These Thirsty Geeks Invented The ‘Internet of Things’


Mater atrium necessitas.

Surely that, or something very much like it, must have been what the four computer science students at Carnegie-Mellon University were thinking when, way back in 1982, they invented what became the first, and still the prime example of what the “Internet of Things” is.

Or maybe they were just thirsty geeks having fun as only geeks know how.

Whatever the case, Carnegie-Mellon student David Nichols had grown irritated by the amount of time and effort wasted walking down the hall to the Coke machine only to find it empty – or the bottle of soda he bought still warm. So he enlisted the help of three buddies from the computer science department – Mike Kazar, John Zsarnay, and Ivor Durham – to join him in creating a soda supply monitoring system.

Their comical, yet ingenious system used micro-switches in the Coke machine to sense how many bottles were present in each of its six columns of bottles. The switches were connected to the department’s main computer and the students wrote software that allowed them to keep tabs not only on how many bottles were in in of the machine’s columns but how long each bottle had been in the machine.

The program could even tell them which button to push to get the coldest bottles, and how long it would be before recently loaded bottles would be cold enough to enjoy. The machine and the monitoring system were actually linked via ARPANET, the pioneering data network created by the Defense Department for use by university-level researchers around the nation. It was the forerunner of the Internet. Later on, the machine was connected to the Internet itself, which earned it its nickname, “the Internet Coke Machine.”

A Bit of Fun Leads to Big Innovation

The four students might have viewed what they were doing as a fun learning exercise. But what they were actually doing was acting out the adage, quoted above in Latin, often attributed to Plato: Necessity is the Mother of Invention.

Nichols, Kazar, Zsarnay and Durham had never heard of the Internet of Things. The term wouldn’t be coined for another 17 years, when British-born technology pioneer Kevin Ashton first used it while teaching at MIT. But using the definition for the IoT that Ashton –  who literally wrote the standards for global use of RFID and other types of sensors – laid down in 1999, it’s clear that what the Carnegie-Mellon foursome created was one of the first recognized examples of the Internet of Things.

And it’s still a great, easy-to-understand example of what the IoT really is, and, more importantly, what the potential of the IoT really is.

Monitoring the supply and temperature of soda remotely and in real time seems sort of goofy, even frivolous, given how much such a monitoring system would cost to design, build, deploy and maintain relative to the $1 price of the product it sells. But the similar – though vastly more capable – data monitoring devices, communications technologies and analytical tools available today can monitor far more critical systems and sub-systems in virtually any machine and provide critical data to human and or automated monitors in real time.

For manufacturers, that can mean identifying machine wear-and-tear, or production line flaws that can cause breakdowns and inefficiency BEFORE they happen.

For supply chain managers who oversee the flow of goods from raw material to delivery to the end-user, such real-time, accurate and detailed monitoring data can help identify and eliminate costly inefficiency and provide better customer service.

For sales people, IoT technology can mean being able to make specific production, delivery and price promises to customers with great confidence that the company can meet those promises. For research and development professionals that can mean having an unprecedented view into how various materials work in concert with or in reaction to other materials.

Consumer Implications

For car owners, IoT soon will – and in some cases already does – mean that a computer at the dealership can silently monitor their vehicle’s performance in real time and send alerts to get minor problems fixed before they grow worse and cause a serious breakdown.

That’s what happens when machines on which we depend are able to communicate detailed information to other machines capable of analyzing that information and acting upon it.

Knowing in advance that we can get a particularly cold soda by pushing this button on the soda machine instead of that button is of some value. Knowing in advance that a short is developing in critical component of a water pump that must work perfectly 24 hours a day, 7 days a week to keep a coastal city from being flooded is priceless.