Tag Archives: Systems Thinking

Cometh “The Internet of Things”, who will own it?

Back in the late 1980s I managed a software support team. In those days, before contemporary enlightenment bought around by the advent of plug-and-play, the constant refrain was the interchange with the hardware support team “It’s your hardware!”, “No,it’s your software”, “No, no, it’s clearly the hardware”….and so on. These days we’re used to dealing with new hardware as one single,easy entity, but back then there was a clear demarcation, made more visible by the practical jokes played by each team on the other.

Twenty five years later I work in a communications service provider (CSP), albeit one that has its own telecommunications carrier network. In my organisation there is much the same demarcation going on between the telecommunications team and the internal IT team, although that partly has to do with maintaining the security of each domain which provides the visible artefacts suporting the separation. Look at some of the other CSPs,and core telecommuncations companies, and you’ll find that this pattern is quite widespread. Even where I am aware of forward looking companies that have merged the comms & IT teams, they still keep their own identities.

The problem we are facing is that, as happened to the old hardware/software split back then, the two worlds of “communication” and “information technology” are accelerating towards each other. To be honest this shouldn’t come as a shock, but I’d suggest that both sides are trying not to pay too much attention to the consequences of this and are carrying on with the hope that it’ll all sort itself out in the wash, “the wash” being the compelling event that forces some sort of activity out of us all.

I must confess here that I had expected the emerging IP Multimedia Subsystem (IMS) architecture to be the driver for this merger of the technology areas but now it’s looking unlikely for a few years yet. What has spurred my interest in this is an interview with IBM’s Andy Piper about the MQ Telemetry Transport protocol, a lightweight transport protocol being proposed for the coming generation of Internet-connected devices.

So why do these architectures make a difference over what we have today?

Yes,it’s true that a lot of the more modern telecommunications devices that are implemented are actually computers and (speak this very quietly) databases. However I’d argue that the fact that you squirt PSTN protocols into them, and program them as you would other PSTN systems, puts them firmly into the telco domain.

The emerging IMS & telemetry systems bring a new set of challenges, for example the handling of complex data & topologies, something where IT departments have been working for a long time now.Similarly the new abstracted protocols such as Session Initiation Protocol (SIP) require a level of orchestration that we find in workflow systems; and they are supported on platforms running modified forms of traditional programming languages such as the JAIN SLEE extensions to JAVA.

And then there is the “killer app”: back in 2009 the BBC’s Rory Cellan-Jones wrote about the IBM researchers using the early MQTT developments to link a house directly to Twitter. As we use social media more as our interface to the underlying technology fabric then we need to think about user interfaces in a very different way, again the challenge being taken up already by IT departments.

No, I’m not advocating that IT departments around the globe prepare to take over the world. Rather the Systems Thinker in me is advocating that there are two skills bases (the IT and the Comms teams) that will converge over time. There are different perspectives here, but it’s not beyond the realms of systemic inquiry to bring them together.

It’ll be interesting to see who the smart Communication Service Providers are that drive that convergence sooner rather than later so that skills transfer can begin now. As Sergeant Stan Jablonski used to say in Hill Street Blues: “Let’s do it to them before they do it to us.”