The internet of things – are all the great things we read about real or hype? Can this be compared to when cloud appeared on the scene, and everything changed to xxx in the cloud? We invited some industry experts to discuss a number of questions that are commonly asked. Here are their opinions:
Wikipedia identifies the internet of things (IoT) as the interconnection of uniquely identifiable embedded computing devices. Do you think this is a good definition? If not, how would you define it?
Steven Yates, director of strategy and consulting, BT says he thinks that the description is limiting in what the IoT is and will become. “The use of the term ‘embedded computing devices’ gives the impression of computers as opposed to uniquely identifiable sensors which could be mechanical, computational or biological in nature, but IoT presents their information gathered in a format which is consumable by a collation mechanism. Another element missing in this definition is the intelligent processing for information gathered for disparate sources. This is the major difference between machine to machine solutions and the IoT. In the IoT world one organisation does not own all the sensors, but the information those sensors provide can be extrapolated across many scenarios across many industries, organisation etc. to enable intelligence.”
Alan Knott-Craig, CEO, Project Isizwe has a different view he says “Perfectly fine definition. Another way of phrasing it is “Today the internet is mostly made up of people connecting with one another. Tomorrow the internet will be made of people and things connecting with one another.”
Willem Hijbeek, CEO, Tempe Technologies agrees with the definition but as long as we understand what “computing device” means. “IoT can range from an extremely simple temperature sensor ‘blinking’ once a day to a 32-bit processor controlling a multi-sensor AC powered LTE connected object. I’d rather refine the definition as ‘any object that can feed the cloud in order to generate value for the end user’. IoT is the next evolutionary step.”
Nick Black, EUC business manager at VMware South Africa and WECA added “Perhaps, however, the phrase “computing devices” is partly misleading, considering that IoT as a general rule also includes a host of non-traditional devices such as household appliances, wearables such as smart watches and even headphones, vehicles and naturally mobile phones. Furthermore, if we look at where computing and IT in general is headed as an industry, one can also say that the concept and notion of IoT is more than just the internet ‘of’ things and it should also include the internet ‘in’ things.”
What really is the internet of things? Is it important? Is it hype? Is it practical for business? Does IoT have economic value?
Brian Andrew of RS Components says IoT is basically the technological concept which allows for the transfer of data over a network or the communication between two devices without human interaction. “I do think this is very important and will continuously change life as we know it. I don’t believe it is hype. It’s very important in the advancement of technology both to businesses and consumers. It’s very practical for businesses. Allowing for the transfer of data or for the communication of information can be hugely advantageous. One device can communicate to another about the current status or to report any issues. This can save cost and man hours.”
Alan Knott-Craig has a tougher view of the question “The internet, at its core, is a serial murderer of middle men. By maximising the flow of information, traditional information monopolies have been destroyed, ie: newspapers. This ultimately results in more economic value for producers and consumers. The internet of things will do the same for the world of parking meters, cars, dog collars, etc. The middlemen that exist will be eliminated, the system will become more efficient, and more economic value will be freed up.”
Steven Yates “Definitely the IoT is not just hype, but rather something very practical and of great commercial value. At its grass roots the commercial value will be gleaned by organisations producing the ‘things’. These ‘things’ currently take the form of smart phones, smart electric meters, heart rate monitors, pedometers, weather stations, power meters and many others which are already prolific (and becoming more so) in the world. The next phase is collating the information received from the multitude of devices and creating something unique that provides value. An example to this is the new airport management systems where it is possible to collate traffic information (from Google maps), smart parking information (such as seen in our OR Tambo airport) and personal location information to provide passengers with an estimated travel time to make sure they do not miss their flight.”
Jeremy Potgieter, group manager M2M and IoT, MTN Group in answering this question reverted back to the definition of IoT “The internet of things, is, at its core an ecosystem of connected sensors and devices for delivering real time efficiencies and services devoid of reactive human interaction.
“As we have started embracing ubiquitous connectivity and unified communications, the importance of IOT has gained more interest. Both socially and commercially, the benefits start emerging from the data we share and harness from everyday interactions recorded by devices. The natural evolution of these interactions would be to leverage off what (information) they offer, to trigger events that would logically follow. Herein lies the true value of IOT, practically and economically.
