Is there a book that discusses different types of indicators of source code?

Question

I have to present a document on the use of source code metrics in software development processes. I thought it would be nice to start by introducing some sort of categorization of the source code metrics (for example, using a VOC plane LOC counter that does some static code analysis), but I can’t find a book or article that will present various types of metrics and maybe some historical background information (books and article approved by my instructor on websites).

Does anyone know about such a source?

Answer 1

Pressman Software Development: A hands-on approach contains a lot of useful information about software metrics (among other things).

Probably the key metric tip is to be careful what you measure. If you measure LOC, then what will people focus on.

"Not everything that can be considered counts, but not everything that is considered can be considered."

- Albert Einstein

"Measuring software performance over lines of code is like measuring progress on an airplane, how much it weighs."

- Bill Gates

Answer 2

The Patrick Smacchia NDepend site has a good set of definitions: Definitions of indicators . His NDepend tool (for .NET) provides many different ways to quantify a code base.

How useful the metrics are, well, is a bit more controversial. They are good for getting a code-level view of the sky, but are they useful every day? How many stores do they use (or care) on a weekly basis? I suspect that most of them are more concerned about the number of errors: the number of errors corrected and the number of errors recorded.

Of course, measures, such as linkage and communication, provide indicators of possible refactoring points.

Another important metric: how many tests exist, and about as much as the code coverage these tests represent. Refactoring without tests and good testing is a risky business.

Software indicator

Code metrics overview

Answer 3

Performance and Models in Quality Software Development

Stephen H. Kahn

It’s not always very pleasant to read, but it’s a great book that covers quality and quality indicators. .

Answer 4

Halstead "Elements of Software Science" has become a major work in this area. When I recall the story, he identified about 15 different indicators, all of which are designed to measure different qualities of a code fragment.

Unfortunately, van der Meulen and Revilla showed that a pair of key metrics correlate very strongly with SLOC (source lines of code). (This was not the first time it was demonstrated, but it was the first document I could find quickly.) Since SLOC is much easier to measure than others, SLOC is what needs to be measured and managed.

Answer 5

PSP: Self-Improvement Process for Software Engineers Watt S. Humphrey

Answer 6

Also check out SLOCCount . This is not a book, but a set of tools for counting the physical source lines of code. Looking through his user guide , and articles on the site about using SLOCCount to estimate the size of the Linux kernel will give you an idea of ​​the effectiveness of SLOC metrics, using it as a metric, using the SLOC metric in the COCOMO model to measure effort.

Answer 7

This is not a book, but the metrics used by Sonar contain a good list of practically useful code metrics. Here is the metric definitions page.

Answer 8

As I just mentioned in my question about visualizing changes in indicators , there are Object Oriented Indicators books : complexity measures from Henderson-Sellers and earlier Object Oriented Programmatic Indicators that I have owned for years.

There Metrics and Models in Software Quality Engineering, Second Edition , is the only book available in Safari that contains a chapter that discusses the above metrics and some industry results.

I also found the 2006 book Object-Oriented Metrics in Practice: Using Programmatic Metrics to Characterize, Evaluate, and Improve the Design of Object-Oriented Systems in which there are no reviews (not a good sign), but it sounds interesting.