Podcast #31: Connecting the Value in IoT – by Syed “Z” Hosain

Good day everyone, this is Ed Maguire Insights Partner at Momenta Partners with another episode of our Edge Podcast. Today we have a special guest with Syed Z Hosain, who is the CTO and founder of Aeris Communications. Z, it’s great to have you with us today. 

Thank you very much Ed, I enjoy being here. 

What we’d like to do to dive right into the conversation is, to get a sense of what’s shaped your view of the Internet of Things, and share a little bit of your background and what has brought you to here? 

I don’t want to make this a sales pitch but let me explain a little bit about what Aeris does, and why we have achieved what I will call a very excellent success in the market of IoT, and prior to that M2M. We’ve been in existence for a number of years, and frankly before the terms M2M and IoT even came into existence, we didn’t originate those terms, but to be honest I wish we had! I guess we were originally a technology looking for a place to happen, and we had an approach to doing data transmission back in the analogue cellular days, which allowed people to eliminate the need for dumb modems, the using of traditional communication methods and protocols on the cellular network, which were measured in voice minutes that could get incredibly expensive.  

So we came up with an approach which allow companies to migrate to an early understanding of small amounts of data on the analogue system, and the analogue cellular system in North America particularly in the United States, and then Canada, and Mexico after that, which allowed them to understand the value of communication, rapid communication, in a matter of seconds, no matter where the device was in that footprint of North America. We stitched together a network of carriers, this is prior to the large United States-wide, nationwide carriers that exist today, and created a network which allowed this data to be sent transparently no matter where that device was, whether it was in Montreal in Canada, Mexico City, or Atlanta, or San Francisco, it behaved the same way.  

That I believe was a very early way of saying that data from these devices, remote control of these devices in a matter of seconds had tremendous implications, in the alarm and security business for example, as well as in the trucking industry, the early trailer-tracking industry for the customers that we had at the time. When the term ‘machine to machine’ was coined, and it’s not original to us unfortunately, I’d love to say I’m the originator of that term, but it isn’t; it captured the imagination of industrial folks, engineers, but it didn’t capture the imagination of the consumers at large. 

The gentleman who came up with the term ‘Internet of Things’, which I mention in my book by the way, so you can download a copy of that from our website, he came up with the term and I think that has become a very nice way of capturing the consumer imagination; ‘Ah, I get it!’. We have the public Internet for websites, for customers to communicate with each other, for social media, for companies to deliver the message of what they’re up to, what they do. What we’re talking about is a network of devices, things, smart things, capabilities that can now be using the internet to provide a function for general benefit of mankind, (I’m exaggerating the point, only a little bit frankly) which allows people to take advantage of information that they might not have had access to before. 

That was wonderful. I think that revelation if you will, the imagination that has been captured in that term, really leads to what could be done in the future. We have seen an explosion of markets that take advantage of what the IoT, the Internet of Things, can provide to them, such as in the healthcare industry, in the connected car industry, the ability to get information about your car, the ability to get information about your home, the ability to access information from things in ways that makes us a better society as a whole; the healthcare devices, Fitbit devices, fitness devices that people can then use to gather data about how well they’re doing personally, and send that data somewhere where they can analyze it later and say, ‘You know what, my data from a month ago showed that was in better physical health than I am today, maybe I need to take some action’. Or, I’m using a medical device such as a wireless communications blood-sugar monitoring system, which do exist today, and now I can send that data to my healthcare provider, so, they keep track of what it’s up to, and they can contact me if there’s some issue associated with my blood sugar etc. So, I think we have created a new paradigm where information from devices, smart devices, smarter devices, are starting to make an impact on our daily life.  

We have a gentleman in our company who is also focused on the social good of the IoT, particularly in third-world developing countries, and I’m very-very pleased to see the kinds of stuff that companies are coming up with, where we can benefit people who are not necessarily as well-off as we are in the US or European markets, who need the benefit of automation that can help them. For example, if a water pump breaks in the middle of a village in sub-Saharan Africa, it gets reported. So, it doesn’t take two months to fix it, it can be done in a matter of days when somebody goes out and says, ‘Ah-ha! This mechanical water pump is not working properly’, a wonderful-wonderful benefit that I think the Internet of Things is shaping for us in a variety of different markets, it’s got to be the next wave. We’ve talked about the waves and revolutions in the past, I think this is the next wave of things we can do with the IoT that is simply not limited by anything that we can imagine today, in the future. 

