Google Will Open Source National Parcel Map
This is my uninformed prediction, but it would be a bold move, backed by the National Research Council since 1980. It is the kind of game changing move that would help solidify their position as the number one search engine for localized business advertising.
A Halloween Treat - Gazing into Google's Future
I alluded to this possibility the other day when I looked at the possible importance of Google Map’s new parcel data layer and suggested that some mapping content that is currently very valuable will be given away for free in order to support a larger location-based advertising market. The release of their free navigation application for Android 2.0 seems to back these thoughts and suggests that mapping is just a means to an (advertising) end for Google. I believe the mapping community should get ready for more game changing moves.
Forbes also made the “free navigation” connection. It seemed pretty consistent with their overall strategy of building on their strengths to leverage into the localized search and advertising market. This local advertising market could be worth $40 B in new revenues to the industry. Since Google has ~70% of the current search related advertising market, a back-of-the-envelope estimate would make it worth ~$28 billion in new revenues, doubling their current revenues. Google has some room to make investments to secure this local search business that others cannot make, and they seem willing to push their advantage without regard to previous players.
The Forbes article specifically mentions a price per device for turn-by-turn navigation data of $5 – 10 per device. Worldwide mobile phone sales are on order of 300 M units, so if Google wanted to “own” the localized search and navigation space, it was going to cost them billions for the navigation data. So it appears they made a classic build vs. buy decision and came down on the side of build. We can argue about whether Google will be a “trusted” data source for navigation. However, the very fact that three very prominent “geonerds” in our industry are fixing Google’s maps for free suggests that there may be some value to the crowdsourcing of mapping data (see also – Open Street Maps).
The key to building a successful localized search and advertising service is the accurate navigation and business listings (and here as well). This requires highly accurate and timely cadastral data. For Google to be dominant in this space and double their revenues they have to solve the problem of accurate parcel data on a national scale.
The desire for a national cadastral map dates back to 1980 with the release of the National Research Council report, “Need for a Multipurpose Cadastre”. There are lots of issues as to why 30 years later we still do not have a national parcel map; and I believe some of these include the “ownership” and licensing of the localized parcel vectors as a revenue source for municipalities.
If Google really wants to create a national parcel map for navigation and business listing purposes, they will need to go through municipalities. They could obviously buy the data from Pitney Bowes or 1st American. These companies have built relationships with these municipalities over the years to provide high-quality cadastral data at the local level, which these companies then aggregate to provide a national-level geo-coding product. However, it does not appear that Google has bought their current parcel data from them; and considering that Google is leaving the existing navigation data providers NAVTEQ and TeleAtlas, it would seem that Google plan may be to go directly after the parcel data without the middle man.
So how do they get the parcel data directly from the municipalities without worrying about individual licensing with the 10,000s of local, county, and state governments? One approach is to commit to creating a national cadastre map, and then open source the project by giving yearly updates to the USGS National Map. What municipal GIS professional is going to argue against opening up their data to their users through an open source cadastre project that is backed by the NRC, federal government, and Google?
Google would still run the use and updating of “Google’s” parcel data via its APIs. They would still have the easiest and most recently updated service to drive their navigation and business listing services. They could sell “geo-services” directly to local and federal agencies based on their cadastral maps, which would save government agencies millions a year in maintaining cadastre data in databases that have little interoperability, i.e. a new business line for Google Maps Enterprise. It would also bolster their “Do No Evil” marketing, and perhaps keep the regulators off their backs (re: Microsoft in the 1990’s).
And oh-by-the-way it would speed their creation of a better turn-by-turn routing system and business listing service that would help them dominant a new $40 billion dollar localized advertising market. They could spend $1B on creating a self-perpetuating National Parcel Map, and still generate an enormous ROI on the expense.
Trackbacks
- Spatial is Special, Spatial IT is Not « Fiducial Marks – Paul Bissett, WeoGeo CEO
- Data is to ESRI as Search is to Google « Fiducial Marks – Paul Bissett, WeoGeo CEO
- Truly Open GeoData Needs a Micropayment Ecosystem « Fiducial Marks – Paul Bissett, WeoGeo CEO
Comments are closed.
