I set a rule that people weren't allowed to send good news unless they sent around an equal amount of bad news. We had to get a balanced picture. In fact, I kind of favored just hearing about the accounts we were losing because ... bad news is generally more actionable than good news.

A bad strategy will fail no matter how good your information is. And lame execution will stymie a good strategy. If you do enough things poorly, you'll go out of business.

The next time someone tells you we can trim the budget by cutting aid, I hope you will ask whether it will come at the cost of more people dying.

I'm going to save my public voice largely for the issues where I have some depth.

Energy is very primal stuff and there are a lot of leads that are promising, still at a fairly risky stage, but over the next decade some of these breakthrough approaches are going to pay out, and U.S. research and U.S. leadership on this should be part of how it gets solved.

Microsoft's philosophy is to 'do things better.' And Vista has given us lots of opportunity to do that.

Programs today get very fat; the enhancements tend to slow the program down because people put in special checks. When they want to add some feature, they’ll just stick in these checks without thinking about how they might slow the thing down.