Code Leader: Using People, Tools, and Processes to Build Successful Software

This book is for the career developer who wants to take his or her skill set and/or project to the next level. If you are a professional software developer with 3–4 years of experience looking to bring a higher level of discipline to your project, or to learn the skills that will help you transition from software engineer to technical lead, then this book is for you. The topics covered in this book will help you focus on delivering software at a higher quality and lower cost. The book is about practical techniques and practices that will help you and your team realize those goals.

This book is for the developer understands that the business of software is, first and foremost, business. Writing code is fun, but writing high-quality code on time and at the lowest possible cost is what makes a software project successful. A team lead or architect who wants to succeed must keep that in mind.

Given that target audience, this book assumes a certain level of skill at reading code in one or more languages, and basic familiarity with building and testing software projects. It also assumes that you have at least a basic understanding of the software development lifecycle, and how requirements from customers become testable software projects.

Who This Book Is Not For: This is not a book for the entry-level developer fresh out of college, or for those just getting started as professional coders. It isn’t a book about writing code; it’s a book about how we write code together while keeping quality up and costs down. It is not for those who want to learn to write more efficient or literate code. There are plenty of other books available on those subjects, as mentioned previously.

This is also not a book about project management or development methodology. All of the strategies and techniques presented here are just as applicable to waterfall projects as they are to those employing Agile methodologies. While certain strategies such as Test-Driven Development and Continuous Integration have risen to popularity hand in hand with Agile development methodologies, there is no coupling between them. There are plenty of projects run using SCRUM that do not use TDD, and there are just as many waterfall projects that do.

Philosophy versus Practicality: There are a lot of religious arguments in software development. Exceptions versus result codes, strongly typed versus dynamic languages, and where to put your curly braces are just a few examples. This book tried to steer clear of those arguments here. Most of the chapters in this book deal with practical steps that you as a developer can take to improve your skills and improve the state of your project. The author makes no claims that these practices represent the way to write software. They represent strategies that have worked well for the author and other developers that he have worked closely with.

Philosophy certainly has its place in software development. Much of the current thinking in project management has been influenced by the Agile philosophy, for example. The next wave may be influenced by the Lean methodologies developed by Toyota for building automobiles. Because it represents a philosophy, the Lean process model can be applied to building software just as easily as to building cars. On the other hand, because they exist at the philosophical level, such methodologies can be difficult to conceptualize. The book tries to favor the practical over the philosophical, the concrete over the theoretical. This should be the kind of book that you can pick up, read one chapter of, and go away with some practical changes you can make to your software project that will make it better.

That said, the first part of this book is entitled “Philosophy” because the strategies described in it represent ways of approaching a problem rather than a specific solution. There are just as many practical ways to do Test-Driven Development as there are ways to manage a software project. You will have to pick the way that fits your chosen programming language, environment, and team structure. The book has tried to describe some tangible ways of realizing TDD, but it remains an abstract ideal rather than a one-size-fits-all technical solution. The same applies to Continuous Integration. There are numerous ways of thinking about and achieving a Continuous Integration solution, and this book presents only a few. Continuous Integration represents a way of thinking about your development process rather than a concrete or specific technique.

The second and third parts represent more concrete process and construction techniques that can improve your code and your project. They focus on the pragmatic rather than the philosophical.

Every Little Bit Helps: You do not have to sit down and read this book from cover to cover. While there are interrelationships between the chapters, each chapter can also stand on its own. If you know that you have a particular problem such as error handling with your current project, read that chapter and try to implement some of the suggestions in it. Don’t feel that you have to overhaul your entire software project at once. The various techniques described in this book can all incrementally improve a project one at a time.

If you are starting a brand new project and have an opportunity to define its structure, then by all means read the whole book and see how it influences the way you design your project. If you have to work within an existing project structure, you might have more success applying a few improvements at a time.

In terms of personal career growth, the same applies. Every new technique you learn makes you a better developer, so take them one at a time as your schedule and projects allow.

Examples: Most of the examples in this book are written in C#. However, the techniques described in this book apply just as well to any other modern programming language with a little translation. Even if you are unfamiliar with the inner workings or details of C# as a language, the examples are very small and simple to understand. Again, this is not a book about how to write code, and the examples in it are all intended to illustrate a specific point, not to become a part of your software project in any literal sense.

This book is organized into three sections, Philosophy, Process and Code Construction. The following is a short summary of what you will find in each section and chapter.

Part I (Philosophy) contains chapters that focus on abstract ideas about how to approach a software project. Each chapter contains practical examples of how to realize those ideas.

Chapter 1 (Buy, not Build) describes how to go about deciding which parts of your software project you need to write yourself and which parts you may be able to purchase or otherwise leverage from someplace else. In order to keep costs down and focus on your real competitive advantage, it is necessary to write only those parts of your application that you really need to.

Chapter 2 (Test-Driven Development) examines the Test-Driven Development (or Test-Driven Design) philosophy and some practical ways of applying it to your development lifecycle to produce higher-quality code in less time.

