The It Measurement Compendium Estimating And Benchmarking Success With Functional Size Measurement Manfred Bundschuh by Manfred Bundschuh, Carol Dekkers 9783540681885, 3540681884 instant download after payment.

Measurement and estimation of software projects has been extremely difficult for both technical and sociological reasons. The technical reasons include scores of poorly-defined and incompatible metrics, gaps or “leakage” from historical data, and the rather sparse collection of accurate benchmarks available to the general software community.

The sociological reasons center around the adversarial relationships between followers of rival metrics and measurement practices. For many years the “lines of code” metrics users have been at odds with the “function point” metrics users. Several other forms of measurement such as earned value, balanced scorecards, and goal-question metrics also have supporters and tend to ignore other forms of metrics.

In recent years the situation has become even more complex. As of 2008 there have been at least 24 function point variants, 5 methods for counting lines of code, and perhaps 15 other forms of measurement such as use case points, story points, object-oriented metrics, and others too numerous to cite.

Dekkers and Bundschuh navigate this tricky area with clarity and objectiveness. All of the major metrics variants are discussed and explained, and their pros and cons are noted.

The book also discusses the organizations that are trying to eliminate competition among the rival metrics camps and achieve some kind of consensus on what needs to be measured and how to go about it. Although there is still antagonism among the various rivals, this new book by Dekkers and Bundschuh is likely to be useful in leading to common goals and mutual understanding of what the various metrics were trying to accomplish.

Prior to the publication of this book, there was no easy way for followers of various metrics to learn about the other possibilities. While there…