On this week’s episode of The Data Stack Show, Eric and Kostas have a conversation with Rob Rastovich, the CTO of ThingLogix, a Bay Area company that provides IoT services and solutions. Additionally, Rob lives on and operates a 200-acre cattle ranch in Oregon that utilizes the IoT.
Highlights from this week’s conversation include:
The Data Stack Show is a weekly podcast powered by RudderStack, the CDP for developers. Each week we’ll talk to data engineers, analysts, and data scientists about their experience around building and maintaining data infrastructure, delivering data and data products, and driving better outcomes across their businesses with data.
RudderStack helps businesses make the most out of their customer data while ensuring data privacy and security. To learn more about RudderStack visit rudderstack.com.
Eric Dodds 0:06
Welcome to The Data Stack Show. Each week we explore the world of data by talking to the people shaping its future. You’ll learn about new data technology and trends and how data teams and processes are run at top companies. The Data Stack Show is brought to you by RudderStack, the CDP for developers. You can learn more at RudderStack.com.
Welcome back to The Data Stack Show. Today we’re going to talk with Rob Rastovich. And he is a fascinating guy, he built an IoT company that he’s sold to Amazon. And actually we’ll find out whether or not some of the stuff that he built is still running on Amazon. But he’s also a cattle rancher. So just a really interesting, diverse guy. Kostas, I think my burning question is, I think about IoT, and I can’t help but think about going to, like a Best Buy or something and looking at a smart refrigerator or a smart washing machine. And I just wonder how widespread is the implementation of that? I know there’s a lot of sort of IoT stuff, maybe in like warehouses, Amazon warehouses where they’re sort of doing inventory, and there’s lots of robots, but I wonder how widespread it is. So that’s what I’m going to ask. How about you?
Kostas Pardalis 1:23
I don’t know, the first thing that comes to my mind when I’m thinking about IoT is like, when robots are going to revolt and you know …
Eric Dodds 1:34
Then it’ll just make the world a cleaner place.
Kostas Pardalis 1:38
Yeah, I think that would probably be a good idea here in San Francisco, to be honest. But anyway, yeah, I mean, I think it’s probably the first time that we are going to have a discussion about IoT. I’d love to get to learn the space more, like see how the technology looks like, what are the use cases, what are the technologies out there? How mature is it? Because there’s also a lot of fluff around that stuff. Because there’s also a lot of like b2c products. So yeah, I think we’re going to have a very interesting conversation. And hopefully, we’re also going to learn something about how you can combine IoT with cattle. So let’s see what he has to say about that.
Eric Dodds 2:22
Let’s do it. Rob, welcome to the Data Stack Show, we have so many things to dive into. But thank you for giving us a little bit of time.
Rob Rastovich 2:31
Thanks, Eric. I appreciate it.
Eric Dodds 2:33
Okay, well give us your background. I mean, you have a fascinating background. And I’m not gonna spoil it for the listeners, but give us the brief background on how you got into data. And then what led you to working with ThingLogix today?
Rob Rastovich 2:47
So I started out actually in marketing. My job was at an advertising agency in Southern California. And that was my background. And when the internet first came around, we started working, we found out that okay, well, there’s another, this is another mechanism for us to sell ads back in the early 90s. And so I started playing around with the internet and then started putting up ecommerce pages. And once I started that I kind of always say that was the point where I got hooked on technology, that was the crack cocaine when you first see you put up a page and you click on something and oh my gosh, something else happens. And you put something in a basket and you make a transaction and the sale number goes up. Back in those days, we had the counters on the pages when the biggest thing was how many pages people were going to your page. So that’s when I really got the bug.
Started doing that for a number of years and then did a conference in San Francisco and saw Marc Benioff and Parker Harris, right click and deploy to a cloud. And I was hooked on the whole Salesforce ecosystem, being able to not having to worry about servers and JDKs and SDKs, and application servers and all those things. And so I drank the Kool-Aid, drank the cloud Kool-Aid back then, and started doing Salesforce consulting. I did that for, oh, probably eight or 10 years. And then some of the guys that were working with us got the bug we said, well, this whole machine to machine thing. We think that’s going to be the next big thing. And so we spun off, and we started a company in Denver called Telemetry. And our objective was, the belief was that, okay, the next big the next big .com, boom, if you will, is going to be IoT. We didn’t call it IoT back then, we called it machine to machine but basically, the idea of a device talking to another device, or talking to you and you talking to your device, the use case back then was being able to turn your, you know, thermostat down from your phone and those kinds of things. So, our goal was to build a message broker that could handle the amount of data that we believed was going to be coming through IoT, and that product that we ended up trying to build had to be something that didn’t exist back then. The idea was, when we were doing servers and we were doing applications back then it was all the size of the server, how many hits you could take, how big the database was. Well, this had to be a technology that could handle a million concurrent connections constantly and keep growing and growing in scale. So that was our objective. Well, and our goal was to, because we had come from the Salesforce ecosystem, was to sell it back to Salesforce. And I remember we were testing it out, we had that customer that said, okay, prove it. You think you can really do a million concurrent connections for a week solid? And we said, yeah, we can do it? Well, when you’re talking a million concurrent connections, that’s a million connections every millisecond, right, not like a million hits a day, it’s a million connections, every millisecond for a week. And that takes a big infrastructure to even do that and to test that, sure. We spun up a bunch of these EC2 instances on Amazon, and we literally got a call from Amazon going, what are you doing? Please stop that. And we explained what we were doing and the test was successful. And it was very quick after that, it was three or four months after that. And we sold the company to Amazon, and what is today the microservice, AWS IoT was the technology we were building back then.
