Cloud Society: balancing the cloud and the device

By handling processing in the cloud, users of mobile devices could have a lot of power in their hands

The architecture is simple, but profound. Take a fantastic amount of server processing and storage capability, running in a highly optimised, multi-tenancy data centre, anywhere in the world.

Next, create a handheld computing device which is powerful enough to support multi-tasking, with a high-resolution graphical screen, a camera and other interfaces. Connect the two via a reliable network connection and as much bandwidth as you can muster.

Now, then, the question is, what can you do with that? One answer - a lot more than is currently being done with smart phones and cloud computing. The opportunity to innovate - to see the device as a telemetric end-point, as a control desk, as a multimedia collection and delivery tool - is clear. And we are starting to see examples of applications and services that exploit both device and cloud.

Take Nuance for example, a company known for both voice recognition and document management solutions. Nuance's flagship voice recognition offering - Dragon Dictate - is pretty good as far as it goes, but it requires a reasonably powerful personal computer, even when it is used with a good microphone and a pre-trained user profile. Equally, the potential of voice recognition on a handheld device offers a compelling use case.

Enter the cloud - and Nuance's FlexT9 product for example, which offers (among its capabilities) a speech-to-text function on Android phones. Only it doesn't - it records the speech, uploads it to Nuance's servers which then return a stream of text. No training or good mike required. This isn't meant to be an ad for FlexT9 but it illustrates the point. As does Apple, which adopts a similar model for its Siri Q&A application.

The fact that a device can do the simpler stuff that needs to be geographically close to the user, and the cloud can be used for the more complex, back-end processing, offers a wealth of possibilities of which we are not yet even scratching the surface. A recent discussion with John West, mobile solutions architect for Nuance, gave some indication of the direction things were going.

"We're putting most of our stuff in the cloud," says West, using FlexT9 as an example. But other examples show how voice recognition is just the tip of the iceberg. "We're looking at the medical domain, at how a physician can articulate a query which can be parsed and results returned from multiple sources," he says. Other examples he cites are around financial services and travel. Indeed, whatever a back-end system, or combination of systems could do, from analytics, to biometric checking, to complex processing, to media handling, could equally be part of a device-plus-cloud service.

What's going to stop progress? Perhaps the problem, as ever, is the availability of reliable network bandwidth - but this is less an issue in wireless-enabled campus environments such as large corporates, academia and healthcare. We're not just talking about highly mobile consumer applications, rather, anywhere that a smartphone can become the remote control onto a globally scalable back-end.

Mobile handsets will continue to become more powerful, with embedded capabilities that once were the domain of high-end devices. But the real opportunity for innovation comes from using both mobile clients and cloud in equal measure, making the best use of each. One day soon, you really might have the power of the entire cloud in your pocket.

Read more about:

Sign up for our free newsletter