Google “Opens The Kimono” On How It Builds Maps
The Atlantic published an article about what Google calls “Ground Truth,” its allegedly top-secret map-development process. As the name implies, the idea is to create digital maps that more accurately reflect the ground truth of the real world vs competitors’ offerings. The trigger for the article is undoubtedly Google’s desire to expose the complexity, layers […]
The Atlantic published an article about what Google calls “Ground Truth,” its allegedly top-secret map-development process. As the name implies, the idea is to create digital maps that more accurately reflect the ground truth of the real world vs competitors’ offerings.
The trigger for the article is undoubtedly Google’s desire to expose the complexity, layers and (by implication) cost of the process of generating maps in the wake of its ouster from iOS 6 and increasing competition from Nokia/Navteq and OpenStreetMap.
There are several things about the article and the revealed Google map-development process that are interesting. The first is that Google’s Street View process is likened to a webcrawler for the real world, along with the notion that it’s “indexing” the physical world just as Google indexes documents and pages online.
The article quotes Google Maps’ Brian McClendon on the use of Street View and the “OCR-ing” of the physical world:
We already have what we call ‘view codes’ for 6 million businesses and 20 million addresses, where we know exactly what we’re looking at. We’re able to use logo matching and find out where are the Kentucky Fried Chicken signs… We’re able to identify and make a semantic understanding of all the pixels we’ve acquired.
The next thing that I found interesting is the degree to which Google uses human beings in correcting and editing maps. This is perhaps the revelation of the article: how much literal “manpower” is involved in the production of Google Maps.
Google acquires third party data as the foundation (i.e., US Census Bureau data). Then it adds its own Street View-acquired data (millions of miles of roads driven). Finally the humans go at it:
Humans are coding every bit of the logic of the road onto a representation of the world so that computers can simply duplicate (infinitely, instantly) the judgments that a person already made . . .
The sheer amount of human effort that goes into Google’s maps is just mind-boggling. Every road that you see slightly askew in the top image has been hand-massaged by a human. The most telling moment for me came when we looked at couple of the several thousand user reports of problems with Google Maps that come in every day. The Geo team tries to address the majority of fixable problems within minutes . . .
It’s a massive, complex process that Google developed over a period of years — at enormous expense. Perhaps Navteq (Nokia) has a comparable operation in place but it’s doubtful. Thus the payoff for Google in exposing “Ground Truth” to The Atlantic: “I came away convinced that the geographic data Google has assembled is not likely to be matched by any other company.”
Anecdotal validation of this for me came this summer in the UK and Ireland in a rental car. I used both a TomTom GPS device and Galaxy Nexus. The Galaxy Nexus-Google Maps experience proved to be far superior to the TomTom GPS experience. Apple is relying on TomTom to varying degrees around the world for its mapping data and (perhaps) navigation, which makes me question the potential accuracy and reliability of Apple Maps.
The article reflects the massive commitment that Google has made to maps. The question is whether Apple is willing to make that kind of commitment as well and whether it can rely so heavily on partners or whether it equally needs to built the kind of operation reflected in the Atlantic article.
My sense is that if Apple is “really serious” about delivering a great maps & navigation experience it will need to go much further than so far it apparently has. But we’ll know much more when iOS 6 is released to the public later this fall.