We describe an experiment whose purpose is to establish our hypothesis regarding the relative “software flexibility” of particular design policies (in this particular case, programming styles) towards particular changes. We present a program with two implementation variants, one using a procedural programming style and the other one using an object-oriented programming style. We instruct the subjects to carry out a predefined set of changes in each implementation, and to test the changes they made. We measure the time each change required, in order to be able to compare the effect of the programming style used.