“IoT is no longer hype or a buzzword thrown around by futurists for idle conversation. The reality is, it’s been here, is here and will be here long after we are gone. Take the sprawling connected car environments, smart TVs and appliances or the ever increasing utilities management drives, these are the tip of the iceberg of what IOT will be in two weeks, six months, one year from now. How we as enablers change our traditional models and services to enhance these environments will determine our success or failure.”
Willem Hijbeek: “As long as we understand what ‘computing device’ means. IoT can range from an extremely simple temperature sensor ‘blinking’ once a day to a 32-bit processor controlling a multi-sensor AC powered LTE connected object. I’d rather refine the definition as ‘any object that can feed the cloud in order to generate value for the end user’. IoT is the next evolutionary step.”
Lindsay Britz, marketing manager, Magic Software SA says since the field is so expansive and can include so many different types of implementations, it’s helpful to narrow the topic by defining sub-segments, such as IIot (industrial internet of things). “As we move down this road, with more services taking up more bandwidth, it may be important to define different classes of service or to separate networks so critical services, such as healthcare, get priority over entertainment services for example, when supply is limited. Surely this will involve a lot of discussion.”
Jeremy Potgieter says whether we specify different levels of IoT or use a generic description is inconsequential if we cannot deliver value to the proposition.”We are of the view that a separation of services is an absolute requirement. Machine related communications should not contend with existing internet usage/users. The environment in which IoT operates needs to be as real time as possible, especially when we talk about health care, emergency services and public safety. These services or activities related to them cannot be queued whilst waiting for the next WhatsApp conversation to complete. If the aim is to enhance efficiencies and provide tangible and realistic solutions, then dedicated networks and associated QoS and SLAs are non-negotiable.”
Nick Black: “With technology advancements we inherit and expose ourselves to numerous other risks. You may remember the Stuxnet attack in Iran a few years back – this was the first piece of computer code that caused physical damage to actual centrifuges. As a refresher, Stuxnet was a computer worm that was designed to attack industrial programmable logic controllers (PLCs). The worm seriously compromised Iranian PLCs, collecting information on industrial systems and causing a myriad centrifuges to tear themselves apart. This attack exploited vulnerabilities in an interface that controlled machinery – so when considering hanging industrial devices on the world wide web we need to be sensitive to the risk this introduces. It is clear that if you consider the much bigger picture here this would also impact traditional labour policies and have an impact on unemployment and skill sets. “
Brian Andrew: “I don’t believe it is important to divide definitions for the different sectors. The same principles apply and the concept remains the same.”
Alan Knott-Craig: “My opinion is that separate networks are a nice idea in theory, but won’t work in practice. No one can compete with the economies of scale of the internet, and economics always wins in the end. Security challenges will be solved. The doomsayers will be proved wrong, again. It will be like Facebook. Facebook is a network of hundreds of millions people, grouped into little sub-networks of ‘friends’ with similar values and interests. The iternet of things will be comprise sub-networks of things, but it will still all belong to one big ‘socia’ network.”
Steven Yates says for him it is a definite no. “We are seeing a move away from separate networks as the industrial type networks ‘SCADA/PLC’ networks are replaced with IP networks. This is extremely important from an IoT perspective is as the value comes in collating information from different sources (not all owned by a single organisation). If the ‘industrial things’ all stay on separate networks we greatly limit our ability to integrate the information and create differential knowledge. To give an example; when we monitor a gas pipeline for leaks (industrial environment in mining/petrochemical), if we can identify only a leak we can act on – this is a ‘machine-2-machine’ process. If however the gas pipeline monitoring system is on a standard network and once the leak is identified (automatically), wind data is collated with the exact geographical location of the leak together with the geographical locations of surrounding villages and all of these data are presented in an easily consumed manner, then we are talking about IoT. Thus, not only can the leak be fixed but the environment impacts can be assessed and potential human risk elements can be negated.”
The predictions of how many connected devices or “things” there will be by 2020 vary between 20 to 50-billion. Do you think analysts are simply grabbing at numbers?
Lindsay Britz believes that perhaps globally the predictions are correct. “However, there may be some lag within the South African market. Case in point: enterprise mobility. It took South Africa substantially longer than most other countries to adopt this technological breakthrough. While some forms of IoT are in use in SA, we need to keep pushing for infrastructure improvements, updated educational programmes, and adopt a this innovative change in order to remain relevant within the global market.”
