is truly about the UML. While there are sparse references to some tools, the text focuses on the UML as a standard and how to effectively and pragmatically apply it to your efforts.

Another concern I had when starting this book was a strict adherence to the UML. Much to my pleasure, this book takes a very pragmatic approach to modeling software systems. There are often statements indicating how "many designers do it" as opposed to the more formal approach. These situations show how making the UML work for you (as opposed to you working for the UML) does not cause any lack of clarity. In fact, it often adds to clarity and simplicity.

I really appreciate the way in which the text suggests approaching adoption and use of the UML. It would be difficult to try and quickly learn and apply all of the details, of all of the diagram types, and which arrows connect what shapes. The book addresses this by suggesting that readers adopt the UML in pieces. It also suggests that not every diagram type is needed for every situation. Once again, the text emphasizes a practical approach.

Although it would seem difficult to describe the graphical nature of the UML in text, the author does this quite adeptly. There is an excellent balance between figures and text. Examples are direct and meaningful. Also, the author does not dwell on how to model a software system. Instead, the focus is on how to use the UML as a modeling tool.

In addition to the UML-centric chapters, the book offers additional information related to software modeling. The first chapter provides an overview of the UML. This gives readers, especially those new to the UML, a nice foundation of vocabulary and purpose. The last chapter, "Effective Diagramming," provides readers with some solid guidelines on using the UML effectively. Here the author explicitly describes "appropriate" modeling techniques (Dare I say best practices?).

Appendix A provides an introduction to Model Driven Architecture (MDA). It really is a brief introduction that should kindle the reader's desire for more information. If you are feeling ambitious, you can also learn about the Object Constraint Language (OCL) described in Appendix B. The OCL is used in addition to the UML for more granular detail in describing constraints in UML models.

Overall I really enjoyed this book. I highly recommend it to people trying to learn the UML for the first as well as those wanting to know what has changed since previous versions of the UML. I recommend this book to both veterans and to those new to the UML alike because of how it is written. Both groups will find the book's pragmatic approach to using the UML quickly educational and beneficial as an on-going reference.

When searching for a very good UML reference book last year, I happened upon the book entitled "UML 2.0 in a Nutshell" by Dan Pilone with Neil Pitman. The book, which measures a mere 8.9 by 6 by 0.8 inches, is both lightweight and highly portable; which is one of the reasons that I decided to purchase a copy. However, it was ultimately the content of the book, and not its compact size, that convinced to me that this would a very useful resource. Condensed within 216 pages, "UML 2.0 in a Nutshell" lives up to its title, as the book is an extremely informative resource in understanding the various graphical elements that comprise UML with its nine types of diagrams.

The book's 12 chapters and two appendices are divided into four main parts: an introduction, static diagrams, behavioral modeling diagrams and finally, extensions and applications of UML.

First Part: Introduction

chapter 1: Fundamentals of UML

This chapter provides a short, but good introduction to the fundamentals of UML. If you have never used UML before, this will help to introduce several key concepts of UML; but you might want to consider purchasing a UML tutorial book, such as "UML Weekend Crash Course" by Thomas A. Pender, to obtain a more hands-on approach to learning UML.

Second Part: Static Diagrams

Chapter 2: Class Diagrams

Class diagrams are one of the most important aspects of UML. With class diagrams, the relationships between classes can be thoroughly illustrated, including the strengths of the relationships between classes. This chapter provides a precise description of the various ways that class relationships can be defined within UML: dependencies, associations, aggregations, compositions and generalizations; as well as association classes. This chapter also discusses class members (variables & methods), whether a class might be abstract or an interface, and templates.

Chapter 3: Package Diagrams

Classes that are contained within a common package can be illustrated within UML using package diagrams. This chapter also includes relationships between packages and use case packages.

Chapter 4: Composite Structures

This chapter discusses composite structures that exist during runtime, including connectors and ports, as well as collaborations.

Chapter 5: Component Diagrams

Components (replaceable & executable pieces of a larger system whose implementations are usually hidden) can be used in UML as either a black-box or white-box view. This chapter discusses both uses.

Chapter 6: Deployment Diagrams

This chapter discusses how the deployment of an application (which may include many pieces) can be illustrated within UML, including artifact instances, manifestations, nodes, devices, execution environments and communication paths.

Third Part: Behavioral Modeling Diagrams

Chapter 7: Use Case Diagrams

This chapter documents how an actor (a person or another application) interacts with applications and their internal components.

Chapter 8: Statechart Diagrams

This chapter discusses the two types of state machines that can be described in UML: behavioral state machines and protocol state machines. This includes states, composite states, submachine states, transitions, activities, pseudo-states and event processing.

Chapter 9: Activity Diagrams

This chapter discusses how activities and actions are illustrated within UML, including activity edges, activity nodes, object nodes, control nodes and more advanced activity modeling: activity partitions, exception handling, expansion regions, looping, streaming, interruptible activities and data store nodes.

Chapter 10: Interaction Diagrams

How UML is able to illustrate interactions between objects is accomplished through interaction diagrams. This chapter discusses usage of interaction participants, messages, execution occurrences, state invariants, event occurrences, traces, combined fragments, interaction occurrences, decompositions, continuations, sequence timing, timing diagrams and communication diagrams.

Fourth Part: UML Extensions & Applications

Chapter 11: Tagged Values, Stereotypes and UML Profiles

This chapter discusses use of stereotypes, tagged values, constraints and UML profiles.

Chapter 12: Effective Diagramming

This chapter is essentially a "do" and "don't do" chapter that emphasizes the need to keep UML diagrams as simple as possible, though that is not always possible.

The two appendices: Appendix A is about MDA (Model-Driven Architecture) and Appendix B is about the object constraint language (OCL), which is an extension of UML 2.0.

Overall, I rate "UML 2.0 in a Nutshell" with 5 out of 5 stars and highly recommend it to anyone learning and/or using UML on a regular or infrequent basis. I have yet to see a book as well written as this in explaining the many aspects of UML in as a concise & easy-to-understand form as "UML 2.0 in a Nutshell".