Groundhog Day


 In Groundhog Day, Bill Murray’s charming 1993 film comedy, a TV reporter travels to a small Pennsylvania town to film the February 2nd Groundhog Day festival.  After filming, he collapses in his hotel room and when he wakes up the next morning it’s not February 3rd as it should have been, but February 2nd all over again.  And the next morning it’s February 2nd again, and then again and again and again...  Everything that happens each time is the same as what happened on each of the earlier instances of Groundhog Day; everybody says and does the same things at the same time in exactly the same way.  Our hero is stuck in a time loop.


Tom DeMarco

   If you’ve done more than a very few software projects, you know that Groundhog Day can happen in organizations.  There are patterns of behavior that repeat from one project to the next.  The lessons you learn on project n are often the same ones you learned on project n-1 and the same ones you’ll be learning in the future.  You’ve already learned what needs to be done next time around, but next time around the same things happen again. 
   The loops we get caught in are patterns of behavior that establish themselves in a corporate culture and then prove resistant to change.  Some of these patterns are harmful, but others are surprisingly beneficial.  Keeping the good ones and routing out or at least reducing the effects of the bad ones is a key to improved performance. 
   But here’s the rub: altering corporate culture is cogitative and, by definition, cooperative work, but we don’t even have the vocabulary to address and think about patterns of group behavior.  As Alistair Coburn has observed,

“We are still in the infancy of naming what is really happening on software development projects.”

   Pre-articulate humans and proto humans may have discovered that water could be used to put out a fire, but without either of those two key words, they could never pass that information on.  It had to be discovered and rediscovered again and again.  Welcome to Groundhog Day.  For lack of a good vocabulary of our governing patterns and a deep understanding of what they are and why they persist, we haven’t much chance to alter them.

   And now begins the part of this little essay that is a blatant advertisement.  The six members of the Atlantic Systems Guild have been working together over the past three years to produce a book about the sociological patterns that affect team and organizational behavior.  We have identified some 89 of these patterns, given each one a name, a concise definition, and a short essay about how to recognize the behavior and in some cases what to do about it. 
   As we’ve gone about this work, we’ve been able to observe in our own activity some of the very patterns we were writing about.  Yes, we had our moments of Mañana (Pattern 7), got sidetracked by a few Happy Clappy Meetings (Pattern 4), and learned to our dismay that Silence Gives Consent (Pattern 25).  We sometimes felt like Project Sluts (Pattern 38), but we did nonetheless respect the Sanctity of the Half Baked Idea (Pattern 58) , and finally Rattled Our Dags (Pattern 26) to get the job done.  In retrospect, we were lucky to have adequate Seelenverwandschaft (Pattern 66) to see us through. 
   The book has already been published as your read this, or is about to be, in both English and German.  Keep an eye out for it:

Adrenaline Junkies and Template Zombies: Understanding Patterns of Project Behavior.  Published by Dorset House in the US and Hanser Verlag in Germany.

Click for free sample pdf.

Tom DeMarco
Camden, Maine,