Craftsmanship is a return to the roots of software development: Good software developers have always understood that programming is a craft skill. Regardless of the amount of arcane and detailed technical knowledge that a person has, in the end, application development comes down to feel and experience. Someone can know all of the esoteric technical details of the Java programming language, but that person will never be able to master application development unless he or she develops a feel for the aesthetics of software. Conversely, once a person gets the feel for software development, the specific technical details become almost irrelevant. Great developers are always picking up and using new technology and techniques; learning a new technology is just a normal part of the life of a software developer.

The term software engineering was coined in 1967 by a NATO study group that recommended a conference to discuss “the problems of software.” The report from this 1968 conference, which was sponsored by the NATO Science Committee and took place in Garmish, Germany, was titled Software Engineering.1 In the report, Peter Naur and Brian Randell stated, “The phrase ‘software engineering’ was deliberately chosen to be provocative, in implying the need for software manufacture to be based on the types of theoretical foundations and practical disciplines that are traditional in the established branches of engineering.”

In the same spirit, it is the intention of this book to be deliberately provocative in implying the need for practitioners to start paying attention to the craft of software development. Software craftsmanship is important because it takes us away from the manufacturing metaphor that software engineering invokes and makes us pay attention to the people who do software development. Craftsmanship brings with it the metaphor of skilled practitioners intent on mastering their craft, of pride in and responsibility for, the fruits of their labor.

Software craftsmanship is not the opposite of software engineering or computer science. Rather, craftsmanship is a different tradition that happily coexists with and benefits from science and engineering. Just as the modern blacksmith benefits from better tools, materials, and understanding, so software craftsmanship benefits from better computers, reusable components, and programming languages. Just as blacksmiths transcend science and engineering with their skill and artistry, software craftsmanship can transcend computer science and software engineering to produce great programs, applications, and systems. UNIX and the modern-day GNU Linux are probably the best-known examples of this—systems that are thriving due to the craft, skill, and dedication of their creators.

Software craftsmanship is a response to the problems of trying to force-fit software engineering into commercial application development. Software engineering was developed to meet the needs of NATO in developing very large defense systems. Commercial application development differs from the development of defense and government systems in that applications are a whole lot smaller and normally have to be up and running in less than 18 months. It is rare for a commercial application to be developed by a team of more than 20 people, and most application developers work in teams with fewer than 10 members. Software engineering is good at handling the problems of really large teams of 200 or more people, but it has little to say about how the individuals in a team should practice their craft.

Software engineering encourages the “human wave” 2 approach to software development. Rather than solving the problem of how to develop highly skilled developers, software engineering attempts to deskill software development by suggesting that every problem can be solved by throwing more people at it.

Although this approach sometimes succeeds, the resulting software is junk. Slow and bloated, it just never feels right. Users are dazzled by the graphics and animation but never really manage to come to grips with the software. They are thwarted by their inability to learn the software and use only a small fraction of the available features.

Software does not have to be like that. All too often I see application development teams shipping valuable applications that provide real, measurable business benefit, but apologizing for not following software engineering best practices. For me, the real test of a team is whether it manages to ship and then enhance and extend the application for years afterward. Timely shipping of the first release is important, but it is more important that subsequent releases occur in a timely fashion and that each new release improves the application.

Whenever I’m asked about hiring developers, I tell people to look for developers who have shipped a few applications successfully and then stuck around long enough to handle the next enhancement or maintenance release. Shipping proves that the developer can make something work; staying around for the next release allows the developer to experience the effects of the way that he or she built the application in the first place. If a developer has done this three times, my guess is that he or she is skilled and experienced enough in the craft of software development to be successful again.

Software craftsmanship is the new imperative because many members of the software development community are starting to chase technology for its own sake, forgetting what is important. The purpose of software development is to create high-quality, robust software applications that deliver value to their users. What matters is growing a new generation of developers who can do that.

Software craftsmanship stands for putting the joy and excitement back into creating applications for our users.

1 Naur, Peter, and Brian Randell, (eds.), Software Engineering: A Report on a Conference Spnsored by the NATO Science Committee,NATO, 1969.

2 Levy, Steven, Hackers, Penguin Books, 1994, p. 88. 0201733862P08202001