Steven Yates says I think it is very difficult to make a guess as to the number of “things” that will be connected by 2020 because of what we include or don’t include – most estimates exclude mobile phones and tables which should be in excess of 5-billion devices by 2020. “Some estimates are specific to corporate or industrial ‘things’ while others include ‘things’ for personal use like HR monitors, power meters, pedometers etc. The bottom line is that the proliferation of connected ‘things’ by 2020 will be exponentially significant to what we are currently witnessing with smart phones.”
Brian Andrew: “I believe those numbers are definitely possible and that the number of devices will be within that range in the next 5 years. If we take a look at the technology of today which includes things like Live!y (www.mylively.com) that adds devices to everyday things to enable someone to track an elderly parents routines and reports on anything out of the ordinary, or with a simple example whereby our cars tell us when we have a faulty sensor or low on oil for example, I can see those numbers being achieved.”
Nick Black:” The real business adoption of the technology that is driving the IoT really is unclear at this early stage in its lifecycle. So that said I don’t completely understand how this can be estimated in the way that it has and to be honest the range indicated here suggests that this is probably a stab in the dark.
That said, I do expect that the IoT would be adopted in consumer technologies much faster, due to the limited security impact this would have to industry and business in general. We are pretty much there already with home entertainment, smart houses and mobile devices but not as much from a business point of view. From an African perspective we are limited by infrastructure and as such I would question whether industry would adopt IoT in volume just yet, although it is very much something that is being tabled by companies as a future consideration.”
Jeremy Potgieter: ”Moore’s law states that processing power and speed will double every two years, a trend/observation that some will contend was grossly under-spec’d when referring to transistors. Similarly the forecast, based on existing trends may well be below the eventual number in this period. As innovation uncovers solutions to yet undefined problems through all verticals and segments, we may indeed see this number grow exponentially. Especially given the ever decreasing size and cost of components like sensors, cameras and GPS units. Smaller size and cheaper price can only equate to higher deployment in higher numbers.”
Is there a roadmap for IoT?
Steven Yates: “There is absolutely a roadmap which an organisation can embark on with regard to the IoT however that roadmap will be different for each organisation as the reason for that organisation embarking on the road will be different. For example a mining company’s IoT road map will look very different form a smart city’s roadmap and much more different from a digital hospital’s roadmap. The most important element is to be clear in why an organisation is embarking on this journey and what are the business reasons and ramifications of doing so. From there a roadmap can be defined.”
Willem Hijbeek: ”There is a road which is certainly winding and bumpy, but I’m not sure that there is a map. At this juncture the road map for objects connected at ‘short’ range is quite clear namely BTLE, WiFi, Zigbee, etc. The long range AC powered M2M segment has some certainties. The largest segment which accounts for approximately 50% of the market has a need for mobility, long battery life and long range connection. This one segment leans toward low throughput public networks operated by TELCOs with a road map driven by cost. How fast this all of this will materialize is not yet clear but it will most certainly be driven by end user applications.”
Nick Black: “Roadmaps are variable at the best of times, I expect the roadmap for IoT will change and adapt as new use cases are discovered. Beyond that I think vendors and industry players have specific technology roadmaps for how they can introduce it into their product offerings. That said remember the IoT is also an ideal and a concept, it’s difficult to map something that is more of a concept than it is a physical technology. Against this backdrop I think the real area one should look at placing a magnifying glass on here are areas such as BYOD (bring your own device) and EUC (end user computing) and how we can manage the proliferation of these aspects into our organisation. Bearing in mind that both of these cannot actually be separated from IoT as a concept.”
Lindsay Britz: ”Gartner and IDC project a rapidly increasing slope of connected devices. However, there is no one roadmap. It will develop as fast as our imaginations, infrastructure and skills allow. In order to benefit from this growing trend, businesses need to start working with it now to gather experience and knowledge. So, where do you start? Small. Look at your supply chain, your customers, your processes. What would benefit from automation? Once you have your thoughts and blueprints, find a supplier and get your feet wet.”
Alan Knott-Craig: ”It started a decade ago with tracking devices in cars. Every car tracker is a ‘thing’ that uses the Internet to connect.”