That’s tremendous, and I think we’re very much of the same mind in terms of our view of the potential of connecting the physical world.  

Let’s go back to the origins of Aeris, and some of the initial technology challenges that needed to be solved, and how did your own personal background lead you to identify where there was a market need, and ultimately that you’d focus on the technologies and business problems which were at the origins of Aeris? 

When we first got started, analogue cellular usage in this country and developed countries in Europe etc., wasn’t very high, they’re what I would call today is now a saturated market with everyone having a cell phone, everyone having a smartphone, this didn’t exist back in the early to mid-nineties very few people had cell phones. The need for communication, being able to get access wherever you were from a voice perspective, was an early data perspective such as text messaging simply didn’t exist, so, there were technological barriers from that perspective. You had a nation which had a number of small carriers, so one issue from a business perspective that we encountered is, we had to have nationwide coverage, footprint, where a customer deploying an application device, one of our early customers basically said, ‘We have an alarm and security unit, the fact that we ship across the entire United States means we put a product into distribution, and then we have no clue where it’s going to get powered up. You need to have a network that allows that unit to transmit its data, no matter where it is’. 

Well, we had to run around sticking together the business agreements from a number of different cellular carriers of the time, there were over 35 to 40 of them in the United States that we had to light up in a consistent way, meaning that we’d go to them and say, ‘Hey folks, this is a business application, it’s not a person and a human and a voice minute, we want to send data in the analogue cellular system, here’s how we’d like to do it’. They were very-very good thinkers in many of those companies that recognized the need for this kind of application, and basically supported us. In the early days… I’ll pick an example and a name if that’s okay. 

Please, yes, I’d love to hear. 

Bell Atlantic NYNEX, the CTO was Dick Lynch, and we went to him and said, ‘This is what we want to do’, we were absolutely, absolutely pleased by the fact that he said, ‘Great idea, let’s make it happen’. So, they were one of the earliest carriers who supported us, and without the personal background and support of Dick, we would have difficulty getting off the ground. So, there are people that we’ve encountered over the years who have helped us in various ways, Ed Reynolds who was the President of BellSouth who helped us get into BellSouth at a time when people said, what is this data, what is IoT, what is machine to machine?’ I think with the hindsight that I can apply, we always had some internal champions at the various places that helped us overcome difficulties. So, we got our network stitched together and we went from there. 

What were some of the initial challenges that you faced in trying to realize this vision of a consistent network. I think you had to deal with a lot of different companies, a lot of different cultures, a lot of different technologies; in your view what were some of the challenges that you overcome, and were there some lessons that you learned which remain relevant and applicable today? 

Absolutely. For example, one of the big differences between us and the traditional cellular smartphone and cell phone industry, is the customer relationship, the relationship between us and our customers is not a one-to-one, or semi one-to-one relationship with an individual who has a smartphone in the consumer space, or even a small family with 3, 4, 5 up to 10, or fewer devices that they interact with. We early-on recognized that we expected to add hundreds, thousands, hundreds of thousands of devices into our network for a single customer who had an application that they wanted to deploy nationwide. So, the relationship had to be very different, you sold the customer on the benefits, enterprise customers on the benefits of what this data could do for them. Then you had to put the systems in place for us to deal with hundreds, thousands, tens of thousands of devices at a time, for activation, for provisioning. 

