With better development tools, better products get developed. This is just as true in the brave new world of the Internet of Things (IoT) as it has been for decades in fields like software and mechanical engineering. IoT development platforms will have to improve in response to current challenges and trends facing this revolution of smart devices. To get a sense of what kind of IoT development platform will be able to meet tomorrow’s IoT challenges, let’s take stock of the IoT trends emerging on the horizon.
Fortification and integration
A much heavier emphasis on security will definitely be one trend that will dominate IoT in the near future, and 2017 provided one glaring reason for this: KRACK, the vulnerability which strips the often-used WPA2 protocol used for securing WiFi. This vulnerability in one of the most fundamental wireless network protocols has thrown the spotlight on the problem of patching the legions of computers which have become embedded in our lives: from the routers which bring the internet to every room in your house, to the webcams perched over your child’s crib, to the internet-connected refrigerator in your kitchen. Constantly connected devices will need to be continuously updated or they will be perpetual security risks, just like the millions of already existing and IoT devices in use all over the world which may never receive the updates that patch this gaping security hole. KRACK will be the watershed event for IoT that Heartbleed and Shellshock should have been.
Another IoT trend to watch out for is continued maturity of the whole IoT ecosystem as startups get acquired by bigger companies, protocols get standardized across the industry, and as IoT integration with the cloud becomes more seamless and mature. Since IoT depends on combining the electronics hardware, device software and firmware, and the cloud - each of them complex technologies in their own right - the success of the entire platform as a whole depends on each of those components reliably working together. This is why you can expect more partnerships and integrations between cloud vendors such as Amazon and Microsoft and IoT hardware vendors such as Samsung and Intel.
How IoT development platforms launch smart devices
The piece at which all of this critical integration of the device, cloud, and app comes together is the IoT development platform, which is a bit of a misnomer since the best platforms are used not only for the creation of smart devices, but for their operation and maintenance as well. This ongoing connection to the smart devices after purchase will be vital to rolling out the updates (and protecting customers) when the next KRACK-like vulnerability is found.
These platforms allow a developer to create the code that operates the smart device, either through a full-fledged IDE or a simplified editor which creates rules based on If-Then logic. After creating the code, the development platform provides a way to actually load the code onto the smart device.
Finally, these IoT development platforms pull the data from the internet-connected device to the cloud, from which customers can control the smart device and review its data via a web app. Also connected to the cloud is an analytics dashboard by which the product team can visualize the data and for some platforms even perform advanced analytics on it and glean actionable insights for both customer support and product improvement. The best IoT development platforms going forward will help smart device OEM’s maintain the tight integrations between cloud, device and app by employing analytics which instantly spot technical hiccups across the whole system before customers experience a disruption.
Minding your device’s behavior (analytics)
In order for those analytics to actually improve decision-making, enhance operational efficiency and predict future behavior - all promised benefits of IoT-enabled products - you need to collect the right data, leverage the right algorithms to analyze that data and turn it into actionable insights in time to use them.
In short, you need behavior analytics, which is a lesser-appreciated, but still vitally important component of an IoT development platform. The best way to implement behavioral analytics is through a model-driven approach, which is the strategy employed by IoT development platform experts Seebo. In a model-driven approach, the rules which govern the behavior of the smart device, and the analytics dashboard which visualizes the actual performance of the device, are both generated at the same time. This makes the comparison between how the system is supposed to behave and how it’s actually behaving, really easy.
With behavior analytics, product teams can see which features of the device or pages of the web app are rarely used by customers, and thus can be dropped from subsequent updates and product versions. This focuses resources more efficiently on features which customers actually engage and rely on. The result: higher customer satisfaction and a revenue-boosting, data-driven approach to product support over its entire lifecycle.
Doing your homework
Not all IoT development platforms are created equal, and those which feature behavior analytics have a clear advantage over those that don’t. When choosing a platform for launching your IoT product, it’s important for your whole team to learn the ins and outs of one you’re considering. Your embedded engineers need to make sure the processing power, memory, storage and I/O options are sufficient for workload your product will throw at it. The developers writing the firmware and connecting web app may require certain languages, frameworks and protocols to be supported, so that they can implement code which is secure by design because it employs defense in depth. The customer support and R&D teams will need to assess the scalability of the cloud supporting the device, as well as the type of analytics supported (KPI visualization, anomaly detection, machine learning, and so forth).
There are many things to take into account when considering an IoT development platform – some more obvious than others. All should be studied in depth for a good result, because ultimately, you want your new product to be the biggest IoT trend happening when it launches.
Edited by
Ken Briodagh
