On the face of it this is hard for me to understand. Here's the way I can rationalize it: imagine Google spends $20 million (or whatever) to build underlying t

On the face of it this is hard for me to understand. Here's the way I can rationalize it: imagine Google spends $20 million (or whatever) to build underlying technology. Then spends $2 billion to roll out a nationwide sales force sufficient to saturate all cities of some threshold size, and another $1 billion for a nationwide marketing campaign to introduce the project to consumers, all over the course of 1 year. On the surface this is a better deal. But does it remind you of anyone? Microsoft. This is what Microsoft does when they try to do something new. And all too often, they fail. One reason they fail is that it's hard to effectively mobilize thousands of people and spend gobs of money in a short amount of time. The effort is so big it threatens to crush itself under its own weight if it is not properly managed. And management is hard. Google is essentially sidestepping this management risk by spending a ton of money. They get something that is in a working state, rather than trying to quickly deploy something huge and new. Money is (relatively) cheap to Google, but opportunities for low risk, significant growth into a complementary space are rare. So this makes sense for them. Here endeth the half-baked disquisition.