We had an early customer for example who was doing on the floor factory testing of their product, it took them about five minutes to get the product tested after manufacturing, they said, ‘We want to do live, over the air testing, without having to call you’, which was the traditional way you got a cell phone activated, ‘We want to be able to send you a request on the Internet at the time, a connection to you where you provision that device, activate it on your systems…’ because we had our own network, ‘… and let us know what the number is, we will program it during the test function, and we will do a live over the air test’. We said, ‘Great, how quickly do you want to do that?’ they said, ’30-seconds’. Wow! Imagine provisioning a cell phone 25 years ago in 30-seconds or less. We set a design target for ourselves of 10 seconds, and we achieve in nominal operating delay, from request to provisioning live on the air of about two to three seconds. That capability is something we pioneered and has stood us in good standing with all our customers, even today, where we can provision and activate, and deactivate and control a device literally within a matter of seconds. 

Another example of that is, we had a scenario where a customer came to us with a seasonal application, notably water irrigation monitoring where the systems get turned off for months at a time in the winter when they’re not being used, then they activate in the early springtime when they start being used again. They came to us and said, ‘You know, we cannot afford to go out and touch these devices, change the numbers and reactivate them. Can you make a system for us where between the months of…’? I’m making this up, ‘… November and March, we will not incur a bill because we’re not using your network, but we don’t want the number to go away, we want to be able to deactivate the device and reactivate the device’, we said, ‘Sure, lets figure it out’. 

We put together a system which they could through an API call, or using our portal deactivate the device, and then in March when the device was turned back on again because they applied power, it would automatically convert from a suspended state, where it would retain its number, to an active state and started working, just completely normally and without any hiccups whatsoever. That was another thing we pioneered is the ability to suspend the device, the concept of being able to turn off a device for months at a time, and not lose the ability to reactivate that device when it came online later-on. So, there were all kinds of neat little interesting things that we were able to do. 

By the way, the reason we were able to do this, and I want to emphasize that, is that in the smartphone arena where people typically pay $50-$100 on average, revenue per month, the expectation of service cost is a little different than when you have an IoT device where you’re paying a lot less than that, and it isn’t just small percentages, it’s a lot less than that, the average revenue per unit for an internal IoT device is tiny. So, can you operate efficiently, effectively, with good profit when you’re generating revenue that is tiny compared to a smart phone? That led us to requiring automation, and that was a guiding principle that our early founding CEO put on me, and I was running engineering and operations at the time and said, ‘How can we make sure that we operate successfully with low revenue, but with good margin, decent margin profitably, and make it happen?’ My instant response to him was, ‘You cannot have people involved, it has to be highly automated’, and that has been a guiding mantra right from day one. So, we automate as much as we can, we automate. 

The minute a customer has to call us, and we have to provide human support for either one device, or ten devices, or a market, it adds to the cost burden, so we try to automate as much as possible. 

As you look at the advances in technology which have certainly made the cost of sensors and processors a lot cheaper, essentially the ability to generate data, what stands out to you as some of the more impactful developments in technology, which have really helped your customers begin to realize their vision much more effectively, and help you as a technology provider deliver more robust solutions? 

I think if you look at the way you we serve our customers, and by the way I should step back and add one other comment which would lead to the answer here; we provide cellular IoT connectivity, that was how we started. We’ve expanded from that to data, we’ve expanded from that to content analytics; previously it was just transport of data, now it’s on behalf of our customers, we look at the content, analyze and predict what’s going to happen. All the things that predictive analytics can do for you, business intelligence, you name it, the traditional phrases associated with the wide data and information, analyzed data, analyzed data which results in information is of value. 

I think the most interesting impact that I see has happened in the past between the two to ten years, and going forward, is the development of new sensors, particularly in the healthcare industry. The fact that we now have devices that are going to allow data to be gathered from the human body, using embedded sensors, skin implants, or even going beyond that, there’s some interesting things going on with measuring devices inside the eye-ball for example. I think that is going to revolutionize how we as a society will benefit, because healthcare is becoming such a paramount cost issue for everybody, and the fact that we can now do some things that we weren’t able to do five, ten, fifteen years ago, and transmit that data to the healthcare providers who need to know that data, is a big deal. 

So, I think the revolution in the sensor markets particularly for healthcare, is going to be the next big thing if you will for IoT in general. Healthcare by the way is the most rapidly growing number of units on our network today, although other older implementations such as the connected car was a natural for the mobile cellular industry, healthcare appears to be the next wave. 

