In

016: How Augmented Reality and IOT are Driving Business Innovation – Ron Dagdag
Unstoppable Talk Interviews

 
 
00:00 / 40:13
 
1X
 

Augmented Reality and IOT are interplaying together in interesting ways to drive innovation and change within businesses.  In this episode, I spoke with Dallas-based Microsoft MVP Ron Dagdag about his experience and interest in these fields, and the metaphors he uses during speaking engagements to illustrate concepts in these technologies.





Sam Schutte:                In today’s show, we have Ron Dagdag. He is a Microsoft MVP in Dallas, Texas. I met Ron while working together with a manufacturer in Texas. And today we’re going to talk about how developers are innovating using augmented reality in IOT. Ron, welcome to the show.

Ron Dagdag:                 Hello.

Sam Schutte:                So thanks for being on. Maybe a good place to start is tell us a little bit about what you do, how you got started and how you got into that field.

Ron Dagdag:                 I’ve always been fascinated with technology ever since I was a kid. Started programming in high school in my hometown, five hours North of Manila, Philippines.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And we don’t have internet back then in our hometown so I would rent computers per hour to do some programming and to explore GW basic back then. And then I got into computer science and then moved here to the US, back in 2000. Most of my career is in manufacturing, automating different processes from sales, to shop floor, to delivery.

Sam Schutte:                Got you. And you said you rent a computer by the hour. So you would go to a internet shop and go in there. Okay. Interesting.

Ron Dagdag:                 We don’t have internet back then, so it’s really just computers.

Sam Schutte:                Oh yes just computer. Okay. And did you get your degree over there or here?

Ron Dagdag:                 Yes, I got my degree in the Philippines.

Sam Schutte:                Okay, great. And so now you are a Microsoft MVP, so maybe for our listeners, explain a little bit about what that is and sort of the specific areas you’re interested in that you were awarded that for.

Ron Dagdag:                 Microsoft MVP is an award that’s given to developers or to technologist because they organized events or they shared their knowledge to the community and their technical expertise.

Sam Schutte:                Okay.

Ron Dagdag:                 Anything related to Microsoft technologies. And then this is my third year to receive this award.

Sam Schutte:                Okay.

Ron Dagdag:                 I usually speak at conferences, and write blogs, organize events here in the Dallas, Fort Worth area.

Sam Schutte:                Okay. And what are the specific areas you have that for?

Ron Dagdag:                 Related to Windows development. Windows development is not just a laptop or desktop anymore. Think about it. There’s augmented reality side of Windows, which is the whole lens and also the internet of things side of Windows, which is the IOT core and such.

Sam Schutte:                Got you. And then the hollow lens is the goggles that they have produced, correct?

Ron Dagdag:                 Yes, that’s correct.

Sam Schutte:                They just come out with a new version of that recently, didn’t they?

Ron Dagdag:                 Yes, they came out with a new version, a Hollow Lens 2 that has hand tracking which I’m excited about.

Sam Schutte:                Interesting. So maybe let’s talk a little bit about some of the background of augmented reality. Can you explain for folks what that is and what are kind of the different types of augmented reality or that spectrum?

Ron Dagdag:                 Augmented reality, virtual reality, when you think about it, there’s a virtuality continuum. If you can imagine on the left side of things is where reality is. And then on the right side of a diagram is virtual.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 So reality is where we live, right? We see it every day. We experience it every day. Virtual reality, on the other side of the spectrum, is everything is computer generated, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And you wear a headset and it’s all occluded and you can only see what is generated by the computer. So in between that there is mixed reality and that’s where the lies between the augmented reality and augmented virtuality is. So augmented reality is you can see the real world and you’re adding some experiences based from some computer graphics on top of what the reality is using a device. It’s kind of like the Hollow Lens. And also there’s the augmented virtuality is where you’re pulling in some information from the real world inside virtual reality.

Sam Schutte:                Okay. What’s a good textbook example of one of those experiences you might do in those environments?

Ron Dagdag:                 So, augmented reality, Pokemon Go would be a good example of that one, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Because you can see from the camera of your phone and be able to see the reality, at the same time have some graphics or some models that you can actively interact with.