Eric Dodds 6:21
Rob Rastovich 6:22
And so I also live on a cattle ranch. I have 200 acres and 400 head of cows; we can talk about that later, too.
Eric Dodds 6:31
We definitely will.
Rob Rastovich 6:34
At the time, I wasn’t … you had to move to Seattle, because during that acquisition, you had to move to Seattle. Sure. And there was no place to put 400 cows in Seattle. And so myself and one of the other partners and one of the other founders of Telemetry, we decided to spin up ThingLogix, because Amazon was more interested in the tech than the customers that we had. So we really started developing professional services around IoT and trying to create the pattern by which to deliver an IoT solution. Because of our Salesforce kind of DNA that we had, Salesforce really did for CRM, what we’ve done for IoT, and again, I imagine Benioff and Parker Harris sitting around one day and going, I got an idea. Everybody out there is creating a database, and they’re calling their table account and the contact, they’re making a join, and they’re doing all the same. Why don’t we just give them all that, and they can log in, and they’re all going to need the same things. And then we’ll give them the building blocks to expand on that? Well, we took the page right at that and said, that’s the same thing with IoT; you’re going to need all these things, you’re going to need context, you need security, you’re gonna need certificates, you’re going to need an API, you’re gonna need all this stuff to interact with your IoT devices. So why don’t we just build that? Give that to them? And that’s what ThingLogix now became. We ended up building a product called Foundry, which we have since open sourced, and it’s free to everybody to install in their AWS instance, and start configuring their way to an IoT solution.
Eric Dodds 8:03
Wow. So question for you. It really struck me that you started out by assuming mass scale in terms of required volume for IoT. And that’s so interesting to me, because a lot of times you’ll sort of when you’re thinking about a new startup or a new technology, you think, Okay, let’s try to prove this out on a small scale. But it seems like you started with the assumption of mass scale. I’d love to just know more about the thinking behind that. And then also, did you estimate too low or too high?
Rob Rastovich 8:44
Well, so the first problem you have to solve when you’re talking about the IoT world is custom. And I say there’s a lifecycle to an IoT project, right? And the lifecycle goes, like this customer shows up and they say, Look, I have this thing. Can you connect it to the internet? So I want a connected thing. So we show you put some little firmware on there and make a little MKTG client. And now you’re publishing data, and you show a little webpage and it says, look, here’s your data. And they go, that is so cool. Here’s my data. I can see it going up there, then the next thing they asked for, is, okay, can you put it on a graph? Sure. So you make a little graph and you watch the other graph go dynamically up and go down, and they go, that’s amazing. That’s awesome. We want to know well, can you make it send me a text message like when temperature gets greater than 100? I want to get an alert. Sure, I can make it send you a text message. Then they say, can you connect 100,000 of these? Right? Because I don’t have a business with one. And then they have what I call the “oh shit moment” Hope I can say that. Where they say, oh my god, how are we going to manage all these? Then turn off the text message because I’m tired of getting it, don’t send me alerts anymore, and get rid of the graph. And now we need to manage this at scale. So we knew that going in if you’re gonna have a business on this, and if you’re gonna have truly a connected product, whether it was a refrigerator, a dryer, or a temperature sensor, that the business model had to be infinitely scalable. And so we set out to prove that, yes, you could connect, you can connect an infinite amount of devices, and have that infrastructure automatically expand and contract for you. So we did start out with that idea. Did we succeed? Yes, we did. Now since that time, Amazon has taken it, obviously to a whole new level. The stack that we sold actually still runs the core of it. But they’ve added their whole ecosystem around and as ThignLogix, because we took, and we started adding their whole ecosystem around the solution as well.
Eric Dodds 11:02
So yeah, and so would you say, I’m also interested and Kostas, I’ll hand the mic to you after this, because I know you have a ton of questions as well, when you think about IoT, when I think about IoT, it’s a little bit novel, right? You think about okay, smart home, you hear a lot of things about smart homes, maybe people have a Nest or smart washer, dryer or whatever. But how pervasive is it from your perspective? Is it one of those things where it’s way more pervasive than we think it is? And sort of all these devices are sending data? Or are we still in early innings as far as mass adoption, and maybe that varies between commercial and consumer?