That’s interesting, because healthcare as a highly regulated industry always comes up as one of the areas, or one of the industries that’s been particularly challenging for technology vendors. Could you share some of the experiences that you’ve had in working in that industry, and has anything changed to boost your optimism? 

Definitely. I like to separate healthcare IoT applications into two classes, one is what I call the mission-critical applications, and then there are the non-mission critical applications. What’s the difference? Well the mission critical application is one in which the technology, the transport, the analytics is so vital that if one of those elements breaks a human is put in harm’s way. That’s a concern, meaning that if I was doing a defibrillator that is implanted in a person’s chest, and that requires me to have data on it continuously and there’s some problem, I would be very reluctant about saying, ‘That’s a perfect IoT application’, because if there is a failure in the system, which we are still evolving, still learning, still making improvements on, we’ve put a human at harm’s way, and of course worst-case scenario where you put somebody who ends up with a fatality, and I don’t want that to be the driving force behind healthcare IoT. 

So, if you look at the non-mission critical scenarios where you might have a person who is doing a Fitbit kind of solution, where it’s a healthcare fitness tracking monitor, or its doing a blood sugar monitor which isn’t just a passive measurement that is reported to the person taking their blood sugar, but is reported to a healthcare provider who gets access to that data. Maybe, just maybe, it says, ‘Hey, you haven’t taken your blood sugar measurement in a week’ call them back and say, ‘What are you doing?’  

There was an interesting example of a new start-up, which I’m hoping they will be successful down the road, which was looking at diabetic patients, particularly in the low income community, because one of the biggest problems with people who don’t have enough money to worry about food, is that they don’t get their quality of food as well as they ought to, they go into the fast food places because it’s easy, and you’re seeing a massive change in their blood sugars because they aren’t watching what they eat. Well, they were going to try and tackle that by taking the blood sugar data, their weight, their blood pressure, sending it to a healthcare management system, not necessarily a doctor, who would monitor that and then work with the local cities and counties to provide food stamps that would only allow them to go and buy healthy food, so that they could be in control of their diabetic scenario, rather than ending up in an emergency room scenario, and placing themselves at risk for worst case scenarios for diabetic patients. 

That I think is a wonderful example of where information gathered from a remote device could play a role in managing healthcare and managing the betterment of the person who is being measured in that case. It’s amazing to see what possibilities exist, we’re limited by the imagination of the ideas that we can come up with, and I don’t mean ‘we’ as at Aeris, I mean the people who are implementing these kinds of solutions. 

Those are amazing use cases. I do think what’s been so remarkable about the vision that’s evolved around what we call Connected Industry, or Commercial Industrial Internet of Things is really world changing on so many levels, so many dimensions.  

I did have a question about the market itself, it has gone through some ups and downs as it were, and I’d love to get your view on the sentiment in the market. If we go back about four or five years ago, when Industrial IoT hype was at its peak, and of course you as a true veteran, an OG of M2M, you’ve seen these same technologies when they were called something different. There was a lot of hype, a lot of ebullient forecasts of exponential growth, and then I think the reality was a lot more deliberate than that. I’d love to get your perspective on what the impact of that initial hype cycle about five years ago was, and where we are today, how things have panned out relative to some of the most aggressive expectations, and how the actual market itself has evolved. 

One of the comments I might make about the aggressive hype has been indeed the fact that people expected way too much, too soon, too quickly, and forgot about the real-world experience of what it takes to get an IoT application done. If you’ll pardon my using a very simplistic analogy, let me give you an example of what it takes to develop and deploy an IoT application, it has nothing to do with technology, it’s purely a business case issue.   

Why do companies exist? I’m going to be very simplistic about this. Companies exist either to increase their revenue, or to lower their cost; whatever they can do to make one of those two things happen, or both which increases the delta is their profit. That’s how they make business happen, that’s how they operate, that’s how they grow in the future. Well, IoT solutions in general will help those if you can prove it, so once you figure it out, where is the return on that investment for an IoT application, whether its internal cost reduction, or increasing revenue because you now have product which you’re selling to the market, you’re going to be better-off. 

