SOA Measures and Metrics
I just recently reviewed a blog by Loraine Lawson on What Metrics Can You Use for SOA. Interesting view, but I wonder if we are doomed as an advanced civilization when we have not learned to leverage the rich history and knowledge of our predecessors. No really, I am always amazed at our collective ability to overly complicate our lives by not leveraging the “wisdom of the ancients.” We constantly try to re-populate the software development knowledge pool every time a new technology appears on our societal door step. “Measures and Metrics” seem to be the latest example since the adoption of SOA as a computing paradigm by mainstream developers.
If you don’t believe me, go ahead and Google the subject “SOA measures” and see what you get. Go ahead, I’ll wait. See what I mean, article after article on “what are right measure to use in SOA development” or “Will SOA survive without real measures and metrics.” These kinds of responses are indicative of a larger problem – one not related to SOA. We stopped being software engineers and programmers (technicians) and became SOA hacks. Very sad.
Software engineering, as a practiced field of academic study and commercial operations, contains more information on measures and metrics than almost any other sub-discipline in computer science. ACM (Association for Computing Machinery) has over 1,000 journal and conference reports in this area, IEEE has about twice as much, and the Internet as a whole has countless articles on the subject. Measures and metrics are not new, even SOA-related ones, they are just not practiced as widely as they should be.
Yes, I know there are those out there that say “well, Dave Linthicum says current SOA metrics are invalid,” or “Hub Vandervoort says SOA metrics are on the rebound, but still are inadequate.” Well, I agree. But the real observation is “why is it that way?” The answer isn’t because SOA has some unique computer-based genome that is resistant to traditional measures; it is because we, as practitioners, has lost sight of the fact that SOA is nothing new and our predisposition to want to make believe that the important things we create must always be “new and improved.”
Leave a Reply