Rob Rastovich 11:47
Well, it’s not as pervasive as I want it to be. I’ll tell you that. I say one of the things about doing a startup is as I say, we have created a solution to a problem that most people don’t even know they have yet. So you can get out to the bleeding edge, but you kind of have to wait for corporate America to get caught up. And I think we are still at the very beginning stages. And quite frankly, the pandemic has actually increased, you know, people’s awareness of IoT, when the pandemic hit, we actually came out with and we built a we called it Work Watch, an application on top of our platform, where it took health surveys, it had a kiosk, it automatically took your temperature and did all that real time interaction kind of stuff. And we were able to spin that up and get going like three or four weeks. But those types of applications now became people’s awareness. A big one for us, during the pandemic became people counting. We had a very simple camera, you put it at the entrance of the store, and you have a little sign on the side, and it counts the number of people going in and the number of people going out so that you have occupancy and capacity compliance. During the heart of it, we had a customer come to us, they were trying to get some people back to work. And their biggest problem was they had to make an appointment for the restroom. Right? Because you couldn’t have more than one person in the restroom at a time. So they had to come up with, you know, a system whereby kind of like the light that you have on in an airplane to tell you when someone’s in and then when you can go and when not. And so all those types of little, little applications, I think, have greatly increased the awareness of people’s need for that. These things are gonna come around. So it is becoming more and more pervasive. But it truly is … I still think we’re at the beginning of it. And one of the things that I’m very passionate about is how businesses can take advantage of these. It’s a subscription based economy that we’re going to where you’re now … the example I always give is we had a pool supply company come to us and they say, well, we want you to update our website so people can do online scheduling, and they can order their products online. And they can do a service call line. It’s the wrong business model. What you need is a connected pump. You put the connected pump in the pool, create a subscription model, the pump now tells you when the chemicals are low. So you don’t have to go order, the chemicals just show up. You don’t have to have that emergency call on Friday night that says I got a pool party coming on Saturday, can you come out and fix it? Because you can do predictive maintenance on pools like that. So it’s a different way of thinking about your business. And that shift in the business I think is what takes time, but it’s coming.
Eric Dodds 14:34
Would you also say … sorry, Kostas, I lied, one more question … we still have plenty of time … Would you also say that, let’s take the pool pump example, and I was thinking about the refrigerator, but you mentioned a couple of other devices. Is there also a time lag with the replacement cycle of those right? I mean, you don’t buy a new refrigerator every year. And so even though there’s a lot of availability of the technology on the market, to actually replace physical devices, especially ones where I just think about like, wow, I couldn’t even tell you how old my washing machine is. Would it be nice to have a smart washing machine so I can sort of manage that on my phone? That’d be great. But I also am not going to replace it just to get that right now, because it still works okay.
Rob Rastovich 15:22
Yeah. And you have that cycle along, like you have the early adopters who are going to go buy anything that’s new in tech. If you had told me 30 years ago, 40 years ago, that, hey, one day, you’re gonna buy a brand new phone every two years, and it’s gonna cost you over $1,000 every time you want to buy a new one. I’d go, you’re crazy. It’s $3.99. It’s on the side of the wall here. Why would I want to buy a new phone? It works fine. So that technology cycle, I think, is getting shorter and shorter. And if you had told me 20 years ago that I would turn off my cable, and I’d only buy stuff off the internet, I would not believe that. But yeah, there is that lag. But I think it’s getting, you know, shorter and shorter. Because the market that’s growing up now is very used to technology in that cycle. Us old guys, we weren’t used to that cycle. So there, it wouldn’t be out of the question. And they’ve got disposable income too. It wouldn’t be out of the question for them to be walking through Home Depot or Lowe’s. I saw this the other day, and I have to have one too, by the way, a refrigerator that has a coffee maker built into it. But that is the coolest thing ever. And then I go home and I’ve got to get a new refrigerator because I got to have a Keurig inside of my refrigerator. Yeah, there is that, but it is getting shorter. Yeah.
Kostas Pardalis 16:57
This is great. Rob, I have a question about something that you mentioned earlier. And mainly because we take many things for granted like today, everyone knows about Salesforce. Everyone knows about the cloud. But what was the feeling that first time that you saw the founders of Salesforce demoing a cloud solution in front of you? And not only yours, but also the people around you? Can you share a little bit more about these? Because I think this is going to be super interesting.
