Editor's note

When contemplating an IoT project, most organizations realize that they will need to journey down roads that are completely new to them and develop an enterprise IoT approach unique to their situation.

There's a lot to be explored and learned before any IoT product or service can be successfully conceived and deployed. Just keeping track of what unfamiliar disciplines need to be embraced (for example, programming or mobile connectivity at scale) requires that each organization makes sure it has a conceptual architecture or stack for its vision of IoT.

There's no single picture that covers every company's IoT approach, but there are nevertheless challenges and requirements that are common to almost every enterprise IoT project. Here, we consider four such common themes as they've appeared in recent coverage.

1The IoT security plane

The IT department has been holding itself up to the measure of adequate security and finding itself wanting for 30 years and more. In the realm of IoT devices, it enters a world where lots of designers and developers don't seem to have security on the requirement list. Beyond the vulnerabilities of small, embedded systems (sometimes running Linux and pushed into production still using default root passwords), identity and authentication from app to thing and from one service to another remain huge attack surfaces for the next generation of hackers to exploit. Building security into your enterprise IoT approach is critical.

2Edge interaction gateways

The cloud is a great thing, but it's not the only thing. Particularly in the case of large-scale IoT deployments, it sometimes makes sense to consider things almost in reverse from the common view. Rather than thinking about how the cloud application will keep tabs on all the devices, think instead about how far a device needs to reach before it communicates with the other devices or hubs it most needs to interact with. If the most immediate transactions are occurring entirely in a local scope, that's the moment to start thinking about moving to an edge IoT approach.

3Analytics for additional value

Right now, our dominant vision of what the internet "is" is the human interaction of readers and websites. In a few years, though, we may come to think of the internet as the machine that reads and processes the entirety of all the sensor-reading data in the world (plus transmissions from a few planetary explorer missions). The value of this machine will depend entirely on its ability to create meaning from data mass. As with all of the internet, there won't be a master plan, just individual organizations and governments finding ways to tease huge data sets for insight. There are enormous opportunities -- and unsettling privacy threats -- to be considered here.