Subscribe via RSS
Subscribe via email
When they switched to their own street data, I was thinking that we’d see them flipping it to OpenStreetMap at some point. I’m guessing that execs at NavTeq and TeleAtlas are hunkering down right now waiting to see if GOOG will nuke them by making it all open.
Yeah I’m not sure the federal government will get in the middle of this.
Why not just ask Google to donate cadastre to OSM
Randy
Wouldn’t take much of a leap for google to pay members of the crowd to update the network and parcel data. Who could keep up with that?
Sure, have an Open Parcel Map and watch surveyors wet their pants.
Geonerds fixing Google for free? Who’s that, and why aren’t they working on OSM?
In reply to Peter Rushforth and nomoreshoes regarding crowd updates to parcel data -
I seriously hope not. Parcel data and street data are two very different things. One major difference being that streets are generally publicly owned and accessible, while parcels are mostly privately owned. Local govs have enough trouble now, explaining to residents that the published county tax maps are not legal documents, and only useful for helping identify taxable parcels. I can only imagine the phone calls they would receive if all parcel data were published on Google, and then updated by their neighbors. Worse yet would be the strangers tramping across private property to “help update” the property boundaries they find on Google. That would be a mess.
GPS vendors already have enough trouble with nitwits driving into lakes or down railroad tracks because their GPS units said “there’s a road”. Can you imagine the conflicts that will ensue if property lines become open to, and editable by all?
An addendum to my comment – open to all is OK by me as long as caveats are known. Editable by all sends shivers down my spine
Excellent points, Roger.
I think your addendum is what I was suggesting. Open source doesn’t mean crowdsource (at least not to me). I think I may have inadvertently suggested such. There would still need to be standards, and probably a quality control facility. I suggested the USGS, but have been told there might be some legal issues surrounding the feds assumption of this responsibility. The important thing would be to have some mechanism to facility what Roger is suggesting – get the data as quickly, and as accurately, as possible into the hands of the public.
Don, you pretty much hit the nail on the head. You are absolutely correct that much of the public ascribes the degree of legal authority to parcel maps that only surveys and other documents signed and sealed by licensed surveyors carry. While I’m all for open geodata and am working hard to push my agency in that direction, I have the same fears you have about the confusion that is already bound to result with Google Maps parcels and would only be exacerbated with crowdsourced parcel data sets. I’ve been letting my County Assessor know what’s been happening with Google Maps because sooner or later it will affect his office.
In my mind, it’s more vital than ever that jurisdictions that have the legal authority to maintain parcel maps get their data out of the vault and into the public’s hands in ways that are easy to use and oriented toward answering the most commonly asked questions. Traditional GIS web apps that throw the entire kitchen sink at the average person are only going to go unused if there are far less painless ways for people to find the information they’re looking for, or at least something resembling the information they’re looking for.
Paul, I somehow missed your comment about “prominent geonerds fixing Google data for free” in my first pass through this post. I assume I am probably one of those geonerds based on this blog post. I just wanted to point out that you appear to misinterpret my intentions in submitting a couple of updates to Google – that was purely to test out the process and verify how well it works. As I point out in my latest post, I really question whether people are going to be very motivated to provide these updates to Google. I think that typical consumers are just likely to switch to using a different free system (MapQuest, Microsoft, Yahoo) if they see lots of errors (and see fewer errors in those other systems). And I think that geonerds will continue to contribute their time to OpenStreetMap so that their efforts can be used by a broad range of organizations rather than just Google (that’s what I will do).
I agree that motivation will be a key moving forward for continual updating of their navigation product. I also wonder (like Kirk on your post) about a sensor network power by android phones. I could imagine a pretty decent solution that used the sensor network as an automated self-correcting solution to the base map. This could greatly accelerate the correction of the base map.