The guy who made the blue dot on Google Maps possible is at it again, this time with an even more accurate location technology.
In 2006, Sam Liang of Google started to work on the company's geolocation project. It was his team, he says, that created the back-end technology that enabled the creation of the blue dot on Google's mobile maps: the one that tells you where you are.
Now, six years later and at his own company, Alohar Mobile, he's working on a new blue dot: one that's both more precise and that uses far less battery life.
What he's trying to do is create an "ambient" location tracking technology. (Mark that word: ambient. It's an emerging thing in tech products.)
Alohar's technology works by using not just the usual location-finding sensors (the GPS receiver and Wi-Fi hot spot triangulation) but also other sensors and algorithms, some of which Liang would not tell me about. It does use the accelerometer and compass, I learned, and also statistical modeling to tell "where you most likely are."
Car navigation apps also use a form of modeling: Since you're in a car, they assume you're on a road, and even if the GPS radios aren't placing you directly on a road, the algorithms will "snap" your location to one by default. Alohar does similar things but uses more data: If you're moving at walking speed, for example, it will place you on a sidewalk or in a building, not in the middle of the street. If the camera on your phone sees fluorescent light, it will try to geolocate you to an indoor location.
If you stop moving, Alohar will shut down the GPS until you start again; that's one way it does better geolocation while using less power.
The goal is to make accurate location data available to apps all the time, without draining batteries. This always-available location data can be used for cool new things. At the Launch conference in March, Liang showed how an always-running app could determine if a person carrying a phone was in a car accident, and send an Onstar-like emergency alert automatically. Or, a user could press a Help button if they were having a medical emergency. Liang points out that standard phone GPS isn't accurate enough to direct paramedics or police to a downed person's location; Alohar pinpoints people much better.
Alohar ties in to a rich database of locations, which the company's proof-of-concept app, PlaceMe, uses. It tracks where you are at all times, builds a list of the places you visit, and for how long you stay at each one. It's creepy, but it can also be useful. If you stop in at a store or cafe for a bite you can see, later, where it was.
By itself it's perhaps only of interest to those into the quantified self thing. But the technology could be very useful for note-taking or camera apps, many of which already geotag items you create on the run. With Alohar technology the tags could be more accurate.
If you want to experiment with it, I have to warn that PlaceMe, while cool, isn't fully cooked. It does affect battery life; on an iPhone 4 I found the impact noticeable but not bad enough to prevent me from using it. On my Android phone (a Galaxy Nexus), though, it murdered battery life enough to be unusable. Liang says updates are on the way for both platforms.
Liang says that the app respects privacy, by the way. While Alohar's servers do record the location trail of PlaceMe users, location data is tagged with a hash of the phone's hardware ID, not with personally identifiable information. Still, if you have an extralegal side business or a friend you don't want people to know about, I'd recommend against downloading this app.
Liang's goal is to get the location technology out there to developers. PlaceMe is just a demo. He says that potential uses for ambient location sensing include apps that learn where you "dwell," even if you don't check in. For example, your phone could now learn what kind of food you like based on which restaurants you stop at. It can tell how much time you spend walking vs. jogging, which could be useful for new kinds of ambient fitness apps (but see also: Jawbone Up, Fitbit, etc). Liang also said that the technology would be really useful for "double-blind dating apps," but I'll leave interpreting that to the reader.
Over 600 developers have signed up to use or try the Alohar technology, Liang says. The technology library is free, as is use of the service for experimental apps. Alohar will eventually charge a license fee for intense users, just as Google does for those who embed Maps in their online apps.