Sam Schutte:                Got you.

Ron Dagdag:                 And then on the virtual reality side, Oculus Rift, Oculus Go would be a good experience because you wear this headset and this goggles and be able to have a different experience. It transports you to a different … It can take you anywhere.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 That’s how I see virtual reality. Augmented reality is something that can bring stuff to you, to your world, right? Virtual reality is something that can take you anywhere you want to. So if I say I want to be in a certain place, I want to experience a certain place like in a video game, then it can take you there. It feels like you’re there.

Sam Schutte:                Got you. That’s a good way to look at it. Yeah. It’s adding to you the real world, the existing world versus just taking you to an entirely different world.

Ron Dagdag:                 That’s correct. Yes.

Sam Schutte:                Interesting. Yeah. And I know folks are using augmented reality a lot now for doing things such as inspections, is a real common case. I saw a presentation not long ago of a oil refinery company using it to provide information sort of within view during inspections for instance.

Ron Dagdag:                 Mm-hmm (affirmative).

Sam Schutte:                So I think it’s something that’s taking off more and more in the industry. How did you first sort of get into these technologies and use that sort of passion you’ve got for it to win the MVP award?

Ron Dagdag:                 One of the things that I was interested in, back four years ago, back in 2015 when I first saw where they released the Hollow Lens, right? And there’s this experience between an IOT device and then someone’s wearing the Hollow Lens and there was a robot on top of an internet of things device. It inspired me to kind of see the convergence between augmented reality and also machine learning and internet of things. How it could change the way we interact with software and how it would impact our daily lives. It’s changing to where it’s not just keyboard and mouse anymore.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 You remember back few years ago, it started with touchscreens and now it’s voice, right?

Sam Schutte:                Yeah.

Ron Dagdag:                 Voice and smart home devices and even hand gestures, like Kinect. And then now the new devices that are coming up, the virtual reality devices, have hand tracking and also eye tracking.

Sam Schutte:                And so how are companies kind of implementing and using those from a practical standpoint to sort of realize value for business problems and for projects they’re working on?

Ron Dagdag:                 These days you see a lot of applications for touchscreens, right? iPads being deployed in the field, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Being able to use the sensors that is on those smart tablets. And then nowadays a lot of homes have started to have voice enabled systems, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Be able to control. And if you just want a simple command or simple question that you can ask. And eventually offices may need these. I don’t have to pull a laptop if I can ask Alexa or Google, what’s my sales going out today?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Those kinds of things. And in terms of the computer is moving from desktop or moving from offices towards the people that does not have access to computers or technology. Not technologically, but it’s harder for them to access computers, like people in the field, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 People that are drilling, or oil and gas companies, or people that are in manufacturing, or in the shop floor. It’s harder for them because they need both of their hands to be able to work on device or on something on their jobs.

Sam Schutte:                Yeah. It seems to me, like really the whole thing is about accessing the information you need in the right place at the right time. I mean, if you look at voice assistance, that’s for when you may not have your laptop booted up or want to pull your phone out and type and search. There’s a certain portion of information you can access via a voice question, right?

Ron Dagdag:                 Mm-hmm (affirmative).

Sam Schutte:                And not everything, but there are key things.

Ron Dagdag:                 Yes.

Sam Schutte:                And there are solutions I’ve been seeing come to market that integrate more with enterprise databases, ERP systems and whatnot that you can ask Alexa, “Hey how many widgets did we sell last week?” And it will tell you.

Ron Dagdag:                 Yes.

Sam Schutte:                Then there of course there’s other information that you need to see with your eyes while you’re looking at a circuit breaker or something.

Ron Dagdag:                 Right.

Sam Schutte:                You don’t want to get your laptop out and try to find the picture or pull the PDF up at that point.

Ron Dagdag:                 That’s correct.

Sam Schutte:                So, there’s another portion that you need to compare to the real world, right?

Ron Dagdag:                 Yes.

Sam Schutte:                So it’s all just about what is the best way to sort of consume that information and can we get closer to it.

