This post was authored by Collin Dailey, a technology development engineer at Phoseon Technology, and Jacob Jaugen, content manager at Portland Engineering.
At the core of systems integration, what do we do? Systems integrators make machines talk to each other and move data. Clients pay systems integrators for knowledge of existing and available resources to design or, perhaps more appropriately, to piece together solutions. The systems integrator relies on industry buzzwords like “virtualization,” “thin clients,” and “high availability” to describe some of the technically complex frameworks available as potential solutions. Systems integrators build a reputation by being experts in controls integration, but the reality is they are only middlemen between complex hardware/software and the end user.
Systems integrators do not build virtual machines (VMs), but rather deploy them. Systems integrators do not create thin clients, but rather configure workstations. Systems integrators do not engineer high-availability systems, but rather integrate the pieces to achieve them. Everything done relies on using tools that the systems integrator “programs” with ladder logic or function blocks. If the client has a problem for which traditional programmable logic controller (PLC) suppliers have not already done the actual heavy lifting of developing a solution—which a systems integrator can in turn configure—the client is probably out of luck and will have to accept lower expectations. And because this is how it is and always has been, we have become complacent in our assumption that this is how it always will be.
A buzzword has come along that old-school integrators seem to universally scoff at, as though it is a contrived novelty that is somehow below them: the Internet of Things (IoT). Let us say this very clearly, the Internet of Things poses the greatest threat to systems integration status quo since the invention of the PLC. It has turned millions and millions of normal day-to-day consumers into systems integrators. Let that sink in for a minute. What you as a systems integrator do for a living—those special skills that pay your bills—has very quietly and very rapidly become not that special and unique. Sure, these in-home integrators are not working on hardened industrial or commercial systems, but they are working on the same problem: making machines talk to each other successfully.
Large-scale networks of interconnected devices now exist at the hobby level as an affordable reality, and normal people are connecting all variety of machines and making them talk without ladder logic and without a professional systems integrator. Not only are they doing it without a professional systems integrator, they are doing it in more technically complex ways.
Every integrator knows what SQL is, but how many can write it? How many systems integrators can write a select query, or do they rely on other experts to build GUIs that take care of all the hard work? How comfortable are you with Linux and writing low-level code? You know what a VM is, because you are up on your jargon and there is a great host of VM software out there, but how many of you understand the step beyond virtualization and can speak to the value of a server-less architecture? If these examples seem above the ability of an average systems integrator or outside the traditional scope of our services, let me assure you that they are certainly not above the ability of people connecting Ethernet cameras and digital thermostats in their homes.
Our industry primacy is largely the result of people not understanding what we do, because honestly, how a variable frequency drive controls a solids centrifuge at a wastewater treatment plant is not very interesting or glamorous. Doing it in a different way violates the “if it ain’t broke, don’t fix it” rule. All of that is changing right now, and systems integrators need to be proactive so we do not wake up one day to find we have been left behind.
So, what are some solutions or ideas on how to wind up on the right side of history? For starters, integrators should not fear text-based programming. Function block and ladder logic are legacy ways of describing relay logic. It is a terminal dead end, and the IoT/IIoT will never be ladder logic programmable. While the barrier for entry to this level of computer programming is beyond the skill set of traditional integrators, it is not beyond the skill set of the millions of over-educated millennials working in coffee shops and tooling around on GitHub for fun in their spare time. These same people are the ones filling our hiring pool. Although we typically just want somebody who understands ladder logic, we have to be open to and accommodating of the multitude of programming perspectives these workers provide, even if it is outside our comfort zone.
The tools that we use are incredibly complicated in their construction. We as integrators do not need to be PhDs in computer science, writing custom Linux kernels, but we absolutely need to get on board with the fact that structured text is not just for the occasional tricky math calculation. Programming languages like Java, Python, and C/C++ are becoming a standard for anybody worth their salt at connecting devices. The only barrier to learning the most basic level of scripting language required is having a passion for making things work and knowing how to use Google.
About the Author
Collin Dailey is a technology development engineer at Phoseon Technology.
About the Author
Jacob Haugen is content manager at Portland Engineering, a Pacific Northwest–based instrumentation and control design engineering firm.
A version of this article also was published at InTech magazine.