Tricky Bits with Rob and PJ
Rob Wyatt and PJ McNerney discuss the latest and greatest news in the tech world and to figure out where things have been, where they are, and hopefully where they are going.
Tricky Bits with Rob and PJ
Why Are Maps Hard? - Interview with Cory Whittaker
Enjoying the show? Hating the show? Want to let us know either way? Text us!
Maps...we use them almost every day of our lives.....but...what makes maps hard? From the paper maps we used to pull out on car trips to the digital and dynamic maps that we have our smart phones, the problems around making and maintaining maps have been with us for a while. Fortunately...we've got an expert to help us out here.
Cory Whittaker, a long time self professed "map nerd", takes Rob and PJ on a journey on maps and how they are used in the modern day.
Come and join us for our first interview episode!
Ierengaym. com ierengaym. com
pj_1_02-01-2024_100239:All right, everybody. Welcome back to Tricky Bits with Robin PJ. Today we're really excited for our first interview with our guest, Corey Whitaker, who's going to help us, or more to the point. He's gonna answer the question, why are maps hard? We've been using MapQuest, we've been using Tom, Tom Google, we've been using Apple Maps for a very long time. Like what's what's left to be done At this point in time, we have GPS I think we're gonna collectively find out, and this ties back to what we talked about previously in augmented reality, that maps are a lot more difficult than you might think, and is a whole lot more stuff to be done there. So with that. Corey, welcome to the show as our first guest,
Track 1:Appreciate it,
pj_1_02-01-2024_100239:Rob and I would love to hear a little bit about your backstory, how you got to where you're at, and let's take it from there.
Track 1:the backstory with this is that, I've always been a map nerd, if that's the best way I could describe it, back in the day before, you'd be taking road trips with your parents, you know, you're sitting in the back of the station wagon, not buckled in, broken glass and rusting ails around you because, you know, that's how we we traveled back then is we didn't have anything to entertain us really as we're driving down the road and where I grew up in Arizona, it, we were going to go see relatives. It was a four hour drive minimum. So, you know, we spent a lot of time in the car. Well, I give my parents credit to keep me entertained. They handed back the Rand McNally and all of those just handed into the backseat. I would sit there and look at the maps for hours on ed. So you would say that from a young age. I, I just love looking at maps and knowing all the intricacies of them. So really that love of maps is what took me into a career first, starting in working for a city government, doing maps, and then all the way to working for the preeminent map company in the world, Esri or ESRI, they are involved in every major city in the us every state, federal agency about anything that you use maps for. It started with that. So anyway, that's kind of my backstory of where where I came to be, this map nerd, if you want to call it that, and I'll probably call myself that through throughout this whole thing. Going from that to helping everyone utilize maps and even if ways, like you had mentioned in directions and everything, there's ways that those maps that I work on and help get people to what they need to go and where they need to go to.
pj_1_02-01-2024_100239:So for folks that are younger than 30, and that might even be generous, R McNally, was a map that was on paper that was folded. And one of the most complicated things was to fold it back into place because of the complicated instructions.
rob_1_02-01-2024_100240:You can still buy these maps too. If you go to the gas station. They're still there. They come in books too, I think. I still have one in my truck and I just like the safety of having the paper map and like Corey said, it gives people in the truck something to do.
Track 1:Exactly.
rob_1_02-01-2024_100240:who've just flipped through the map while you're driving on a long road trip. Rather than play with the phone I mean, people just to look and you can just see, I think there's context. You can see things that are near you. It's like, oh, let's go here, let's go there. And I think you lose some of that with the digital maps. So the, uh, the paper maps are still in full force.
pj_1_02-01-2024_100239:Other than folding them, what makes maps a hard problem classically, Corey, and then what makes them a hard problem today? Because we've got a number of companies that are doing this, obviously, Esri, as I believe, the oldest and most preeminent one, and then a whole host that are also trying to solve these problems. But what problems are trying to be solved and why are they so difficult?
Track 1:Well, let's go back to those paper ones for a minute. And I'm sure everyone has seen a map of the world. You know, that's when we see all the continents, everything in there. But have you ever noticed that certain maps look different than others? And where I'm going with this is take Greenland, as a matter of fact. Have you ever seen those maps where Greenland looks the same size as South America?
pj_1_02-01-2024_100239:Sure.