Ron Dagdag:                 And I just also want to add cameras for safety zones, right? And that’s another extra sensor that can help someone like the person detection, or people detection, or something that would also listen to sound. One thing use case that I’ve heard or looked at is being able to detect if there’s something wrong with saw saw mill area.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And if there’s something wrong with that device and be able to detect that up front before an accident happens or something like that.

Sam Schutte:                Yeah. Well and of course the feeding all that sort of visual information into a machine learning algorithm that can not just react to the obvious problems but sort of the other data points that indicate a trend.

Ron Dagdag:                 Yeah. That’s why it’s really neat that the convergence between augmented reality and machine learning and internet of things, they’re all coming together to produce different types of products that actively can make a difference in terms of productivity, changing lives, safety, security,

Sam Schutte:                Mm-hmm (affirmative). And when you’re looking at projects like this or for projects you’ve been involved in what do you think the best practices are for starting the initial stages of a project like this? How do you execute those and where are some of the sort of sticking points where a project might struggle or fall flat?

Ron Dagdag:                 In every project that I’ve looked at, finding the flow first of how the interaction happens. Sometimes being able to figure out what the goal and what the value it can bring to the business.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 I like building proof of concept.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Sometimes that can be throwaway code, but it kind of gives you a reference to understand the whole problem. And that’s why I enjoy attending hackathons and even in just a few weekend you have something that might not necessarily ready for production, but it shows you the value and ways to explore that idea if it’s worth pursuing or not.

Sam Schutte:                Got you. What are some of the most interesting things you’ve built that a hackathon?

Ron Dagdag:                 Few things that I’m interested about is how you would … There’s one project that I’ve worked on where I can, inside virtual reality, kind of control a Rover. So just just thinking about how you would be able to remotely control the Rover inside virtual reality and then be able to bring cameras inside virtual reality or what the robot is seeing or what the rover is seeing is sent up to the inside virtual reality.

Sam Schutte:                Just sort of like a remote control car. [crosstalk 00:13:14].

Ron Dagdag:                 Kind of like a remote controlled car.

Sam Schutte:                Got you.

Ron Dagdag:                 Those kind of things.

Sam Schutte:                Interesting. What are some of the biggest ways that technology sort of all in this area has impacted companies you’ve worked with?

Ron Dagdag:                 Lately I’ve been working with voice assistance, AI smart voice assistance like Amazon Echo and also Google Home. And I’ve noticed that the rise of containers, serverless computing, machine learning, internet of things, and also computing at the Edge. And that’s what I’ve been … how it’s impacting how you developing these technologies now. So one of the examples I’m seeing at is because this AI assistance they’re actually running servers as computing at the end. It’s where the Cloud provider runs on the server and dynamically manages the allocation of resources.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And it’s the idea that you only pay what you use and it’s because … I don’t know if you’ve noticed lately computing is becoming a utility company. It’s just like a utility company where it’s not like you’re buying a bunch of servers and installing it to your … having IT in your team install it in house, but you’re actively renting computers.

Sam Schutte:                Yeah.

Ron Dagdag:                 And now it’s even more, is you’re renting just the time that you use it, that you run these functions.

Sam Schutte:                Or that you’re just renting this specific sort of functions.

Ron Dagdag:                 Yeah.

Sam Schutte:                And I think that’s the interesting thing about that. When you talk about serverless and containerization and everything else. For instance, if you look at Digital Ocean, they have a completely managed database platform. Just give me a database connection and that’s all I need. I don’t worry about the server, I don’t worry about anything. And those pieces of it managing a Linux server and administering the Linux server are the things that typically a lot of developers struggle the most with and also just don’t really want to have to do because you’re not creating anything when you’re dealing with that sort of administration. Right?

Ron Dagdag:                 Yes.

Sam Schutte:                And of course usually that means maybe you have to have somebody who’s an expert in that stuff to do it for larger companies, but it’s like if I can just point and click and the database just works, that is a huge time saver. I’ve been migrating a bunch of stuff to the Cloud recently, so it definitely hits home for me. What are some interesting applications of machine learning you’ve seen in terms of what kind of models you’ve seen being used and such?

