I just can't figure out why Google didn't buy TeleAtlas or NavTeq.
I keep getting this strange feeling that Google is secretly collecting their own map data, so that they'll eventually be free from having to lease the info.
Maybe they will. It might be so they stay the right side of a monopoly legal problem.
I guess it doesn't fit into their strategy - Google provides a whole load of layers on top of it - it created the KML standard in effect, and it has a fair bit of work that adds on - Panoramio, Street View etc.
Google it seems will provide a huge user base to help Tele Atlas update its maps. Google in return strengthens there position. It's not like there are too many other companies out there that Google can turn to, due to Nokia's acquisition.
With Google's work on Street View, why couldn't they start making their own maps? If they had a decent tracking system, they could port/mesh on the data from it to a map. I've always wondered about their acquisition of the modelling company. I can see part of it in Google Earth, and some potential in the small scale as a on-birds eye - more angled framewire building view for cities, but it could be part of a larger picture.
It wouldn't be too much of a stretch to create their own maps? Talking about huge data sets - if they had the satellite data, they could map on the gps requests from iPhone users, say, and then with an average of the data, then kind of recreate the roads through a kind of gps trail. Kind of hard to describe. A bit like pointillism.
I think you could pretty much recreate the map data, once you had the street view data, and were able to make sure it was aligned to known co-ordinates. Any which way, the LBS , GPS and Photosynthesque areas are going to explode pretty soon...