Google’s “My Location” feature for Google maps on mobile is remarkable. It allows phones without GPS to be located. In one fell swoop it achieved the following:
Google have undermined the value of GPS in phones
Why buy a GPS phone when google can ‘upgrade’ your existing phone? A lot of people dismiss cell tower based location finding as useless due to its accuracy of 300 metres to 1.5 kilometers – it’s not much use for driving directions but its very useful for cinema times, traffic conditions, weather, news and find a friend functions. The less clicks the better on mobile and google just eliminated 90% of the pain of entering your location.
Additionally GPS doesn’t work indoors, using the last location that the GPS received can be woefully inaccurate (“why does my mobile think I’m in my back yard?”).
Google have drastically reduced telecommunications companies location based services stream
Telcos have been holding their crown jewels of cell tower based location finding close to their chests for years now. In most of the world there is no cross telco location service requiring developers to negotiate and intergrate with each telco separately – slow, painful and expensive.
The revenue model I’ve seen most often is to charge the equivalent of an SMS cost for locating a user e.g. if you have an app such as ‘find a friend’ and the application want to locate the user then the API cost of that is one SMS credit – typically 0.15 Euro. Google can now do this for all phones for free. Add this up and billions of dollars has been lost by telcos world wide – its entirely their fault – such a costly service means things like tracking services or location based advertising are unfeasible.
I was anecdotally told by a large telecom equipment provider that soon after the Google press release went out telcos were calling asking to ‘block’ access to this information – too late – the cat is out of the bag.
Google have removed one barrier of mobile services, especially search
A very smart person at Vodafone once told me mobile services are all about the user context. A key context is location – know where a person is, the time of the day, their previous searches and you know one hell of a lot about them. Yellow pages types search are greatly enhanced by a default location e.g. local cinema times, traffic jams, public transport.
Enhance PC searching services
Google now have the opportunity to enhance their PC based searches by offering a ‘locate me using my phone’ option on PC google maps, iGoogle etc. As with most of the privacy concerns about google, if the service is compelling enough most people will use it. Imagine a search box like the following
[mexican food ] [Search] [I'm feeling lucky"] ["Search using my mobile location"]
High barrier to entry for competitors
So now that we know it can be done (build a cell tower database without carrier permission or assistance) why not just copy what google did?
Its hard for several reasons
1 You need to install an application in GPS phones.
GPS phones are relatively rare – its seen as a smart phone feature, this greatly reduces your size of your crowd needed for crowd sourcing a worldwide cell tower database.
2 You need a pervasive installed mobile app on GPS phones
Google have the excellent Google maps for Mobile, lots of people have downloaded it. Its has secretly been collecting cell phone tower locations via GPS for months. What other native application is as pervasive as google maps? Games? Most are written in java and generally do not have access to cell tower information.
3 You need smarter than average mobile developers
Cell tower information is not normally available to developers, it has hitherto served no useful purpose to application developers, it is system level information like the MAC address of your ethernet card. Since the previous two points greatly reduced the data collection ‘crowd’ size you should read the cell tower info from as many phones as possible. You also need to work out timing information to roughly calculated the distance from the tower, this may involve some very low level hacking of the phone modem software – this means smart developers hacking windows mobile, blackberries, symbian, linux etc – this is expensive.
So who best to compete with google, preferably to create an open source DB of cell tower locations?
I think the only companies who can compete with this is phone manufacturers, in particular if Nokia added this to an update of Symbian and release some new phones they could create a rival database very quickly. Microsoft could do it too.
There is one other option, its cheap, effective and a secret – contact me if you are interested.