Ron Dagdag:                 One of the things that I’ve been seeing lately is in terms of having pre-trained models that exist out there and using it and just adding it to your existing system. One of the things that I’ve been working on lately is being able to use, like for custom vision. I want something that I can customize pre-trained models out there to do computer vision stuff.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 So if I don’t have billions and billions of pictures to detect if this item, in this case, I’m trying to say if it’s hotdog or not hotdog, right?

Sam Schutte:                Right.

Ron Dagdag:                 So I have pictures of hotdog and pictures of not hotdog and how you classify those and being able to use a pre-trained model and customize it to train and build new models for me. And one of the great things that I see where the serverless computing and containerization is, is being able to just rent the training where do you do your training up in the Cloud.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And then being able to manage and run the models at the Edge, right? Being able to deploy it locally on the computer or in a raspberry PI, or an IOT scenario.

Sam Schutte:                Mm-hmm (affirmative). Exactly. And what about IOT? I mean, what kind of devices and use cases have you seen there that are interesting recently?

Ron Dagdag:                 In the IOT I’ve noticed that there’s more and more need for running your models, running your device or being able to deploy it locally. So managing the deployment side. I’m seeing more and more technologies that helps you deploy it easily in making it more secure because that’s one of the things that it important in IOT devices. Right? How do you safeguard connected devices and networks in the internet of things world, right?

Sam Schutte:                Yep.

Ron Dagdag:                 And one of the things that I have noticed is being able to easily update and deploy from the Cloud to the Edge device to the internet of things device.

Sam Schutte:                Yeah. Well I always feel like there is an issue with the sort of data gathering of IOT devices where often times these things just get set up and sort of just pumped into a table somewhere, but the actual table structure that it will populate is usually not great. And then you end up with millions of pieces of data in a sort of a really flat table.

Ron Dagdag:                 Mm-hmm (affirmative).

Sam Schutte:                And I think part of that is just the tools allow you to do stuff that … almost lets you be a little too powerful if you’re not a database designer. Right?

Ron Dagdag:                 That’s true. I mean you were talking about like a database in the Cloud a while ago. Right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And lately I’ve been seeing where I can deploy on a container, that database to my internet of things device as a container.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 That containerization of internet of things. And for me that was interesting how I can, as a developer, I can create a container right? For my application, I can create a container for my database. And when I deploy it, I can deploy it in the Cloud at the same time. I can find a way to deploy it now to an internet of things device as a container.

Sam Schutte:                Mm-hmm (affirmative). Interesting. So it’s actually running on the device.

Ron Dagdag:                 Yeah. Running on a device. So in terms of code-

Sam Schutte:                [crosstalk 00:19:49] have to upload somewhere.

Ron Dagdag:                 Yeah. What I’m shipping that runs with my computer would be able to run on one of the internet things device if the device is capable of running that payload or that load.

Sam Schutte:                Yeah. Yeah. And what is Microsoft doing in the space as far as that ability to run say C# code on some of these devices. What have they sort of put into place that helps with that?

Ron Dagdag:                 One of the things I’m seeing is the .NET core.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 .NET Core 3.0 is being released at the end of this month, actually sometime next week, but being able to be cross platform that it’s not just Windows devices anymore. That it can also run in Linux devices.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 So these devices like Raspberry Pi or some of the gateways, it can run these Linux devices.

Sam Schutte:                Yeah. Which would be pretty huge. Save a lot of time. What about some of the sort of trends in the economy that are affecting these technologies in terms of IO. What are some things that are happening that are affecting people’s decisions?

Ron Dagdag:                 Lots of the IT projects have noticed that they’re focusing more insecurity and how do you safeguard these devices that are connected to the internet? One wake up call for everyone is that [inaudible 00:21:35] Botnet. It is a malware that infects Smart IOT devices and turning them to a zombie or a bot that can be controlled remotely by someone else. And so and typically a botnet is often used for a DDOS or attack, or denial of service attack. So security has changed the way we looked at these devices.

Sam Schutte:                Yeah. And I’m not super familiar with that particular botnet, but I mean essentially the lesson is folks just left their devices wide open, I imagine.

