Programeter vs. Hackystat

I just came across Programeter and these discussions about the value of automated measurement. For the past seven years, I have been leading the Hackystat Project, which is an open source framework for automated collection and analysis of software engineering process and product data:

http://www.hackystat.org/

Over the years, Hackystat has been used by thousands of developers and hundreds of organizations world-wide. In my experience, I can say that automated measurement definitely has a place in software engineering, but it also has a tremendous potential for misuse.

One simple way to look at it is as follows: automated measurement can evaluate the program, the programmer, or the programming (i.e. the process). The more you focus on the program or the programming, the safer you are. As soon as you start to focus on the programmer, you tend to get yourself into trouble (at least in our experience).

Hackystat has mostly focused on the program and the programming, with results that you can read more about here:

http://code.google.com/p/hackystat/wi...

For example, we developed a rule-based system for automatically inferring whether or not a developers was practicing Test Driven Design, which was the subject of a recent Ph.D. dissertation by Hongbing Kou.

This is just one of the many ways we have applied the Hackystat infrastructure in the past. Perhaps the Programeter development team can profit from our experiences.

Philip Johnson