Chapter 3 (Continuous Integration) explores the Continuous Integration philosophy and how you can apply it to your project. CI involves automating your build and unit testing processes to give developers a shorter feedback cycle about changes that they make to the project. A shorter feedback cycle makes it easier for developers to work together as a team and at a higher level of productivity.

The chapters in Part II (Process) explore processes and tools that you can use as a team to improve the quality of your source code and make it easier to understand and to maintain.

Chapter 4 (Done Is Done) contains suggestions for defining what it means for a developer to “finish” a development task. Creating a “done is done” policy for your team can make it easier for developers to work together, and easier for developers and testers to work together. If everyone on your team follows the same set of steps to complete each task, then development will be more predictable and of a higher quality.

Chapter 5 (Testing) presents some concrete suggestions for how to create tests, how to run them, and how to organize them to make them easier to run, easier to measure, and more useful to developers and to testers. Included are sections on what code coverage means and how to measure it effectively, how to organize your tests by type, and how to automate your testing processes to get the most benefit from them.

Chapter 6 (Source Control) explains techniques for using your source control system more effectively so that it is easier for developers to work together on the same project, and easier to correlate changes in source control with physical software binaries and with defect or issue reports in your tracking system.

Chapter 7 (Static Analysis) examines what static analysis is, what information it can provide, and how it can improve the quality and maintainability of your projects.

Part III (Code Construction) includes chapters on specific coding techniques that can improve the quality and maintainability of your software projects.

Chapter 8 (Contract, Contract, Contract!) tackles programming by contract and how that can make your code easier for developers to understand and to use. Programming by contract can also make your application easier (and therefore less expensive) to maintain and support.

Chapter 9 (Limiting Dependencies) focuses on techniques for limiting how dependent each part of your application is upon the others. Limiting dependencies can lead to software that is easier to make changes to and cheaper to maintain as well as easier to deploy and test.

Chapter 10 (The Model-View-Presenter Model) offers a brief descr...

這本書適合那些想要提升自己技能或專案水平的職業開發者。如果你是一位有3-4年經驗的專業軟體開發者，希望在專案中引入更高層次的紀律，或者想要學習技能以從軟體工程師轉變為技術領導者，那麼這本書就是為你而寫的。本書涵蓋的主題將幫助你專注於以更高的品質和更低的成本交付軟體。本書介紹的實用技巧和實踐方法將幫助你和你的團隊實現這些目標。

這本書適合那些理解軟體業務首要是商業的開發者。編寫程式碼很有趣，但是按時按成本編寫高品質的程式碼才是使軟體專案成功的關鍵。一位希望成功的團隊領導或架構師必須謹記這一點。

鑑於目標讀者群，本書假設讀者具備閱讀一種或多種程式語言的程式碼的能力，並對構建和測試軟體專案有基本的熟悉。它還假設你至少基本了解軟體開發生命週期以及客戶需求如何變成可測試的軟體專案。

這本書不適合剛從大學畢業的初級開發者，也不適合剛開始從事專業編碼工作的人。這不是一本關於編寫程式碼的書；這是一本關於我們如何在保持品質和降低成本的同時共同編寫程式碼的書。它不適合那些想要學習如何編寫更高效或更有文采的程式碼的人。關於這些主題已經有很多其他書籍可供參考。

這也不是一本關於專案管理或開發方法論的書。這裡介紹的所有策略和技巧同樣適用於瀑布式專案和使用敏捷方法論的專案。雖然一些策略（如測試驅動開發和持續集成）與敏捷開發方法論一起興起，但它們之間並沒有必然聯繫。有很多使用SCRUM方法但不使用TDD的專案，同樣也有很多瀑布式專案使用TDD。

哲學與實用性：軟體開發中存在許多宗教性的爭論，例如異常處理與結果碼、強型別與動態語言，以及大括號的放置位置等。本書試圖避免這些爭論。本書的大部分章節涉及開發者可以採取的實用步驟，以提升自己的技能和改善專案狀態。作者並不聲稱這些實踐方法代表著編寫軟體的唯一方式。它們代表著作者和其他開發者在實際工作中取得良好效果的策略。

哲學在軟體開發中當然有其重要性。例如，敏捷哲學對專案管理的思考方式有很大影響。下一波可能受到豐田汽車開發汽車時所使用的精益方法論的影響。因為它代表著一種哲學，所以精益流程模型同樣適用於軟體開發，就像適用於汽車製造一樣。然而，由於它們存在於哲學層面，這些方法論可能難以理解。本書傾向於實用性而非哲學性，偏向具體而非理論性。這本書應該是你可以拿起來讀一章並從中獲得一些實用改變的書籍，這些改變可以使你的軟體專案變得更好。

話雖如此，本書的第一部分名為「哲學」，因為其中描述的策略代表了解決問題的方式，而不是具體的解決方案。有很多實用的方法可以進行測試驅動的編碼。