Rob Rastovich 17:28
Well, yeah, technology is one of the things that excites me. Technologies and cows, right? Those are my two hot points, right. And I had spent most of my career at Harbor Freight provisioning servers. And it was always a fight. And then we had a DBA. And every time we needed to table we had to do that, how you did indexing. And there was just so much that went into that infrastructure. And when I first put up my very first website, I had to provision the T1 line, I had to figure out what a DNS was, I had to figure out how to do routing tables, you had all of that, which kind of now is abstracted to anybody and you just upload your website and there it is, you take it for granted. So when I saw that I thought, oh my god, this is where things are going because it made so much sense. And I had created those web type of applications in the enterprise. We were doing those, but not on the scale that Salesforce was doing. And so I remember going, I want to be a part of this. So there was a company Appirio, it was one of the very first little boutique consulting agencies that started up to do exclusively Salesforce consulting. So I was like an employee, and I finally said, okay, I’m going, I actually went back and I said, I told the tool company that didn’t want to put up a website, because they never thought anybody was gonna buy tools or put their credit card online. I said, hey, let’s move all of our data to the cloud. And they looked at me like you now you are completely off your nut. So I wanted to be a part of it. I ended up getting recruited by these guys at Appirio and went into that thing. And at that time, we had to, I remember the sales cycle, when you’re going through and you’re trying to sell a Salesforce implementation, the sales cycle wasn’t about, is Salesforce a good platform? Or is it a good idea? It was, well, what is the cloud? Where is it? Where is my data going to be? And why should I do that? And security and all those same questions that we answered 20 years earlier, in the .com boom, we were now asking and answering for this, but it was a very exciting time because in a small startup like that, everybody else had drunk the Kool-Aid too. It’s like, this is game changing and the whole notion of IT becoming a utility, like your electricity or like your water or like the gas instead of the server closet. That, to me, was also game changing. And there’s a book, I forget the name, The Great Transformation, where they talked about how when electricity became utilitized, that really changed the way business could work, where you could just plug in and you could use only the electricity that you needed to use, you didn’t have to have a generator outside your factory to create it. That changed the way business was. So when IT becomes a service, and IT becomes more of a utility, where you just pay for what you use, I thought that was a game changer. And, you know, Appirio ended up doing very well then we left and started Telemetry before they were sold. But Appirio grew to, I don’t know, 2,000 or 3,000 people. Was acquired by Wipro. It’s still going strong today, but it’s part of the big SI now.
Kostas Pardalis 20:54
Nice. And how did you go from CRMs to IoT? How did this happen?
Rob Rastovich 21:03
One day, I had been doing it for almost 10 years, I think. And there’s only so many times you can say, well, this is an account and accounts have contacts. And then here’s a lead. When a lead comes in, it converts an account into contact and creates an opportunity and then the opportunity. And then they go in and every customer, but I’m really different. My world is different. What I need is an opportunity that has a different value in the stage field. Okay, so we did that. And I ran a really big implementation for NetApp. I spent three years implementing their CRM. And at that point, I was like, I want to do something new. And quite honestly, I don’t think I’ve ever told this story. You guys pull out all these little stories from me that I haven’t told, so I’m actually out in the field. One day, it’s Saturday. Yeah, with the cows. And I’m irrigating, I’m changing the irrigation. And I remember thinking to myself, like, well, so the .com came around, I won and lost fortunes. And then the cloud kind of came around, we won and lost fortunes, because I love the technology. I’m a much better technician than I am a businessman, that’s for sure. But I’m standing out in the field and saying, okay, you’re so damn smart, Rob, you think that you knew that the internet was going to be ecommerce was going to be anything, you knew the cloud. So what’s the next big thing? And I remember thinking it’s got to be IoT. That ability to connect things and have data move back and forth. And have that messaging and things take care of that’s gonna be it? Well, I think I was right. But I was way too early. We were way too early there. But that’s really what it was, at some point you’re out. And people always ask me, Well, how can you be a farmer and a technician at the same time and I say everybody should do it? Because it’s those moments, right? We’ve all experienced those times, like, I can’t solve this problem, how do I do this? You get away from it, you go to sleep, you get on a plane, and all of a sudden, the answer comes to you. Well, every day, I do that same process, and working on the computer and either solving a problem or trying to write a piece of code, I get stuck, I go out and I change the irrigation or move the cows here and there. And you get out and you kind of do something with your hands. And it kind of frees your mind up like that. Those things come to you. And then I think quite the opposite is true. If I had to do that all day long, I’d go crazy. I love that grounding that gives me the freedom, I think to explain, explore and be more creative on the tech side.
Kostas Pardalis 23:43
That’s great. So IoT, how is it different in terms of technology and infrastructure? Like why is it different from anything else that we have built like in technology? What are the differences there?