There are other drivers, that was a very-very grossly simplistic view and I’ll be the first to admit that. Regulations, whether it’s an industry regulation, or a government regulation, comes into play. One of the earliest examples that I can point to of what is now the smart meter reading industry, was a company who came to us about 23-years ago and said, ‘We’ve just had this county…’ I forget if it was a county, a city, some locality, ‘… on the east coast, which has legislated that, ‘They shall provide peak, off-peak, utility rates that are different for residential consumers’, and they said, ‘We can’t do this. We go out there and we measure the meters mechanically with a meter reader who walks around once a month and takes a reading. We can’t do these two or three times a day, there’s no way we can solve this’. Frankly, that led to the growth of the utility meter reading industry, where the ability to follow regulations where consumers were being benefitted, customers were being benefitted by the ability for data to allow a better rate, a public utility commission managed rate from the utility which would give them  off-peak billing, and enable more efficient use of energy, allow people to off-set and shift their usage to perhaps off-peak times because money talks, incentives are good, that’s an enabler function for IoT in the utility market. 

You can go on from there, you can take other examples where that kind of benefit is a big deal. If 15, 20, 30-years ago, flights and airplanes were frequently delayed because of a ‘mechanical problem’. Well what the heck, when a part fails that’s when you go look at it? No, you try to figure out before it fails so you can take some preventive action, and make sure that the part is either replaced, or appropriately serviced so that parts don’t fail as much. Well, that has happened, aircraft now report what happens during flight; we’re working with a couple of manufacturers who are basically taking data, and understanding what is likely to fail, not just what has failed but what is likely to fail. Should I hold the plane, should I tell the pilot, ‘You’re good to go, but we will have a maintenance tech, or maintenance engineer waiting at your next port of call, with the right part to replace so that we can get you on your way as quickly as possible’. You don’t have to figure out what’s wrong, we’ll tell you what’s wrong and we’ll fix it before you have to worry about it. 

The mean time to prepare for an aircraft has gone down so dramatically that I have yet to hear in the past year or two, anybody, any airline, and I take a lot of flights, tell me we have been delayed due to a mechanical problem on that plane. That’s the kind of revolution of efficiency, and better use of assets that IoT is enabling, all the time. 

Another quick example I will point to. About 15-20 years ago, we were in a building where the HVAC system on the roof of the building failed, and the offices got warm, but we didn’t know about that until they got really warm. Then the guy came out and figured out, ‘Oh, you know what? This compressor fan failed, so the compressor shut down’, now he had to go back to get the fan, then come back to fix it. Well, a $20 part resulted in a compressor failure that was well over $5,000 for that commercial HVAC system. What if that fan had been reported by the system saying, ‘Hey, this fan is no longer on, the compressor is going to fail. Come out and fix it before the $5,000-part fails’. That’s the kind of things that IoT is solving for people today. 

They are some very powerful examples of the kind of changes we’re seeing. What are some of the notable characteristics of projects that have been able to succeed early-on? There’s a lot of discussion about the prolonged amount of time that it takes for proof of concepts to be successful, and I would say the more measured pace of industrial companies, particularly in rolling out your connected technology solutions. Are there some examples of companies that have been particularly successful with pilots? And then along the way being able to establish best practices, or templates as it were, that have been able to accelerate subsequent similar projects? 

Yes, that’s a darn good question, and I must admit that I’m not necessarily the right person to answer that question. I have seen the reports that say 60-70 percent of IoT applications remain stuck in the proof of concept phase for a long-long time. I don’t see that, because one of the things we like to do is basically tell people to figure out what the heck you’re trying to do before you get started. What is the outcome you want, what are the metrics you’re going to apply, what’s your ROI? It’s a little fuzzy, I will be the first to admit that sometimes it’s a little fuzzy, until you get there you won’t know, but figure it out. 

