We've been busy lately at Location Labs executing on Storage, Geofencing, and building out our more general purpose architecture for managing devices remotely from the network. The evolution of device capabilities, lead in particular by the rapidly evolving Android platform is driving a sea change in the way network operators, IT administrators, parents, service providers and individuals manage devices remotely and in turn perceive business and service opportunities related to these capabilities.

With our evolving Sparkle technology, embodied in part by our recent release of our Spatial Storage and Geofencing solution for smartphones, as well as the work we are doing with network operators we are working to help enable this sea change.

So what's new?

Well, for one thing working with our developer partners we've learned a lot about what the needs are around spatial storage. As a stand-alone capability, extending spatial primitives to the already mature cloud storage market, it's clear that this is not a screaming need. Even for moderate to large applications (requiring, say, a single spatial index of say 1 bn records), COTS solutions both cloud based and otherwise are available and adequate. The trick of course is how this data is associated with other data types (often referred to as "context") and in turn how it is queried. For example, consider check-in data. Context might include the social graph (i.e. check-ins associated with friends or friends-of-friends, etc.), temporally (associated with a particular event), or by keyword (say check-ins with a comment including the word "revolution".) This makes the problem much more interesting, and is something we're working on.

Our first vertical integration in this area is around geofencing. A common problem brought to us is this: notify me when the user is near X where X ∈ S, and S is some dynamic set of location-tagged records, such as venues or check-ins. OK, that's more interesting. Making this work requires some heavy lifting for the developer, including: running code on the handset 24/7, efficient use of evolving location technologies in a way that minimizes power consumption, synchronization with the network relating to changes or updates in the set of location-tagged records, etc. In a nutshell, this is the driving capability behind our recent product announcement.

To better understand this concept, consider the following possible applications: a) Notify me when one of my subscribers is close to a tornado or other weather event, b) Notify me when one of my subscribers is near some expiring asset, like a parking space or discount movie ticket, c) Notify me when I'm near one of my friends who just checked into foursquare, etc.

Of course to make this happen there were a number of dependent technologies and capabilities that came along with it. For one, we had to provide hosted spatial storage and a set of API's for accessing it. Click here for more information on how to get started with Spatial Storage. For now it is a straight extension of cloud storage including efficient range search (in 2D) against a particular layer, but as hinted to above, there is more to come.

The geofencing solution is provided by a mobile library (available for iPhone and Android) that is linked into your application. You register a geofence instance and refer to a particular spatial layer managed in the Spatial Storage solution. The library runs in the background, and if/when your user comes in proximity of an element in the layer you are notified via callback. This capability, unique to the industry, is proving to be fundamental to a wide range of exciting new services.

For more info on how to get started, you can visit our developer portal.

@sahotes