Ron Dagdag:                 Yes.

Sam Schutte:                Because of this challenge, I mean I … Any project we’re working on, we see a challenge of sort of getting a proper network access or getting too much access that it’s wide open and people came in and took advantage of it.

Ron Dagdag:                 Yeah. Sometimes some port is open that was not secured.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And those things we need to look at and be able to make sure that security is important in it.

Sam Schutte:                Got you. So what are, outside of IOT and augmented reality, what are some other sweet spots and areas you’ve been sort of investigating in there you’re excited about?

Ron Dagdag:                 Well that AI Smart Assistance, I mean if you think about it, because of that technology I started to use … I’m really happy about serverless architecture.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 You would often hear the term function as a service. At the end of the day there’s … you hear the word serverless, there is a server somewhere. It’s hidden from developer. Essentially what you’re saying to your Cloud provider is that, “Hey, here’s my code. Run it somewhere in your Cloud. I don’t care where you run it, just run it for me.” And it would execute.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And so that’s one of the things that I’m really excited about. How it’s changing the way we do development.

Sam Schutte:                Got you. Okay. As far as projects you’ve been working on, what’s one of the most personally rewarding projects that you’ve worked on recently?

Ron Dagdag:                 Lately I’ve been speaking at conferences about machine learning and IOT and I really enjoy when that “Aha” moment or that spark on someone where their mind is blown away with all these different technologies that I’m talking about.

Sam Schutte:                Yeah.

Ron Dagdag:                 And the possibilities and what they would bring. Lately I’ve been talking about machine learning and the difference between a data scientist and a data engineer. You think about it in machine learning or artificial intelligence. When do you need the data scientists and when do you need a data engineer? Data scientist usually is your best baker in town. They create secret recipes for your cakes or your pastries. And they try different ingredients to make the best cake. Data engineers or typically software engineers are the ones that are expert in building a bakery. They know how to deliver fresh bread to the customer. Where do you source the ingredients? Which oven to use when you’re building the bakery? And where do you put the cash register in?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And I think that’s where the value of AI and machine learning projects would come from is once we started integrating it to our businesses.

Sam Schutte:                Yeah. Well, and it’s interesting because of course any AI model is only as good as the data you feed it that.

Ron Dagdag:                 That’s correct.

Sam Schutte:                So if you have a data scientist building an amazing model, but the information that’s receiving is not real world, it’s useless or at least weakened.

Ron Dagdag:                 Yeah. And then also based from the data that we’re collecting and there’s that feedback loop, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 So let’s say we started the bakery and we’re getting bad reviews from our customers, right? Or bad data from our customers. We can send that back to the data scientist or to to the baker and say, “Give me a better recipe.”

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And that would help them. It’s a feedback loop for them and say, “Hey, this ingredient didn’t work. This combination didn’t work. Let’s try a different one.”

Sam Schutte:                Okay. Yeah. And when you’re speaking on these topics, you mentioned you’re doing some speaking engagements around this, what’s sort of your approach to … or what do you think works well to speak to an audience? You’re talking about a group of people that may not know a lot about these things. And like you said, it’s kind of blowing their mind. Do you give examples? Is that part of your sort of approach giving real world examples like this or metaphors?

Ron Dagdag:                 Yeah. Metaphors, analogies. That’s pretty much how I learn about certain technology. It’s like how does it fit in the real world, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 I mean at the end of the day, whenever you’re building a software system, you’re actually you’re just not necessarily copying, but you’re actually extending how the business works, how the communication in a business, right? How they communicate and you’re actually just making it faster for them to communicate. That’s why you have all the … Depending on how the business communicates with each other and that’s typically how you create your workflows or your ETL, right? Or your load transfer. Right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And how you could send data from one to the other. It’s communication between one side of the business to the other side of the business. And if they change their communication model, or the way they communicate, then your program would change accordingly.

Sam Schutte:                Yeah.

