<strong>I busted the 12 - The Myths of IoT</strong> Myth busters IoT Picture1

I busted the 12 – The Myths of IoT

Internet of Things (IoT) is for many technology-enthusiast the single most important thing that has ever happened. The impact of this technology on our lives is growing close to mythical proportions. Working in this industry for years, made me encounter some of the myths and irrational expectations of IoT. Some of these myths are adding to false assumptions, inadequate economic effects, and disappointing expectations for all of us.
I took the task of debunking the 12 myths of Internet of Things. I am sharing these facts based on my personal experience and hopefully, they will inspire us to use IoT as it is intended: creating value for humans.

Myth # 1: IoT is about things.

Saying IoT is about “things” is the same as saying office work is about pressing a keyboard. It is part of the process, but it is not the most important part. The “things” are just the devices generating data about the real world. We need to process this data to generate meaningful information that forms the basis for decisions. And, depending on the decision, an action is generated to adjust something in the real world. And if IoT is implemented in the correct manner the action is executed automatically and generates value in the real world. The value is what counts. This can be a change in a setting for an electrical engine to reduce energy consumption, an anticipating climate schedule to optimize the climate for the increased number of people in the building, or a change in the recipe for a food factory to optimize flavor and nutritional value. Every successful IoT implementation needs to be linked to a use case creating value for some part of our organization or community. IoT is definitely not just about things, it is about generating value in the real world.

<strong>I busted the 12 - The Myths of IoT</strong> Picture1

Myth # 2: IoT is just machines talking to each other

Machines talking to each other (M2M) are as old as the first SCADA and PLC systems. The ability to exchange data and commands between machines accelerated production and optimized the operational processes. However, M2M is not the same as IoT. M2M is mostly localized to the individual machine or a plant. Internet of Things exposes the data and commands to an external (cloud-based) location where it can interact with other (meta) data and logic. It creates the opportunity to compare data from these machines with each other, monitor them from a distance and to optimize the way these machines and devices operate with each other. M2M is a form of IoT, but lower costs of bandwidth, compute, and storage (by introducing Cloud) made IoT far more extensive, advanced and valuable than the existing M2M.

<strong>I busted the 12 - The Myths of IoT</strong> Picture2

Myth # 3: An IoT engineer is just an advanced Software engineer.

It is a myth that anyone who can write software can become a good IoT engineer. It is a common misconception that IoT application engineering is just a complicated form of software development. The contrary is true. IoT involves more competences, and the chain of data is much longer. Software engineers assume hardware, connectivity, software, and storage is just there. An IoT engineer must consider the sensors, the embedded systems, the different forms of mobile and wired connectivity, mass data processing logic, data storage, dashboarding and integration with external systems. On top of this he needs to anticipate defects and malfunctions in all parts of the IoT chain, support fast data processing, security and privacy and low latency high volume data processing. In some cases, with a very limited set of resources. Especially on the edge of a sensor, the IoT engineer is limited to the physical constraints of the hardware and cannot add extra memory or scale up the compute resources. In general, there is no generic role for IoT engineers since the required expertise is very diverse. Most of the time you see engineers that are specialized in embedded systems, cloud data processing, dashboarding or device management. An IoT Engineer is an advanced Software Engineer, and a hardware specialist, connectivity engineer, performance engineer, security specialist, user interface designer, and much more.

Myth #4: The goal of IoT is to collect as much data as possible.

IoT is often associated with big data. This is another great myth about IoT, that the goal is to gather as much data as possible. This is founded in the idea that when you have enough data, patterns and relationships will automatically emerge from this pile of data. Big Data is the idea that useful information can be derived or extracted from a large set of data. With the advent of IoT devices, the amount of data is exploding. Data is being gathered automatically and sensors sample data points that are not needed to support a business case. But who cares; we have a data lake that stores this surplus of data. And in the future, we apply algorithms or machine learning, and a plethora of patterns will emerge. This is the wrong approach; just because you can generate a lot of data, does not mean you have to. All these data points are hampering the view on the business use case and cluttering the data lake. This is a very inefficient approach. IoT needs to be focused on business outcomes and generate data that is useful for the end user. Only data that is useful for the end user applications needs to be collected, not just to gather data for the sake of big data.

