Review of book: Software Estimation Demystifying the Black Art

Author: Steve McConell

Publisher: Microsoft Press (2006)

ISBN: 978-0735605350

Paperback: 338 pages

Rating: 8/10

Contents

The content of the book is divided into three parts. The first one presents critical estimation concepts. You will learn what effort estimation is and why people often misunderstood it with commitments or plans. The second part presents a set of well-known software size and estimation techniques. you will find there expert-base methods, estimating by analogy, COCOMO, Proxy-based estimates, Wideband Delphi, Function Points and many more. The last part gives an overview of some additional issues regarding software effort estimation (e.g. what is the relation between effort and schedule).

Pros

Steve McConell is a briliant writer, and that's for sure. I would say it is a kind of book you can read during a single winter evening. Although the topic is difficult and complex, Steve has a real gift in presenting things in a simple and interesting way.

The composition of the book is also impressive. Steve starts from introduction to effort estimation techniques by focusing on its role in software projects. Presenting arguments you just can't deny, he will introduce you to the world of effort estimation, where objective analitical process of estimation is often mixed with subjectivity, commitments and plans. If you ever had a feeling that estimation is useless or your estimation never met real effort, you have to read this chapter for sure. The portfolio of sizing and estimation methods, presented in the second part, contains many well-known methods. Although in some cases presented description won't be sufficient to use the methods in practice, however it is a good starting point to look for more in-depth information elswere.

The book is intended to be read by practicioners e.g. you have lots of practical tips; each of them is numbered and presented separately from the paragraph text.

Cons

Actually it is quite difficult to say something wrong about this book. The only thing which comes to my head is that descriptions of the methods are superficial in many cases. On the other hand, I perceive this book as a great introduction to software effort estimation methods. That would justify the lack of in-depth information about some of the methods.

Summary

In my opinon it is a must-read book for anyone interested in software cost estimation. Beginners will find it as a non-painful way for introducing themselves to the topic, more advanced users will find the first part especially interesting from the practical point of view, where they can learn from Steve how to talk with people about effort estimation (and how to read the goal of estimation between the lines).