Agile!: The Good, the Hype and the Ugly

The quote he attributes to Samuel Johnson on page 12 summarizes the agile movement precisely: “Your work, Sir, is both new and good, but what is new is not good and what is good is not new”Interestingly, later in the chapter under a section called “New and good” - the supposed agile innovations he listed were again all good but not new, we have been doing them in aerospace for 30 years. He mostly was tossing agile a bone I guess for at least mainstreaming some good ideas that the better software teams in industry were already doing

I come from a background of serious, large-scale software engineering, that uses standards (mainly IEEE), proper design and iterative development; with a heavy emphasis on re-use and extensibility. The first time I heard about 'pair programming', 'code is all that matters' and how the right design will emerge by constant code refactoring I mentally wrote off anything agile as a lunatic fantasy.In the following decade I slowly became aware of useful things like continuous build, Scrum sprints, standups, and TDD, and realised there might be something useful in it after all. In the end there is: the agile movement has provided better ways of solving the 'how' of incremental development than traditional software engineering did. It still does contain many completely lunatic ideas that will harm any medium to large-scale development. But how to figure out which is which?This book does exactly what is needed - taking a sieve to the dirt and finding the diamonds. It assesses all the agile practices in a way that appears very balanced to me. Meyer, coming from a quite different background than myself, finds that the horror of 'big upfront anything', documentation and other such artefacts is probably dangerous to your project (it is), while the 1 month Scrum sprint is one of the best ideas in recent times.He analyses pretty well everything, including all the main methods (including Scrum, XP, Crystal) and pulls useful small ideas even when the main idea is obviously ridiculous (noone would seriously run a project with all developers doing 'pair programming' for example; if it happened to be useful for a few devs some of the time, that's their business of course, but routinely applied? Sorry, normal people like to concentrate on their work!).Being uninterested in most mainstream IT publications these days, I suspect this might be the only Agile book I ever read, which is an irony in itself. If you are similarly lazy, but need to catch up on Agile, get this book. It's short as well.Note to Amazon: why is this book substantially cheaper on Amazon US than Amazon UK?

As a former senior software engineer, I have watched Agile destroy solid software engineering practices in every company I've worked for, both as a contractor and employee. Many of the points I made to them over the years, and some of the reasoning, has now been validated by someone who is going to be much less vulnerable to the type of character assassination and intimidation I faced. Professor Meyer systematically separates the good from bad, the hype from the real, and shares the results of his research and experience with the reader. He begins by taking the Agile manifesto to task, first by defining terms, then pointing out the false reasoning, change of context, and "guilt by association" used by the founders to develop the three major agile approaches over the years. He pulls no punches -- but he also enthusiastically endorses Agile's positive contributions to software engineering. He points those out as well, much better than Agile's founders in some cases. The author points out the risks of using a "pick and choose" methodology as it applies to software engineering practices in general, and Agile in particular. He also describes in detail where Agile falls short of providing the guidance it should when it states a technique is no longer valid. He backs up his claims with time proven engineering practices, debunks some of Agile's false and/or incomplete claims, and provides historical background to reinforce his arguments. Of special importance, the author provides guidelines, "fills in the blanks", and points out what to avoid and how, so each of us can engineer the best software at the lowest *reasonable* cost to the companies we work for. For those of you who are members of a certain professional computer association (which I cannot mention by name in this review), you should have access to a one-hour webinar with the same title by Professor Meyer. I encourage you to watch it. Then, you'll *want* to buy this book!

a very solid, brief overview of all agile.

This book is exactly what its author appears to have intended it to be--an objective, well-reasoned analysis of both the good and bad of 'Agile' as an approach to producing professional software. Bertrand Meyer is not a methodologist and doesn't have any major axes to grind here--what he is, however, is a pioneer and world authority in the area of OO language and software construction--so much so that not knowing his name and work would be cause for doubt about any OO software programming job applicant. Accordingly, his standard is always goal oriented: will doing 'x' help or hurt the effort to build solid, maintainable software in an efficient, reliable manner? If that's your question (rather than interest in methodology-as-religious-war), then this is your book. Meyer neither dismisses agile nor canonizes it--there are no easy answers, and no sense that any methodology is perfect or always appropriate. But in attempting to run agile, transition to it, or simply borrow from it, a development manager would be well-advised to have Meyer's analysis and terrain map at hand when making decisions about development practices.

This book has the great virtue of being sceptical about agile, while acknowledging its good points. The author has some harsh things to say about agile stories as a substitute for requirements capture. The agile dislike for Big Upfront Design also comes in for some waspish words.It's quite expensive for a short book, but it is well written.

What a refreshing book this is. An effortless debunking of the Agile movement. I especially liked the identification of the tactics often used by the Agile community to argue their case. However the book is balanced in that it does pick out the aspects of Agile that are worthwhile such as the emphasis on testing. Even if you are an ardent Agileist I would strongly recommend buying this book.

The book covers the basics of agile methodologies and provides good practical advice on implementing agile within the work place.IMHO it is balenced and easy to read.Bertrand has also done two webinears with the ACM on the contents of the book.

Insightful criticism of "agile" in its commercial practice. You may agree or not, but worth reading.

This book should be read by committed agile practitioners and newcomers alike. It identifies excellent aspects of agile software development as well significant weaknesses, thus equipping programmers to take advantage of the best that agile has to offer.