There are some difficulties with that statement I just made, IoT projects do take time, it’s not an industry where you can buy an off the shelf solution all the time. Many applications, particularly at scale are going to require design efforts to reduce the costs, reduce the size, reduce the power they might be using etc., and it takes time. What often tends to happen in the scale of things we do today is, that the people associated with the project are pulled off elsewhere, maybe the project leader has gone onto a different company, etc. etc. So, you end up with a scenario where things stagnate, not because of the lack of success of a project, but sometimes due to external factors such as personnel being moved, projects being changed, the focus of the company might change, that’s rare but things like that happen. 

What has worked well for us and usually helps any of our customers is, there’s an internal senior champion who says, ‘A-ha, this is going to help us figure out the ROI, figure out the solution, and make it happen’, then gets that pushed through in the ways that a typical CIO for example would say, ‘This is a project that’s worth doing, because it’s going to save some money for us’, then proposes it, champions it, and makes it happen. That’s not always the case but certainly one of the successful ways of making things happen, is to get an internal champion. 

You’ve mentioned your focus has evolved from managing connectivity across multiple networks into analytics, I’d love to get your perspective on some of what’s been driving that, and how this has changed the way you interact with customers. What have been some of the inputs and some of the considerations that you look at, as you start to incorporate advanced data analytics onto your solutions? 

The reason we moved from connectivity… beyond connectivity is the word I would say, not ‘moved from’, beyond connectivity to supporting the customers with analytics is, for a variety of reasons that has been impacted by evolution and changes in the industry as a whole, the existence of cloud solutions which one could argue is basically servers that are not in your own data center, they’re somewhere else for someone else to deal with the headaches, but that’s fine; the point is that network speeds, performance, cloud solutions where you can rely on somebody to store the data for you, worry about the issues of poor tolerance, worry about the issues of power distribution, worry about the cost of data centers, is somebody else’s headache, has revolutionized the fact that we can gather data, put it somewhere, and analyze it. 

So, we recognized this about 10-12 year ago, around 2006-2007, we said connectivity is great, it is becoming easy relatively speaking, to do. We can manage the devices, we know how to get the data from point A to point B, what the heck can we do to help our customers analyze that data. It was driven by a customer who came to us and said, ‘I’ve created a solution, I’ve got a $1 million data center that I’ve put together, and now I have a problem with geographical fault tolerance’. Sure enough, they had a data center in one place, and they wanted to have the ability to put up another data center in another place, and they said it was going to be cost-prohibitive for them to do it, because they simply hadn’t paid off the original datacenter yet.  

We said, ‘Great, let us take on the burden of managing that for you’, because our network solution, our infrastructure, our deployment of the products and services we sell, required us to have multiple datacenters, and we have them, and we said, ‘Can we help you put your systems into our datacenter?’ That was great, step two was, ‘Ah, it’s now in the cloud, we can make it happen in a very-very nice fault-tolerant redundant way for customers’. Then the natural evolution of that was, ‘We’re transporting and storing the data for you, what business function does that data serve for you? Can we analyse the data to give you the information you really need?’ Meaning, if this threshold, or this value changes to the point where you want it reported it to somebody who could take some action, can we do that on your behalf? 

So, we put in all those systems, all those what I would call a streaming data analysis, not necessarily predictive data, analysis where we were starting to go into the content on behalf of the customer. We never go into the data without the customer being aware of what it is, and we do it on their behalf. We help them with fault-tolerant data solutions which allows them to say, ‘Somebody is managing my IoT data for me. I will deal with the application, the sale, the reporting of the data that matters, the information that matters, and then use that in my business’. That’s what we started out doing about 10-12 years ago, and we evolved it over the years, to where we now have cloud analytic solutions, and data storage solutions, data analysis solutions, where frankly we have some customers who don’t even use our connectivity services, they come in because we understand IoT data, and how to analyse it. 

Have you any views on the potentially transformative, or impact of machine learning in AI? And to what extent is that playing a role in accelerating time to return, or business value amongst your customers, and the partners you’re working with? 

