SoftwareDevelopment

UML Xtra-Light: How to Specify Your Software Requirements
$28.99
Effective communication between people is a very hard task, independent of the subject. When the information is imprecise and the languages are different, there is a lot of literal and figurative hand-waving and assumptions based on context. Many studies have shown that one of the major problems in software development is the difficulty that managers and developers have in communicating business needs and goals. The managers think in and speak manager-speak and the developers generally communicate in geek-speak. Since it is the responsibility of the managers to give the direction, it is reasonable to expect them to learn enough of the language of the developers to communicate the goals. The primary language of software development is now the Unified Modeling Language or UML.
UML is a large and complex language, but the fundamentals can be learned in a short time. The purpose of this book is to explain the basics of UML in a form that managers can understand. Presented in the style of a study of businesses, the words used are those of managers rather than developers. After reading the book, any manager will be much more effective in their ability to describe the purpose of the software that is to be built. However, developers with an interest in learning UML should look elsewhere.
Unclear and ambiguous software requirements cost the software industry billions of dollars every year. The one hundred pages of this book can help you substantially reduce your costs, provided you take the time and effort to understand the contents. In the new atmosphere of cost reductions, no manager of a software project can afford to ignore such a high level and rapid Return On Investment (ROI).

The Road to the Unified Software Development Process (SIGS Reference Library)
$55.00
The compilation of articles from the 90's does fit together, but its integrity is loose due to this structure. It spans genres (history, survey, reference, howto). Even though the second section is listed as history, most of the book can be considered history. This is because the articles range from the early to late the 90's. The mini interview sections after each article titled "In Ivar's Words" (not 'Inar' lol) often clarify or update the content of the article. This book is a decent mix of history, description, and clarification. I found the history and the somewhat conversational tone of the mini interviews most interesting.

Formal Specification and Design (Cambridge Tracts in Theoretical Computer Science)
$70.00
Formal specification is a method for precisely modelling computer-based systems that combines concepts from software engineering and mathematical logic. In this book the authors describe algebraic and state-based specification techniques from the unified view of the Common Object-oriented Language for Design, COLD, a wide-spectrum language in the tradition of VDM and Z. The kernel language is explained in detail, with many examples, including: set representation, a display device, an INGRES-like database system, and a line editor. Fundamental techniques such as initial algebra semantics, loose semantics, partial functions, hiding, sharing, predicate and dynamic logic, abstraction functions, representation of invariants and black-box correctness are also presented. More advanced ideas, for example Horn logic, and large systems are given in the final part. Appendices contain full details of the language's syntax and a specification library. Techniques for software development and design are emphasised throughout, so the book will be an excellent choice for courses in these areas.

UML by Example (SIGS: Advances in Object Technology)
$109.00
A great book, although a bit smaller then most UML books, this is one of those instances that quality definitely came out on top over quantity.

The first section reviews UML in general, what the symbols are and when and how to use them.

Section two is the real meat of the book, showing what every person wanting to learn programming always has a problem with, and usually turns out to be that turning point where they give up learning to program. Its not the learning of a language syntax and its nuances that is the real challenge of learning to program (although at times . . .). It is trying to come up with an application or object oriented script that fits a need you must address, and then the process of figuring out how you should build it from an OO point of view. Section two has 2 chapters. Each has an example of taking a high level idea and breaking it down using the "Bridge" method used in this book, until you finally have a complete UML class diagram. At which point you can then start your coding in the OO language of your choice.

Section three is more or less a repeat of section two only the answers and examples are not provided for you. Section 3 is a set of exercises. Concepts of applications (customer requirements doc) are provided and you do the UML designs.

Section 2 though is where all the real action is at.

I give this book 4 stars and not 5 because of two things:

I. An authors version of the answers to section 3 are not provided to you for comparison to the results you come up with after going through exercise.
II. The UML used in this book is still based on UML 1.x and not 2.x. This book is great but screams for a second edition written with UML 2.x in mind.

Brad