Ron Dagdag:                 Or sometimes you’re introducing new ways and how they would communicate. So that’s typically how I would explain. I would look for different ways on how can I create analogies or how I would be able to give examples on how it is in the real world. One example that I’m talking about is like McDonald’s versus Subway. We were talking about machine learning and between the data scientists and data engineers, right? And what is the Cloud and what is the Edge? A lot of people are kind of confused the difference between when do you deploy it in the Cloud and when do you deploy your program at the Edge? And my analogy on that is, you know McDonald’s, right? McDonald’s, where do they buy their bread, right? Where do they cook their bread? Typically they cook it off site. It gets delivered to them at the store.

Sam Schutte:                At the factory or something.

Ron Dagdag:                 Right.

Sam Schutte:                Yeah.

Ron Dagdag:                 And compared to Subway, right? Where they bake the bread in the restaurant and that’s where, the way I see it, that’s where the Edge is. And there’s a difference depending on the need.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Right. In case of McDonald’s they’re working on make sure that it’s consistent offsite. And so there they have a big company or they have a supplier that send them that information or they send them bread, but in subway scenario, we’re baking your bread and at the Edge. You have the capability to adjust and give fresh bread to your customers. Right? And so depending on how fast you want to be able to adjust how to service your customer makes a difference.

Sam Schutte:                Well, and I think you’re right because if you look at that example, part of the reason they do that is because Subway is all about it being a custom experience every time and such, which was also sort of lines up with the idea of when you would sort of do Edge computing. Whereas, like you said, McDonald’s … A Big Mac anywhere in the world is always exactly the same, but probably any particular Subway club’s only slightly different every time you get it. Right. Being able to handle those sort of exceptions, you might think of it.

Ron Dagdag:                 Yeah. And the way you can customize it because having it on the Edge, the restaurant knows the capabilities or how much traffic they’re going to have.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 And so they can plan ahead or plan because they know they’re going to have this much customers around lunchtime and so they would bake bread accordingly, but McDonald’s once they run out of bread at the store, they’re not going to sell that product anymore until the next delivery.

Sam Schutte:                Well I think the costs make sense in that analogy too. Right? Because obviously Subway’s baking costs, if they’re doing it at the Edge, are always going to be a lot higher than if they did it in one central location just because of economies of scale and stuff. And they would, you know what I mean? Having to buy ovens for every location so and so forth. I mean it’s a good analogy I think. For sure.

Ron Dagdag:                 Yeah. In terms of costs, sometimes it gets expensive also too to … If I’m doing computer vision right, if I send my images up to the Cloud versus I send my images locally, it’s cheaper because once I bought the oven, it’s just there. I can just cook it every day and be able to process the data or process the … make bread every day.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Does that make sense? But if I have to send in the Cloud, I have a third party that I’m ordering every time. Right?

Sam Schutte:                Yeah.

Ron Dagdag:                 And that can get expensive depending on contracts.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 Does that makes sense?

Sam Schutte:                And transportation and everything else.

Ron Dagdag:                 And transportation. And so there’s, in terms of traffic wise, sometimes it’s … What if there’s a bad weather and you can’t get the bread delivered on time?

Sam Schutte:                Well and it’s interesting too, because I imagine if you think of how that sort of translates to the computer vision AI and machine learning models world. So if you had terabytes of images on the Edge that you were pumping into your local AI model and then moving the sort of training for that model into the Cloud, right?

Ron Dagdag:                 Yes.

Sam Schutte:                From there, well now you don’t have to move all those images everyday or something. You just have to move the sort of the learning basically. Right? And of course if you’re a company that has a hundred sites you don’t have to have a hundred sites all gathering terabytes of images. You could just do one as long as what they were doing was all the same. Right?

Ron Dagdag:                 Yeah.

Sam Schutte:                Essentially. So yeah. I mean these little unique or sort of use cases, you’ve got to get that right I think because you can make … It throws off the return on investment if the way that you deploy and architect a system is just way too expensive. Right?

Ron Dagdag:                 Yeah. And there’s also use cases like if you have oil and gas, right? If you think about it that it’s too expensive to send data to an offsite, right?

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 It has to be processed locally because they’re using satellite to connect to the internet.