Not enough yet, let’s put it that way. I think that’s still a very… I don’t want to use the word ‘research.’, but that’s still an area of expertise that the value of that has still to be proven for our customers. Which is not to say that there isn’t value there, I really do believe there is, the ability to respond in an automated way, which is where machine learning could come into play so that humans don’t have to be involved in making some of those business intelligence decisions, and the actions there-of, makes a lot of sense. I have not personally looked at that enough to be able to answer your question in a more meaningful way than that. 

That’s an interesting answer though, because you guys exist where the rubber hits the road as it were, and of course in the technology industry we’re always enamored with the brightest shiniest objects, so certainly it’s no doubt that there’s a lot of potential! But again, getting an understanding reality where customers are, I think that’s a very helpful perspective. Are there some technology hurdles or enablers for adopters that you think may be under-appreciated?  

That’s a very good question. To me, IoT scale, if we’re going to achieve those billions of units, then we need to be able to put the processes and systems in place to be able to grow at scale, deal with the data at scale, in ways that we’ve never imagined before, for the IoT world, let alone all the data we generate elsewhere. I’m about to use a phrase that is not original to me, I wish I knew who had done it, I wish I could take claim to it, we are starting to get to the point where we are creating data museums, and that’s a phrase that I love; meaning that we gather data and we stick it somewhere, where the value of that data erodes over time, sometimes to the point where its meaningless.  

An alarm company once told me that they wanted us to make sure we had perfect fault tolerance, which is tough to do, because the value of an alarm five hours after the event has taken place in a residential or business alarm system, is absolutely of no value. They were doing traditional systems where 30-seconds to a minute was a big deal, and if we were reporting the data to them in two or three seconds then they loved it. But, if we had an outage they would never see that data, ‘Don’t tell me the data after the fact, because it has no value’. So, if we start sticking lots and lots of IoT and other data to data museums, and we don’t analyze that data to get the relevant information out of it, first it was a waste of connectivity, storage, transmission and cost, and second… and this is now my phrase, going back in there, looking at that data with archaeologists is meaningless. If there’s value to making pattern recognition happen so you can predict the future, that’s okay. But if the value of the data, and you never look at it, you don’t go in there and analyze that data, the point was lost. 

I think we need to learn that, those things need to be managed a lot better, we are in a world where somehow gathering data is more important than looking at it, and I wish people would get away from that thinking, and look at relevant data, look at analyzing that data, and taking the actions necessary to deal with the outcome of why we gather the data in the first place. 

I think that’s a great point, it really does cut to the value, the business value, and the reason for collecting data, so it does have to deliver, there has to be meaning behind it. 

Looking forward, as you look at the market evolving over the next decade, what are you optimistic about, and are there some concerns that continue to linger? 

One of the things that always is a concern in the space that we’re in, which is cellular IoT, is that we are piggy-backing frankly on the systems that are in place for cellular communications for smartphone users. When you look at it from that perspective, because it does help us with the cost of deployment, and it does help us with the ability to deploy data in markets and coverage, which you wouldn’t otherwise get is that technology evolves, and you’ve seen transitions and sunsets in these cellular technologies that will continue to hamper the growth of the future. So, I am particularly both concerned about the next sunset if you will, we’ve had a 2G sunset in the US, a 3G that is imminent, well when will the 4G sunset happen, because we’ve all heard the hype about 5G. I hope it’s 10, 15, or more years away, and ideally never, but that’s a tough requirement. 

The real key point is, can we make sure that the usage of that available technology matches the lifecycle of the products that need to use that technology? Can we find alternative data transports, particularly in areas where the cellular IoT systems cannot possibly go, like in the middle of the ocean, or deepest darkest Africa, where they may not have cellular because there are no people living there etc., can we provide services? 

So, I’m actually looking forward to some of these next generation, lower-orbit IoT satellite systems that are coming up with the technology using nanosats, it you’re familiar with that, at the cost; because the cost of operating and deploying a satellite system is so high that its cost-prohibitive, unless you can made the cost of the nanosats low enough that you can put up tons of them, and then create a lower-orbit data transmission capability where you can reach everywhere, and that’s going to be phenomenal. I think in the next 2 to 10 years, some people are more bullish about it than I am, but if it happens it’s going to revolutionize the transport of IoT data in certain markets, and certain locations. 

