In the waning years of the 20th century, programming technology was remarkably stagnant. Most academic computer science programs taught essentially the same programming principles and concepts in 1999 as they did in 1970. But that period of stagnation (or stability depending on your viewpoint) abruptly ended at the dawn of the new millennium.

We have entered a new era where industrial practice is rapidly changing and academic computer science programs are scrambling to keep up. In industry, momentum is gathering behind a new approach to software development characterized by:

* the use of safe programming languages ("managed" rather than "unmanaged" code);
* an emphasis on object-oriented rather than procedural problem decomposition;
* agile, test-driven methods for software development rather than more monolithic methods such as the waterfall model;
* growing enthusiasm for sound, sophisticated static type systems;
* explicit, carefully specified language constructs to manage concurrency.

In this talk, I will describe how the Rice Computer Science Department has kept abreast of this paradigm shift by developing an innovative programming curriculum focusing on object-oriented program design and test-driven software development. I will also speculate on how software engineering research is likely to further impact industrial practice over the next decade.