Rob Rastovich 23:56
So that’s a great question. And one of my other passions that I tried desperately to explain to people I think, for the last 15-20 years, we have lived in what I call a request response world, right? Maybe 30 years. A request response world, meaning, and think of a simple web page. You click on a button on a website, you click on a link, it sends a request to a server, that server sends you a response. And now you see a new page, and you rinse and repeat that and you go through all these different processes. Our businesses have actually mirrored that technology. And there’s a great book called Man as a Measure that always talks about how we as humans, actually mirror our technology. There’s this underlying influence that technology provides and we model our world and it goes back to the clock and how the mechanistic period when the clock started became very precise mechanisms. But in business, we do the same thing, we request from our customers, we send them emails, we send them advertisements, we request that they ask us for information, and whole industries are designed around this request, sending out information and asking for the request. And then we get a response from our customer. And then that customer goes into our systems. And it goes through a journey of marketing, these marketing automation systems called journey and you take them through the journey, and ultimately, they buy your product and, and then you give them that kind of back and forth. Where I think IoT is changing it fundamentally, is changing this idea that now instead of that request response, it is the new UIs, the Alexas, the Google Homes, the SMS messaging, the new UIs of tomorrow, are we going to require an immediate action on an event, it’s going to be, and it is today, Alexa, send me some more dog food. Well, that’s not a request response kind of thing. It is an event. I spoke, and something needs to happen and a whole bunch of things need to happen down the road. And the data from my speech has to be translated into an order, and my address, and all of those events, I mean, go down there and I, we talked about the pool company that comes to us and says okay, well, we want to build a new website, and this new website we want it to be, we want it to have a calendar on it. So you can request maintenance and we want people to be able to order their products. And we want to, you know, schedule our service and our cleaning those kinds of things. It’s a very traditional business model, and we’ll drive people to the website and they will buy our product. Well, to switch that model around. OK, let’s not do that. Let’s take a connected pool pump, that sends us the pH, the alkaline, the chlorine. It sends us data about how the pool the pump is working. And do that as a subscription model. Now I have a customer that is very sticky, like I’m not gonna lose him, I got their chemicals arriving automatically, their pump is being on a maintenance schedule that is predictive. In other words, it’s not okay, if we have a pool party on Saturday, and it goes down on Friday, I don’t have to go out and do an emergency because I can see the predictive maintenance schedule that needs to go on there. So it changes that model in there about how we do businesses, and how businesses interact.
We did one called Cleaning as a Service. Hoover, actually Hoover vacuum cleaners, installed our technology and they connected their vacuum cleaner. Now we offered the Roomba, iRobot’s cleaning thing, but Hoover went one one step further and said, what if we did Cleaning as a Service? What if we said okay, we know that 99% of the reason that people return our vacuum cleaners is because their filter is dirty, and they don’t know how to clean it. What if we just sent them a new filter when it got dirty? And what if those parts and pieces came to you instead of would you use that vacuum cleaner? More? Yeah, and you would probably do a tech refresh more often than you would if you just had it and you would be more loyal to the brand. Because if you think your vacuum cleaner is not working because it just doesn’t work. When really it’s just you didn’t change a filter, you’re going to throw it away. Instead of getting Hoover you’re going to go get a Decker or whatever they’re called. But if you know that, Oh, I get my parts and pieces and it gets fixed. And when the new one comes out, I get a new one. I think it changes how you do business and how the stickiness is of a customer to a company.
Kostas Pardalis 28:29
Yeah, that’s super interesting. And okay, from what I understand, like the way that you describe it, when we are talking about IoT, we are talking about the connectivity of the physical world with, let’s say, the digital world. Yeah. And there’s a lot of telemetry happening, right? Like you have all these sensors, like all these devices, the remote devices that they live, like on whatever, like equipment we have out there, like the pump, for example. So there’s a lot of data there. Right? What does like in terms of scalability in terms of technical requirements, what do we need in order to realize like this vision that you’re describing right now?
Rob Rastovich 29:11
That’s a great question too. And the first thing you have to do is kill the SQL Server. Right? Because tables and rows don’t work anymore, right? I’ve seen so many IoT implementations where they’ve tried to do it, okay, we’re gonna we’re gonna spin up well, okay, we can’t use SQL Server. We can’t use MySQL, we’re gonna have to spin up a Redshift or something that really grows. But IoT data is changing. Version one of the firmware probably sends you temperature and maybe battery life, and version two has latitude and longitude. Version three, has humidity and temperature and battery life and latitude and longitude. And so you can’t be living in a structured data world, trying to put an under unstructured data world into a structured data world. So that was the first thing and unstructured data is very uncomfortable for IT people. It’s like, okay, they understand indexes, they understand tables and rows. And I even remember as a traditional Java programmer and traditional database developer. I remember going, no SQL? That makes no sense. How can you have unstructured data? That didn’t make any sense. I go, Oh, you mean, it’s a Word document? No, it’s not a Word document? And oh, you mean, it’s a PDF? No, it’s actual data, but you don’t know what the columns and rows are. And that’s, that’s number one. So now that you have that, you need a place to say, Okay, how do I put logic? How do I, where do I put my code? I’ve got code that I want to write, I want to create a solution. And where do I drop that code? So that when the temperature value comes up, and the temperature is 100, and the temperature sensor is in Arizona, the logic does something different than if the temperature value is 100 and the temperature sensor is in Alaska. Right? Where do you put that logic? And how do I, you know, put context around that like, so? Where are my devices? Who owns it? Who do I bill? And is there a rogue device? And how do I protect it and all those kinds of things. So we put in, everything’s event driven, everything turns off an event. So a message comes in, something happens. And there is a series of events that you have to have an infrastructure that is pretty robust to handle this. This is not something that’s going to go in the server closet. So it has to be cloud computing, it has to be AWS, it has to be Azure, it has to be Google, it can’t be anything other than that. And that, to me, is also I think one of the exciting things where technology has come. And I remember, without the advent of the Internet, this sounds almost naive, right? And with the advent of ecommerce, there would be no Amazon, right? Without this cloud computing, without the evolution of cloud computing going to and I mean, not just Salesforce cloud computing, where we had CRM in there, but I’m talking enterprise data 100% in the cloud, with that kind of compute power. Without that kind of compute power, we couldn’t have these types of business models, we couldn’t have these types of devices and data coming up.
