Wednesday, February 1, 2006

The old truths about software are the best

Well, it has been 25 years of studying software methodology. I remember Dr. Edward Vondrak grading our flow charts in 1979. If the arrows weren't corrected properly it was a point off. Was there a method to this madness? Yes there was, he used the flowchart to see if the student really had thought through the assignment and designed.
In 1980, I had the rare privilege of obtaining a part time consulting position through Daniels Associate, Inc. in Indianapolis, Indiana. Why was it a rare privilege? Because I was going to be able to do something called on-line computing. We actually programmed on CRTs, there were no punched cards! The client was the Allison Gas Turbine Division of General Motors. At GM they were learning something new, Structured Analysis and Design. Most of the folks there had taken courses from Ed Yourdon's company.
In 1994, I had the responsibility to help introduce object oriented technology at The Options Clearing Corporation. Machiavelli, "there is nothing more perilous than introducing change..."
Now, reluctantly, I find myself helping a client improve their software process that is based loosely on Agile Methods. The newest, new thing.
Well... I am tired of new, new things. After 25 years, I prefer to see what truths and what benefits can be found in the "new" things that were actually principles recognized many, many years ago.
The folks I respect in the "meta" side of software development - the development of software development processes, are the ones that have been around uncovering truths and realities for 30+ years. Folks like Barry Boehm, Tom DeMarco, Robert Glass, Fred Brooks to name a few of my heroes.