herraiz.org | Blog

Main | Blog | Research papers | PhD thesis | GnuPG (PGP)

In natural sciences, intensive metrics do not depend on the size of the system (or amount of material) being measured. For instance, the temperature of an object or the hardness of a material are intensive properties.

In the case of software projects, many of the metrics used to study them are extensive, that is, the metrics depend on the size of the system. Thus, it is hard to compare different software projects using these metrics.

For the case of the Apache Software Foundation, we have found some ratios that are intensive properties, and we show how to use them to study the evolution of different projects, comparing them regardless their size, intended audience, technology used or maturity state. The metrics are obtained from the ratios between communication and development activies in the project. Interested? See a tag cloud of the keywords of this paper:

The paper is available in PDF, and it can be cited using this BibTeX code (also available as a text file for your convenience):

@InProceedings{gala2013:apache,
  author    = 	 {Santiago Gala-Perez
		  and Gregorio Robles
		  and Jesus M. Gonzalez-Barahona 
		  and Israel Herraiz},
  title     = 	 {Intensive Metrics for the Study of the Evolution of Open Source Projects},
  booktitle =    {10th IEEE Working Conference on Mining Software Repositories},
  year      = 	 {2013},
  address   = 	 {San Francisco, California, USA},
}

We are presenting this paper at the 10th IEEE Working Conference on Mining Software Repositories, with the following slides (see the slides at SlideShare.net):

Finally, this work is derived from Santiago Gala's master thesis, available in PDF format and including full details about the study of intensive metrics for software evolution. There is also a replication package that you can use to verify and extend our empirical study.