Kostas Pardalis 32:27
What I understand is that the way that I visualize it is we have like, all these sensors, the sensors are sending data to the clouds, and the clouds were out in some logic, right. And we do something like for example, we make a request on our CRM that this customer, his vacuum cleaner needs a new filter, and someone is going to fulfill the order. Do you see this being also bi-directional, like in terms of like, the clouds and the device on the other side, start interacting and not just being passive, this kind of like the pump, for example, like opening and closing the pump or this kind of thing? Do you see this happening?
Rob Rastovich 33:13
Absolutely. Yeah. And we’re always talking about the lifecycle of an IoT project, right? When somebody says, Alright, can you connect this up to my, to the cloud? I got a water sensor. So yeah, we connected up, you see the data coming in? Oh, isn’t that cool? Isn’t that interesting? Can you put it on a graph? Yeah, you can put it on the graph. Can you make it send a text message? Yeah, you can make it send a text message. But really, that’s not a smart device. That’s just a dumb device, sending data up to the cloud, right? What a smart device does is say, oh, if there is water, tell the pump to turn off, not just send you a text message so that you can go turn the pump off, go and turn the pump off. And when the pump turns off, it should notify the utility company that it’s got a leak or something like open up a service ticket in your utility company. So yes, these devices can’t be just dumb data collectors, where they’re sending data up. They got to be both ways. And really, the interesting part of it becomes that I think businesses have to be that way too. So one of our clients is the United Way. And the United Way, you’re like, okay, well, what devices does the United Way use? None. They don’t use any devices. But we use the same paradigm to do agencies as a device, if you will. So Cisco actually sponsored an initiative to get 1000 people out of poverty, was the name of the program. And they contacted us and said, hey, what we want to do is be able to coordinate care across multiple agencies in the San Fernando Valley, sorry, Silicon Valley in San Francisco. And they want to be able to coordinate their care but they’re all disconnected systems. And we have a person that isn’t really going to fill out a form or the whole data input is a problem, someone’s going to do it on their behalf. So we enabled each of their devices to basically create a database as a device. So we put a small piece of code, very, very unobtrusive piece of code on these databases that would chirp and say, Okay, well, this record just got updated so it sends a message. The message comes into Foundry and we can coordinate that against its other care, get referrals to other networks. And so their entire networking system became an event driven thing. So once a person says I’m leaving the homeless shelter to go to the job training shelter, you know that immediately. And you can, you can track them, you can do the end. And in California, it worked out really nice, because the state also provided cell phones. And so as state provided cell phones, you could track and coordinate their care and whatnot. So it is a situation where it’s got to go both ways. It’s not messaging or just listening to it. But it’s got to be interactive. That’s why I think it’s the internet of everything, and not necessarily just things.
Kostas Pardalis 36:23
Yeah, there’s a lot of discussion also about like, like the impact that 5G is going to have on stuff like edge computing. So the compute actually, is going to move away a little bit from the clouds and go closer to these devices. Because, I mean, we end up connecting everything. That’s a lot of bandwidth, right? Back and forth. So what do you see in terms of the maturity of the technology right now, where do we stand? I mean, the vision is amazing and great. Like, we have all these connected devices, we can become as lazy as we want, everything will happen automatically. So we are cool. But where are we today?
Rob Rastovich 37:05
Well, it’s interesting. You mentioned 5G, and a lot of people think that okay, 5G is gonna take us to the next level of that. But ironically enough, one of the biggest use cases for IoT is agriculture, right? I know that because I live on a ranch and plus most of our stuff, we have a lot of agricultural clients, and in agriculture, and in IoT, it’s not bandwidth. It’s not all vertical bandwidth, not how much you can, because the message is very small. But it’s distance. It’s like, can I get a temperature sensor out of 200 acres, and there’s not Wi-Fi around my 200 acres, but can I get a temperature sensor on the backfield to send me moisture data back and so technologies like LoRa, where it’s very small payloads, but very long distances? That to me, I see as defining technologies in terms of the IoT world, because it’s not the amount of bandwidth we need, we need the reach of it, and we need to be able to go everywhere. But where are we in terms of AI? Is this all kind of pie in the sky? Or is this actually happening? And there’s a couple things. So it actually happens. But there’s two things that slow it down. Number one we talked about, I think it’s enterprises learning the new student business model, corporate America doesn’t change quickly. Number two is fear. And we see this almost daily. And when we were developing this technology we had some people who were very interested in our technology that made us a little nervous that they wanted to invest in us. And we just were very nervous, because reality is we’re building big brother, right? There’s no question about that. And all things connected. And that is a scary thing when everything’s connected all the time. And one of the other discussions that I also love to have is ethics and morals, that I think we as technicians have a responsibility to when we have this technology to make sure that it is properly used.
