Get started Bring yourself up to speed with our introductory content.

IoT: Forever? Or should we plan for end of life?

Google Nest’s shut-off of the Revolv product it acquired in 2014 is just one example of what may become a common occurrence. E-consumers expect everything to be warrantied forever and always connected. When those devices are on subscriptions or contracts, as they often are, they can be kept functioning, supported and moving forward. But, as demonstrated by the Revolv bombshell — and that’s not too strong a word, if you are, I mean were, a Revolv customer — that’s not always the case.

Even in the industrial world, embedded systems often fail to be updated, and oftentimes become old and less possible to secure. This can be seen in old SCADA systems that manage large multisite industrial complexes, providing data collection and aggregation. SCADA systems were most often air gapped — not connected to the outside world — making security far easier and more realistic. These systems in many ways are the predecessor to IoT. By connecting these systems to the Internet — creating IoT — the security issues are endless, and now they must be updated, patched and protected.

When we as consumers buy something — whether it be a computer, phone or other device — we expect it to run forever. Or do we? Tech-savvy consumers recognize that most devices have a finite lifespan before the hardware or software is outdated or inoperable. We’ve all seen the phone that gets too slow and the computer that crashes too often. These are not just due to the age, but the complexity of the software as you upgrade to new versions with additional features that require more horsepower and elbow room to function well.

This changes considerably with IoT, as the device has a very focused function. In the case of the Nest Hub, it was a simple gateway controller. This means that it helps with home automation tasks. In order to function, there are costs paid for by Google (Nest) associated with securing and upgrading the gateway, hosting the Web services, and maintaining the application and infrastructure. When this becomes less strategic, companies will force an update to a newer platform and deprecate old services or devices. For example, if you have an iPhone 4, you cannot run the newest operating systems from Apple; you are forever stuck on iOS 7.1.2 with its security issues. Remember the iPhone 4 came out in 2010, less than six years ago. Similarly, if you have a PC with less than 1 GB of memory, you cannot run Windows 10. These are both good reminders that often times older devices or hardware are not supported forever. The same is likely true for IoT, especially consumer-facing devices with shorter lifespans.

End of life occurs based on the strategy of the company, the cost and complexity of supporting the technology, and the need to eliminate technical debt in order to innovate. With vehicles, for example, you pay a subscription to get the newest software, navigational maps and other services. Alternately, you can decide not to subscribe, and your car still functions with less accurate, less up-to-date software and systems. As these cars become increasingly connected, we will see the same types of issues we saw when SCADA systems became connected — namely, increased attacks and issues — which may impact safety. In order to mitigate this, manufacturers of vehicles may either offer subscription services to keep the vehicle updated and secured, or risk having upset buyers who own vehicles that are unsafe or less functional.

In enterprise software and hardware, you see various strategies employed by vendors. Many will offer specialized support for end-of-life products, where the user of the software or device pays an incredibly high yearly service contract to keep it functional. Otherwise, you see end of life occurring regularly. Most vendors publish end-of-life timelines to allowing customers to assess the risk.

This is the IoT buyer’s dilemma we find ourselves in today. The questions remain: What model will dominate in the future? Will it be subscription-based? Will users expect software and functionality for life? Which model will present the most compelling business case and ultimately win out? We need a connected crystal ball to see the answer!

All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.