That’s exciting. What keeps you up at night? I think security is certainly on everybody’s mind, I’d love to get your sense of what some of your concerns are. 

You hit the nail on the head, security is the one big thing that keeps me up awake at night. We have seen financial security problems, we have seen denial of service problems occur because of breaches, we have seen the impact that security can have outside IoT with identity theft etc. To me, security for IoT is the one paramount problem that can hamper us, that can hinder us from growth because people will react negatively to security breaches that cause harm. Now, harm can be measured in different ways, it could be financial harm, that’s one thing, but if it turns out to be human harm that’s where I start getting very worried. If we do not plan for secure solutions in the IoT space, and we have a breach that results in human harm, that will be a major set-back for the industry. So, I’m on a soapbox when I talk about that, and I preach about it with just about everybody I can. 

I feel like I coined this phrase, ‘Security by design’, about 8 or 9 years ago, but it has taken off in the industry which I think is fantastic. We need to think about security as a part of the development process, as a part of the deployment process, as a part of the management process, not as an after-thought. Far too many people think of security as something we can tack on at the end, or we can check for at the end, and we can make changes at the end, it is just not practical or realistic to expect that, from two perspectives; 1) You will miss something, 2) Even if you have a very good solution today, the security problems evolve, leakages happen, new state actors can come in and cause problems, we have to plan for evolution of our solution down the road, and we have to plan for being able to modify the devices as we need to down the road. So, that’s what keeps me awake at night, is if a security breach results in human harm, then I would be very concerned. 

Yes, no doubt. You mentioned low orbit satellites, which is interesting, but are there any interesting technologies or start-ups that you’re keeping your eye on? 

I must admit that I have not done that extensively, we like to help start-ups get going. I have personally taken on a couple of small companies, not necessarily financially, but helping them understand how IoT can help them. One example was that blood sugar monitor that I talked about, simply because I feel there’s some benefit to those kinds of applications, particularly in the healthcare industry, but I haven’t really spent any time looking at that. I think companies that are at the forefront of deploying security solutions and looking at ways of managing data, to convert that data into actionable information is probably a good place to be, and to understand what could happen in the future. I don’t have any start-ups in mind though. 

The last thing I always like to ask our podcast guests, is a recommendation of a book or resource that you would share for our listeners. 

That’s a very good question, let me phrase it with a little bit of context for my answer. I am a science fiction buff, I have a large number of science fiction books in my personal portfolio, well over 6,000 science fiction books. It’s a little insane but that’s what I do! I tend to like an author and then buy every one of the books he or she has printed/published, read through them and try to interact with the author. My favorite author in the science fiction world of all time is Jack Chalker, unfortunately he is no longer with us, but his books make you think. If you can get hold of his books when you can find them, because he was somewhat of an iconoclastic individual who didn’t always interact well with his publishers, but you can find his used books everywhere. Jack Chalker – get hold of his books, in particular, there’s a four-volume book series that he called ‘The Lord of the Diamonds’ series. I would highly recommend that. 

My current author that I’m excited about, and I’ve been reading everything he publishes, is David Webber, his books tend to be space operas, but I enjoy space operas, so why not? His premier series of books about a captain of a fleet, first of a spaceship and then eventually of a fleet, the Honour Series, if you will. That would be something I would highly recommend. 

So, those two authors and those two series of books. 

Those are great recommendations. I love the answers we get when I ask that question.  

It’s been insightful information, and enjoyable speaking with you Z, and just to recap here; this is Ed Maguire, Insights Partner of Momenta, with another episode of our edge podcast. With us we’ve been speaking with Z Hosain, CTO and founder of Aeris Communications. We will be posting show notes. Z, thank you so much for joining us. 

Thank you, Ed., I really enjoyed it, I love talking about IoT, and I love talking about science fiction, so both of those topics came up! 

Maybe we’ll have to do a follow-up with a bit of a deeper dive. With a collection like yours, I think that’s the biggest collection of anyone I’ve heard of, I bet you can provide an enormous amount of insight there. So, thanks very much. 

Thank you.