What are the biggest challenges in home automation?
In the short term, the slowdown in housing — the end of the construction boom — is definitely affecting the market for distributed audio and control. About 60% of our business is for houses that are under construction. When someone is building a mid- to upper-level house now, they almost always include distributed audio and control of distributed audio. So when the construction market slows down, obviously sales slow down and that's a short-term challenge to the market.

Another challenge — and it's a good thing/bad thing — is the transition to HDTV. It's a good thing because it's exposing more people to surround sound, home theater, plasma TVs, and built-in sound. It's a bad thing because there are so many manufacturers, especially on the flat-panel-TV side, that the price deflation is damaging the specialty retailers.

And another challenge is that systems have become more complicated because there are more components and more sources that have metadata — data that describes what's going on with the source, like with an iPod or satellite radio. Controlling those sources requires very extensive programming and set-up.

You could give people flexibility — do you want XM or Sirius, an iPod?; do you want to be able to see that menu in the bedroom when you control a system in your basement? — by writing very complicated software programs. But that's expensive and time-consuming. We've developed a system where the program is wizard-based and the sources are modular. We saw this coming about five years ago. If you want XM, you buy the XM module and plug it into the card cage. And you can mix and match. As you control the system from a remote or from a keypad in another room, you can actually see the menu that goes with your iPod. The GXR2 receiver that's part of our IntelliControl System (ICS) has a Web server that surveys all the sources and user interfaces — touchscreen, regular keypad, or remote control — that are connected and programs itself to write the control screens on the touchpanels automatically, based on what it finds. If it finds an iPod, it will draw an iPod control screen and bring up all the iPod metadata without requiring the installer to do the programming to make that happen. It took us about three years, 15 software engineers, and about $7 million to figure out how to do that. It was what was required to address this market challenge.