<strong>I busted the 12 - The Myths of IoT</strong> Picture4

Myth #5: IoT is easy and cheap

The most common misconception about IoT is that it is simple; IoT makes it possible to deliver a quick application, and business value drops in your lap automatically. Just use a cheap development kit and you will be operational in a couple of hours. This is exactly why so many IoT projects get stuck in the start phase. You can make something simple very quickly but then it gets complicated when it grows. Scaling IoT is hard; you must manage multiple devices, process the data, implement reliability, privacy, and security. And when something goes wrong, and believe me something will go wrong, you need monitoring to notice and analyze errors and act on this disturbance. When it grows even further you must manage device provisioning, data processing, data storage, cloud management, and cost management. It is like an unstoppable train, which is configured and improved by an engineering team while it is running at full speed. Before you know it, you reach the point where you cannot turn your IoT solution off to do maintenance. Suddenly the opportunities created by IoT have become so essential to your business operations, that it becomes mission critical and needs to be available at all times.” The most challenging part of IoT is that you constantly must be one step ahead in a journey that is rapidly growing and fast-changing.

<strong>I busted the 12 - The Myths of IoT</strong> Picture5

Myth #6 Things are smart

Enthusiasts of IoT always talk about smart sensors, smart meters, and smart equipment. This is not an accurate determination. What is called “smart” is in most cases referred to devices that are digital and connected. But to be honest, the things are not the smart parts. The software on those things is sometimes smart, but the smartest parts are residing in the cloud where logic is processing the data from the things and convert them in conclusions and actions. The cloud is the place where engineers apply machine learning and develop algorithms. Actions, instructions, or algorithms are sent back to things to make them look smart. But the smartest thing of IoT happen in the cloud.

<strong>I busted the 12 - The Myths of IoT</strong> Picture6

Myth #7: Once installed sensors will need no updates.

It is a common misconception that once installed, sensors and actuators do not need maintenance. It is easy to manually update ten sensors in your own home, managing twenty thousand sensors is a whole different thing. You are not going to honor them with a bi-monthly field visit. Consider a lifetime of 5 to 10 years for sensors. It is highly probable that in the upcoming 10 years there is a critical security update, changed settings or changed logic required on these sensors. I am not even talking about replacing batteries and other physical maintenance. Visiting the physical location of sensors is the most expensive activity of IoT. This will invalidate your business case. Once installed, the people responsible for business operations prefer remote maintenance on these sensors. Thinking about remote maintenance, monitoring, updates, and patching is a necessity for IoT solutions. This process needs to be foolproof, resilient, and highly dependable. Then build your IoT solution in such a manner that you can frequently update the embedded software on the sensors. Security and cost savings should be sufficient to validate this business case.

Myth #8: Once connected all devices will talk with each other.

Once devices are connected the assumption is that they can connect to each other as well. This is in most cases not true and not needed. For devices to directly exchange data with each other we see the following options. 1) devices from the same vendor usually talk to each other. Especially in home automation it is easier to use the same remote or app to monitor and control these devices, 2) devices with the same type of protocol have the ability to talk to each other, 3) when required for a use case you can make devices talk to each other when it generates value, 4) devices from totally different domains will not talk to each other. There is no use case for your smart car to notify your smart TV that you are arriving (at least not what I can think of). Cars may talk to each other, and media devices in your house may talk to each other. I do not see a future where all devices will talk to each other. The market is still very competitive and scattered that a generic communication protocol is still far away. Except for home automation where a generic communication protocol from different suppliers of hardware is emerging. The diversity of use cases does not require devices to talk to each other at all. Devices will eventually be logically grouped per domain. Standards for communication are dictated per logical domain and communication protocols will be derived from them. It will be an illusion to see all IoT devices work together.

<strong>I busted the 12 - The Myths of IoT</strong> Picture8

Myth 8: Everything needs to be connected

Industry experts will certainly state: Everything will be an IoT device, and all assets needed will be connected. By 2025 there will be 27 billion IoT devices active. This does not mean that everything needs to be connected. It is a matter of timing and costs. The costs for a single IoT device are still too high to validate a positive business case. There is no need to connect everything just for the sake of it and there is no magic formula to create positive use cases derived from the fact that things are connected. Be careful connecting everything and assess the business case upfront. We still do need parts of our world to have a low technology density.

