Construction engineers have solved all the problems software currently faces - *not*

I can’t count the number of times I’ve heard the wish: “if only
software construction was more like building construction, where all the issues are
clearly known up-front, everything is rigorously defined, and construction follows
a predictable schedule.”

Well, I read a recent post blows
the lid off that theory!  It turns out that building construction has all the
same problems; the plans aren’t fully defined when construction begins, and
managing dependencies to keep the work queue full is a major problem.  Construction
teams are looking at adopting multiple shorter phased development loops to solve these
problems – it looks like the construction industry has things it could learn
from computer development practices.


(Link found at Soapbox 3.0.)