So it will always be used for ill and good. I don’t doubt that. But I do think that that is coming. So that fear. And it’s the same fear we saw with people putting their credit card on the internet, like they would never do it. Why would we put our credit card on the internet? Well, eventually you will, eventually they get comfortable with it. Now, hopefully. And we see this all the time now. And it’s getting scarier. You do a search on Google. And all the ads pop up on Facebook, for what you did a search on Google for and all of this stuff. It’s too scary. We can do really connected kinds of things. But because you can do anything doesn’t mean you should do anything. And I think we’re kind of at that spot where I saw Facebook announced today they were gonna change their ad algorithm. Or there was something about that, that they said, okay, okay, enough. We’ll stop. We’ll stop doing this crazy kind of stuff. But it also is, there is a value to that too. And I go back to like, my, my beef customers, right? Like if I just texted everybody in Bend, Oregon, and said, hey, buy ground beef from Rob Rastovich. Well, most of them are going to get all kinds of expletives back about why you’re texting me, how did you get this number, whatnot. But the ones that I do engage with, ones that choose to have that kind of relationship, it creates a new kind of relationship between me and my customer. That is much stickier, much more personal. And even you look like telemedicine and the things that we can start doing in telemedicine and all those things. There’s a lot of value in this, but I think we’re just starting.
Kostas Pardalis 40:44
Yeah, the last question is about three words. One is security. The other is safety. And the third is privacy. You touched on privacy a little bit. But I think what is interesting with connecting the physical world is that we can have scenarios that are from funny to tragic, right? Like, I can imagine someone hacking my Roomba, for example, and my Roomba attacking me, right. I wouldn’t want this to happen. So what have we done towards that? Like, where do we stand about that? And what is needed so we can feel as safe as possible around IoT?
Rob Rastovich 41:23
Yeah, so I mean, that is always our number one concern, the question we get from our customers, like how can we make sure that this doesn’t fall into the wrong hands? And I think we’re doing a lot of those things. So the monitoring systems, we’re doing the certificate systems, being able to make sure that what is sending the data truly is it. We have all seen certificates on websites that when you put your credit card in, you want to make sure that it’s a secured site and whatnot. Well, each one of these devices actually has a certificate on it. And it has been, because early on, we had all of that on there as well. We had many stories, and you can hear these stories over and over again, about how IoT devices got hacked. And one of my favorite stories was when there was a rental car company that said, hey, we’ll make a mobile app that will unlock the car. And they thought, well, that’s kind of cool. You don’t have to get a key to unlock the car, you get in, you can drive off. Well, they did that and this elderly couple drove out into the desert of Nevada someplace and got out to look around. And there was no cell phone signal and they couldn’t unlock the car. They couldn’t get back in the car. Because there was no connectivity there. And then there’s stories about people turning lights on and off through other devices. So it is a main concern.
And so the audits that we go through, one of the things we do … one of our customers is Emirates MDB, we actually have an office in Dubai, and we do a lot of IoT in the Middle East, they’re, they’re very much willing to they seem to be more open to these IoT solutions, then we are in the States. But security is a huge concern. So going through the security audits and doing all those best practices. And today, even today, I think most people think and I think there’s a common misconception that the biggest danger is some hacker is going to figure out your password and get into you know, your system, that that is a danger. But we have such great protections against that, the real problem. in all this is the social engineering part of it. And there’s a great YouTube video and I don’t know if you’ve seen it, but the guy is a news reporter and he has a hacker is telling him, she’s a lady and she says, you give me your phone number, I’ll close your bank account, have your social security number and turn your phone off. And he goes, there’s no way you can do that. She says yeah, I can do that. And all I need is your phone number and your carrier. He told her her phone number and who the carrier was, she found an audio recording on the web of a crying baby. She gets on the phone with this dear, sweet, wonderful, well-intentioned customer service rep for the carrier. And she explained the crying babies going off and she goes in and she explains to this person how her husband’s gonna get mad and she’s got a baby and she’s got to get this and she doesn’t have a social security number. And she goes through this whole thing and at the end of the day, she has all those, she’s turned off his cell phone, has his bank account closed and no security number. Not through a process of hacking into bits and bytes. But through a process of social engineering and that’s really where the danger lies. And I think the misconception is that it’s all oh, we got hacked. Yes, that is a threat but it is nowhere near the threat of phishing links. Like here click on this link and all of a sudden you’re given you’re giving your password away to somebody.