If IoT takes off as predicted, are current security systems robust enough?
Steven Yates: “The reality is that current security systems are not robust enough even if IoT does not take off – this is demonstrated almost weekly where another cybercrime is being reported – most recently the theft of more than £1-billion from hundreds of financial institutes globally, before that JP Morgan, Sony, and many others. The technology and expertise exist to manage the security threat, it just needs to have a change in focus at board level and not be considered as an IT security issue but rather be included as part of corporate risk mitigation.”
Nick Black: “In short – absolutely not! IT security is consistently in the top three of CSO’s (chief security officers) and CIOs concerns, this includes featuring at the top of budget considerations. The reality facing executives in this area include the growing proliferation of cyber threats and cyber-attacks. The more these target the man on the street and the device in his hands – the more reluctant IT professionals are to allow these into their business environments. Thus it is safe to say that the IoT makes this an even higher priority. IT security simply cannot deal with the volume of threats that are developed every minute. But then lets look at the consumer angle here and ask yourself – does someone really have an interest in what is in your fridge for example? This is a key-driving factor as to why the IoT, in my opinion, will be adopted in the consumer sector rather than business initially until security is improved or even reinvented.”
Lindsay Britz: ”In order to serve real commercial purposes, IoT applications require integration with an organisation’s core back-end systems including CRM, ERP, finance and billing applications. Therefore, system integration platforms play an important part in the success of IoT applications by supporting event-triggered processes while ensuring the necessary precautions to keep data secure.”
Jeremy Potgieter: “In June 2014 a leading political party website is hacked and defaced; September 2014 the UK’s Home Depot robbed of $56-million through cybercrime; and just in February 2015, the world witnessed what was to be called “One of the largest bank heists in history” wherein a cyber-gang (Carbanak) digitally stole upwards of $900-million. The point I wish to illustrate is that even the most secure institutions like banks, governments and large corporations are susceptible to cyber-attacks where the criminal believes value to exist. We always place extremely high emphasis on securing our environments and endeavour to mitigate all possible security breaches, but the golden thread simply does not exist at this point.”
Brian Andrew: “Like with any internet application or device that is connected there are potential security issues and in the IoT case the actual hardware can be hacked as has already happened. The standards and protection systems will however continue to evolve and improve over time.”
Alan Knott-Craig “Like with any internet application or device that is connected there are potential security issues and in the IoT case the actual hardware can be hacked as has already happened. The standards and protection systems will however continue to evolve and improve over time.”
Willem Hijbeek: “This is a topic on its own. The implementation of security measures for the IoT is paramount to safeguard connected nodes which includes the data being communicated to and from public and private networks. The strategy needs to focus on safeguarding the network infrastructure as well as the confidentiality of the data by means of current and evolving encryption and decryption technologies.”
So IoT is real and we have our 50-billion “things” connected, what would we be talking about next?
Lindsay Britz: ”Who can predict what will happen in five years with any certainty? The personal computer was a game changer. The internet was a game changer. The iPod was a game changer. The iPhone was a game changer. The cycle of life – haltering technology is so short – how can we even conceive what technology will be available and in use five years from now?”
Jeremy Potgieter: ”Once connected we will invariably find multiple uses for them. Hypothetically creating learning machine environments that increase the capability of the existing eco systems. From the IoT we derive even ‘bigger’ data and this needs to be analysed to equate to measurable and actionable information spawning the next generation of solutions. This may not necessarily be actions that relate to physical events, it might well relate to enhanced experiences. The question then becomes what’s after that? Are we on the brink of a Skynet or Matrix society?
Nick Black: ”Health issues and more to the point failing health due to the fact that people do not have to leave their homes in order to get anything done. So issues such as obesity and even socio-emotional issues, where people lose the ability to interact face-to-face, and lose touch with society”
Willem Hijbeek: ”IoT will be weaving itself into thousands of different application affecting everyone. So now we have a massive amount of data, applications and services. Creating value from this huge amount of data is the next logical step.”
Brian Andrew: ”Biotechnology and 3D printing!”
Alan Knot-Craig: “Space.”
Steven Yates: “Next is smart organisations or intelligent organisations (both public and private) where the IoT is utilised to change the way we live, the way we interact with each other across the globe. Very exciting times.”