Sam Schutte:                Mm-hmm (affirmative). Because they’re out in the middle of nowhere [inaudible 00:33:30].

Ron Dagdag:                 Because they are in the middle of nowhere.

Sam Schutte:                Yeah.

Ron Dagdag:                 So it just makes sense to just send your secret formula or, I call it your models, down to the Edge devices rather than sending each image or send each data to the Cloud to be processed.

Sam Schutte:                Yeah, it’s interesting for a solution such as that. You think about oil wells that because of their remoteness and the sort of like heavy, heavy industrial use, they probably have been pretty limited in sort of the ways or amount that they’ve been able to deploy technology for solutions until we’re sort of getting to this point, right?

Ron Dagdag:                 Yes.

Sam Schutte:                Because if your only option before was to build some sort of embedded device that ran microcode and whatever, because you wouldn’t be able to have a PC sitting there running on a generator with no cellular connection. I mean, this is also your limitations, right? But if you can put, like you said, I mean heck a battery powered Raspberry PI, right?

Ron Dagdag:                 Mm-hmm (affirmative).

Sam Schutte:                That is running in particular running code that’s easy to maintain, such as .NET code or .NET core. Well now there’s a whole lot of things you can do cheaply, right?

Ron Dagdag:                 Yes. So, but in terms of deployment, one of the things you have to think about when you’re deploying workloads up in the Edge versus the Cloud, right? Cloud, when you deploy it, there’s only maybe hundreds of places where you would deploy it, right? You deploy in the Cloud and you can easily scale it up there, but if you’re going to deploy it in the Edge, you’re thinking about thousands or millions of devices that you’re deploying to.

Sam Schutte:                Mm-hmm (affirmative).

Ron Dagdag:                 So that’s one of the things else also have to consider. So if you think about the McDonald’s versus Subway, McDonald’s only had to talk to a few vendors to deliver them the bread, right? Make sure that that one is … that they cook the bread or they bake the bread, but up to that certain standard. But in terms of each restaurant and training each person to bake the bread properly so that it’s up to standard. That can mean cost too.

Sam Schutte:                Sure. Yeah. And I imagine a restaurant such as Subway probably does have a lot more training costs than McDonald’s for their employees kind of for that same reason. Right.

Ron Dagdag:                 Mm-hmm (affirmative).

Sam Schutte:                Interesting. Well, so Ron, tell us a little bit about some of the upcoming speaking events that you’ll be doing in the next few months here.

Ron Dagdag:                 I will be speaking at Fort Worth .NET user group. I am actively speaking. I have a lot of speaking engagement around the Dallas, Fort Worth area and I will be speaking at also at Liberty JS in Philadelphia this October.

Sam Schutte:                Okay. And what does Liberty JS, what is that event about?

Ron Dagdag:                 It’s a conference about a lot of JavaScript and development over there too. But I will be speaking about web USB, web MIDI, and also web Bluetooth. How you can connect your Chrome browser directly to a Bluetooth device or directly to a USB device without needing to install device drivers.

Sam Schutte:                Interesting. Okay. And what’s the best way for folks to contact you if they want to learn more about what you’re working on or pick your brain about any ideas you’ve got.

Ron Dagdag:                 You can easily contact me via Twitter @RonDagdag, also on LinkedIn. If you type in Ron.Dagdag.net, it would point to my LinkedIn profile. So Ron.Dagdag.net is the best way to contact me.

Sam Schutte:                Okay. And that’s D-A-G-D-A-G.

Ron Dagdag:                 Yes, that’s right.

Sam Schutte:                Well, Ron, thanks so much for coming on the show and talking to us about these technologies that are sort of at the edge of augmented reality and IOT and what you’ve seen people doing with it. It’s been great to see sort of the reputation and exposure you’ve gained over the last few years by really putting a lot of your own time, personal time into this to help educate users in the community. So you should be applauded for that for sure.

Ron Dagdag:                 Thank you.

Sam Schutte:                Yeah. Well, and thank you and have a good one.

Ron Dagdag:                 You too.

 

Recommended Posts

Leave a Comment

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

Start typing and press Enter to search