<strong>I busted the 12 - The Myths of IoT</strong> Picture9

Myth 9: Scaling is not an issue with the cloud

The whole purpose of IoT is processing device data on scale. Cloud technology is applied to enable this scale since cloud is the most effective way to dynamically add resources (compute / memory) to applications. This does not mean you have nothing to think about when you want to scale IoT solutions. There is no magic dial you can turn to make your 10-device IoT project suddenly process data from ten thousand devices. The challenge of IoT scaling is a diverse problem with several optimum and tipping points depending on the number of devices, level of reliability and total size of the processed data. You can characterize these tipping points as paradigm shifts where it was obvious you used a certain solution for a specific task, till you reached a certain size and then it becomes nearly impossible to continue doing the same task and you must apply a completely different approach or design. Take the number of devices: when you have less than 100 devices it is not hard to register and provision them manually. When there are more than 1000 devices this becomes a challenge, and you want to prepare registration/provisioning in batches. When there are more than ten thousand devices you want to implement self-registration and provisioning. The same for data processing: when you scale up you need to use different and more robust components to ingest, store and process data. Some components are simply too expensive for a small number of messages, but they are essential when you scale too many devices. Then it becomes important to queue messages, and implement parallel processing, caching, and cleaning messages before processing them. Without expertise, scaling will only grow your cloud costs exponentially, and that will invalidate your business case. Considering scaling IoT solutions, you always must consider the next tipping point and design the next paradigm for handling IoT data at scale. Only this approach prevents your IoT solution clogging up and becoming unusable.

<strong>I busted the 12 - The Myths of IoT</strong> Picture11

Myth # 11 IoT is technology innovation with things

It is true that most IoT is seen as something belonging to the innovation department where they can tinker around with sensors, circuits, and cloud components. However, IoT is not about experimenting with technology. Unless you are working for large OEM manufacturers or a research organization, most IoT projects are not about technological innovation. They are revolving around business model innovation using data. In most cases IoT is the foundation for a new business model or an industry innovation using devices and data. Technology is (sometimes) complicated, but the real challenge and innovation is to be found in the changing business models and organizational changes.

Myth #10 IoT Security is just like IT security.

It is a myth to assume regular IT security is sufficient for securing IoT solutions. Regular security is the junior league of security compared to major league of security for IoT devices. Your “things” are left somewhere in the field (on a truck, office building, public space, a factory or in a domestic residence) with no supervision. This “thing” is potentially connected with the core of data processing for your organization and decision-making. That means anyone has unlimited time to research, open, modify and replace your “things” with anything that can harm your organization. This could result in sending modified data, hacking attempts, denial of service or transforming your edge devices to a digital zombie army completely out of your control. When it comes to IoT security you must work under the assumption that your “things” will get compromised. If that happens, you must be prepared to assess the impact quickly and focus on a quick resolution. Design with security in mind, especially when you also use actuators to modify the state of things in the real world. Where an IT accident could lead to some loss of data or a system administrator working late to restore your database, an accident in IoT can lead to dangerous situations in cities, plants, traffic, and the energy network. The impact is much more severe compared to “standard” IT systems.

Myth # 12 IoT does not work with legacy assets

Some business leaders believe that legacy (older) assets are not suitable to integrate into their digital transformation. They think that all equipment needs to be replaced with new (digital) equipment before they can start their digital transformation. It is an illusion that you can wait till a greenfield situation to implement IoT. Most (all) cases are implemented in a brownfield situation: connecting existing asset infrastructure and optimizing them using the data and adding new (digital) machinery on an ongoing basis. Adding sensors to existing equipment can generate so much valuable information for the organization. From the operational shop floor, in the value chain, in optimization programs, and in management reporting. Simply adding a movement sensor, energy sensor or temperature sensor on legacy equipment can generate sufficient data to compare the performance between these machines, detect extreme differences in energy efficiency, and plan maintenance on actual machine usage.


I am imagining there are many more myths of IoT to bust. I am open to discuss them in the comments below or send me a direct message. The main conclusion is that IoT is a serious profession that needs serious expertise to operate. The main driver should always be the business value that IoT brings for individual organizations, businesses, or the whole society. This value can be categorized in 4 simple groups: cost reduction, increase of revenue, higher reliability, and sustainability.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.