Eric Dodds 44:57
Fascinating. Tons to think about. We’re close to time. But the question that I had to ask was that you started to talk about some use of IoT in the context of your ranch. Can you just tell us how you use IoT technology? If and how you use IoT technology as a cattle rancher?
Rob Rastovich 45:20
Yeah, you wouldn’t think it would be top of the list. Well, and I’m probably the exception in the cattle ranching industry in terms of IoT, right. So what I’ve started to do is, so my model is where I take in cows, and we feed them beer mash. So we pick up the spent grains from the local breweries and we feed that to the cows. So then we sell the meat back to the pub. So when you come to town, you have a burger and a beer, you’re eating a burger raised on the beer you’re drinking. And everybody needs to show up at Bend. But the process of handling cattle is when you handle cattle you want to keep a cow as least stressed as you possibly can. I mean, if you’ve ever been in a 24 by 24 pen with a 1,600 pound animal that is very stressed, you don’t want to do that. Right, that is a bad place to be. So each cow has an RFID tag in their ears. So we know which one they are. So what we’re just what we’re starting to do, and what we’re trying to develop, it’s not completed yet, but we’re working on it, is a series of, we built new corrals, and we have RFID readers so that as a cow passes through on its way to feed or to get drink water or something like that, we’ll read it and see what the number was. We could then punch in the inventory of cows we can see, okay, which one has been here the longest, which one weighs the most, which one is ready to go to market, because today, the one that goes to market is the slowest one, the one that you can catch and put in there, that may not be the real one that should be going to market right. So as the cow passes through these readers, we can open and close a series of gates, that just very casually moves a cow into an isolation pen and then they close back in and the rest of the cows go to where they normally go. So by having that type of system in there, you’re able to keep very less stress on the cow because they just move in and get their feed, the safety of the handler is improved because he doesn’t have to get in there with a 1,600 pound animal. And so that’s one of the use cases where you can now use IoT to move and change, you can change gates and move them and move cattle around. The other one we use grew industrial hemp a couple years ago, after it became legal in Oregon. And as a result, we could start using fertigation systems. So in other words, you put a sensor out. And because in hemp, or row crops, the row crop is like anything that you plant in a row, you call them row crops like celery or carrots or potatoes or cabbage or those kinds of things. And typically they’re individual plants, they’re usually covered and they have a drip irrigation system underneath there, we put a sensor in there for not only moisture, but for nutrients. And then the sensor will send data back to the pump, and it will increase or decrease, the water is going to go through no matter what. But say it needs more nitrogen or it needs more phosphorus or whatever it needs, it’ll increase or decrease those nutrients to the plant so it gets exactly what it needs when it needs. Very efficient. And water use, which is another passion of mine. I think that’s a whole podcast in itself, we could talk about water. But one of the other things we’re working with the USGS, the United States Geological Service, and they’re using our technology to measure the volume of water in rivers so that you can do flow analysis. And that flow analysis, and while the logical thing is there going to be a drought or is there not a drought. But more importantly, if whatever’s in the Columbia River, or in the Colorado River, probably right now, today is really important to the guy in Arizona, because that’s going to tell him what kind of crops he can plant two or three months from now and what that’s gonna look like. So being able to create that water, I think, is pretty important.
Eric Dodds 49:17
Fascinating. Rob, this is amazing. I can’t tell you what a treat it’s been to just hear you talk about all the amazing IoT technology and then how you’re actually using it on your own ranch. I mean, that is just incredible. Thanks for sharing your story. We’re out of time, but we’d love to have you back on the show sometime and hear about how implementing some of those IoT projects on the ranch goes.
Rob Rastovich 49:40
No, absolutely. Yeah. Thanks for having me, guys. I really appreciate it. It was really fun.
Eric Dodds 49:45
What a fascinating guy. I mean, I just love talking to someone who is trying to just deal with the normal things you deal with as a cattle rancher and growing crops but who also sold an IoT company to Amazon and works with so many interesting businesses. I think one of things that really stuck out to me was his point about social engineering being more dangerous than hacking in the sort of movie sense where we think there’s someone in a dark room hacking into a system with code. And that was just very thought provoking. And I really appreciated that perspective.
Kostas Pardalis 50:28
Yeah, absolutely. I think the way that he put it, it seems that humans are the weakest link in this situation. And probably he’s right. You cannot manipulate like a machine. I mean, a machine is zeroes and ones unless like, there is some kind of issue that we can exploit for that. Yeah, humans can be manipulated. That’s true. So social engineering is quite a thing.
Eric Dodds 51:07
Indeed. All right. Well, we will catch you next time on The Data Stack Show. Thanks for joining us.
We hope you enjoyed this episode of The Data Stack Show. Be sure to subscribe on your favorite podcast app to get notified about new episodes every week. We’d also love your feedback. You can email me, Eric Dodds, at Eric@datastackshow.com. The show is brought to you by RudderStack, the CDP for developers. Learn how to build a CDP on your data warehouse at rudderstack.com.