Location triggers are the essence of LBS...

Location sensing is a necessary feature of all mobiles. The whereabouts of each mobile is needed by the core of the mobile network. This is how calls and text messages get routed through to the mobile.

Location information is also useful for services. Many users have probably already experienced services like "Find" on the 3 network in the UK. These services are static, request-based services. The user requests information and at that time their location is calculated. There are many ways to calculate location and I have a whole chapter on this topic in my book.




However, location information can be much better utilised with location triggering. This is summarised in the above slide from one of my courses. It shows that when a user moves within the vicinity of an event or place/person of interest, there exists an opportunity to notify the user.

My previous posting (idea #7) required just such a capability. Recall that I wanted a 3rd party call to be initiated when I moved within range of my home.

The biggest challenge with triggering is the huge processing task in tracking all the users and all of their possible points of interest, which could be many per user. Many moons ago I blogged about this issue and it is also discussed in my book. I also drafted an architecture for a J2EE solution, which was part of a project to build a location platform for an operator in HK.

Techniques like in-memory processing allow for high performance calculation of the trigger points, but there is still the problem of demand on the network.

I was very interested to come across WaveMarket who claim to have solved the processing problem with their location platform. Their claims may well have validity, if their customer sign-up and recent funding round (9.2M USD) are anything to go by.

As WaveMarket has identified, location triggering is an essential component to any commercialy successful location-based service. There are many services that could utilise triggering. My own favourite is Splash Messaging, which I recently heard referred to as "Splat messaging". It is also a theme that I continue to research, especially in conjunction with graffiti.

WaveMarket are proving their technology through their very own location service, called Crunkie, which combines social networking, splash messaging and blogging - all potentially hot ingredients for a "killer app". However, ingredients are one thing, but usable and compelling services are quite another. I would like to know more about what users think of Crunkie.