Rather than stop selling data to Google, what if Nokia/Navteq simply raise the price? That would open the door wider for TomTom/TeleAtlas to sell their data. Or it might make other schemes of data collection more economically competitive.

It may make TomTom/TeleAtlas an attractive takeover target for Google. What better way to get into the device selling business than with a company that has a pretty sizable market share with a zillion units in the field? All you have to do is take a look at TomTom's "PLUS Services" page to realize that TomTom is a hair's breadth away from selling a local search device capable of delivering advertising-supported content. What if the Google Phone is really a personal search assistant, not a phone?

But let's say that Google doesn't buy TomTom. What else could they do? They could use OpenStreetMap data as some have pointed out. But stopping there is pretty conventional thinking. Let's assume that Google, the primary sponsor of the Google Lunar X PRIZE, isn't going to take the conventional route. Maybe they will take up automated feature extraction from their imagery and send the tricky parts to a bevy of human helpers much as was done in the Fosset and Gray searches. Or, maybe they will help foster an entirely new means (seen via BlinkGeo) of collecting data. Human-augmented feature extraction (for pay, with redundancy built in for safety) or data collection drones, or a combination of the two, may not be so far-fetched. The potential of having all the mobile GPS devices in the world upload their locations is largely untapped as well (Inrix is showing how it could work). The same goes for mobile phone location data.

Bottom line. I don't count Google out. Instead it's fair to say that there will be interesting developments.