The Internet of Things (IoT) has been in the news over the last few days, because an estimated 100.000 internet connected devices were used to launch a malicious distributed denial of service (DDoS) attack, bringing down sites including Twitter, the Guardian, Netflix, Reddit, CNN and many others in Europe and the US. But should we consider the Internet of Things as a blessing or a curse? Can we harvest the benefits without having to put up with the drawbacks?
Let’s start at the beginning, in simple terms the Internet of Things is “a network of devices that contain embedded technology to communicate, sense, and interact with the internet and other devices.”
In other words, it is a way for everyday objects to talk to each other and to you. For example, your refrigerator might alert your smart phone when the milk has run out. The forecasted numbers of devices are staggering. Cisco, the largest networking provider in the world, estimates the number of connected devices to reach even 50 billion by 2020.
From a mind+machine perspective, this opens up a whole new world of promising use cases, with millions of new data sources, connected directly to where the data is being processed. Most of the data from these devices comes in the form of simple sensor readings, such as: locations, temperatures, speeds, vibrations, pressures...etc.
Increasingly this data is streamed, for example a machine sensors might send temperature or pressure readings every few milliseconds or cameras stream video feeds in real time. Streamed data adds a new aspect to data analytics, as it is simply impossible for humans to be involved in the actual digestion of the data since mind is simply too slow to get involved at the streaming level, full stop. However with the right machine, this will allow you to get the information and even insights much more immediately.
So what is the role of the mind in use cases involving streaming data? When the tools required for a use case grow more sophisticated, it also becomes more important to focus on the development and the governance of the use case, and the usage of any Level 3 insights generated. ‘Use case engineering’ might become a new function, with people working on useful Internet of Things opportunities, creating or fine-tuning prototypes that work and discarding the ones that don’t, and then building production versions that run semi-autonomously, supervised by humans that are assisted by machines.
Regardless of how the use cases are managed, it is clear that opportunities in this fields are plentiful and very interesting. However the Internet of Things is still in its infancy in general and before all the promises can be realized, a few fundamental issues need to be addressed.
-
- Security: We need to be aware that every connection in the Internet of Things is a potential opportunity for unauthorized access, especially when the data flows through the open Internet. Cyberattacks like the one last week were only possible because there are thousands of unprotected devices out there which can easily be infected and used for malicious purposes. Before connecting your camera or fridge to the internet, think about who else might be able to access it. How many companies are really on top of this in every aspect of the Internet of Things?
-
- Standards: Interoperability in the Internet of Things is still an unsettled and unsettling matter. At a minimum, this is slowing down the evolution. What language should all these 50 billion devices speak with each other? Not-for-profit standardization bodies in various domains and consortia set up by vendor communities are trying to address the issue. Of course all this will work out in the long run, but there will certainly be a lot of duplication, lack of interoperability, and reduced numbers of choices for customers in the meantime.
-
- Data Privacy: Another major, completely unresolved, and worsening issue with the current Internet of Things is the complex topic of data privacy. What information needs to be protected, why, for how long, by whom, and where?
-
- Intellectual property rights (IPR): This topic gets completely overshadowed by the hype and by the standards, privacy, and safety issues. The Internet of Things will need a very clear and easy-to-implement economic model for owning, pricing, selling, and using various types of Levels 1–4 data, information, insights, and knowledge if participants want to get their proper returns for their ownership rights.
For the sake of brevity, I will have to skip the issues of liabilities, accountabilities and audit trails, but if you are keen to learn more about the Yin or the Yang of the Internet of Things or how mind+machine can help you exploit its potential, why not take a look at my book, where I discuss these topics in more details.