Track 1:And there's others where it doesn't, where maybe we see a, a stretch of see one state. Well, this comes from the problem. I don't wanna say a problem, but the quandary. Of taking a spherical surface and putting it on a flat map. When I was going to college trying to explain this. This is what we call projections is how we manipulate the map itself to sit on a flat surface. And we did this, experiment to really help you understand this. And that is take an orange spherical object just like the earth, but peel it and take that peel and flatten it out. What happens? You have gaps in it. you have to tear it open to get it going, but to lay it completely flat, it doesn't go into a rectangle. It doesn't, it wants to stay in, its, spherical shape, so usually have gaps in it. That putting a spherical shape on a flat surface, and that's where these projections come in. It's basically how we manipulate those tears in the orange to be able to fit onto a square map and look somewhat decent. Because if you were to take, like I said, that orange peel and pretend it's a map, put it in there, on a table or whatever it might be, and then you overlay where the continents are, you'll see a big tear in, in, especially around the polar regions, you know, Greenland, the, Nordic countries, especially, Antarctica. You see some maps. Antarctica looks massive, but that's just because of this projection issue. So that's one problem
pj_1_02-01-2024_100239:This is not a new problem, right? This is a problem that's been around for centuries at this point in time.
Track 1:We're talking about the days of, you know, those early maps of when, explorers were first coming to the new world, where they are trying to draw those out. they were handmade then. But once we started to be able to see the world as a whole, that's when we really run into this problem of these projections. So that's one thing that really makes maps hard is not every one of'em looks the same. And the other part of this that I really think makes meps, I would say hard but difficult to understand the overall complexity of them, is understanding all the aspects of our life that have a, I'm putting in quotes where component. So this where component, I'm gonna take my example of my house here. We already talked about directions. Well, okay, how do I drive to my house? Alright, we have a point on, on the map. I have an address that's another set of information about this where component, but also I have electricity going. How does that electricity get to my house? We have to map out where all the, electric lines are and sewer water. Anyway, I'm, going over this as I want to instill on your minds of we have this one location, my house, and what all kind of items go into making that house function in a modern society like today. And that's just the example of my house. Let's think about supply chain logistics. Of all the things that go into bringing my. Keyboard that I'm looking at right here, from where it was made to the manufacturer, to the retail store, to this or this. Those are all where components. So this, uh, almost dance of all of the things that make our life today. The where component is key through all of these. So one of the things we had talked about earlier was, okay, driving directions. That's a tip of the iceberg, really with all the things that come into, are modern lives.
rob_1_02-01-2024_100240:I think, one thing that people don't realize is what you said of like, directions at the tip of the iceberg. Everyone interfaces with Google Maps and Apple Maps and they think that's it. that's the mapping systems that we know. But even those two systems are the tip of the iceberg. Like I said, the people you work for are backend map providers who feed data into this entire pyramid of mapping systems. And there's other companies who, do that too. Even open street maps does it in a communal, public sort of, way. it is interesting. That's just how big this, where problem is and how things get from A to B and how, not even people or things. It's like electricity gets from A to B
Track 1:mm-Hmm.
rob_1_02-01-2024_100240:the, the, uh, GIS databases that governments keep for exactly what you said, utilities, roads, whatever, and agreeing on anything is across those is gonna be next to impossible. So I assume there's a lot of fuzz in all of this too.
Track 1:Oh, absolutely. Keeping everything up to maintaining it, making sure that when a new road goes in, how does it get from the city? Or state that paved that road to those databases to Google and Apple Maps, that whole process is key for that end user of who will be utilizing these. So the, the dance I, I talk about that dance of how the data flows through. There is really a dance is the best way to describe it because there's so many ins and outs and everything, so very, very much so the last part I want to talk about why maps are difficult is when I went to college I had to take a cartography class. This is basically how to make a map, but one of our required readings, and this is the name of the title of the book, is How to Lie With Maps, and it's by a gentleman by Mark Moner, I believe, if I butcher his name, I'm sorry. But it is really talking about how maps. Can tell the story of what the Mapmaker wants to tell. So how to lie with maps is, is just as easy as understanding where an address is. Because if you think about it, making that council district map was telling a story. These political districts are telling a story., I was making a map of it was transit ridership based off of certain population characteristics. Well, if I wanted to tell the story of we're getting more, minority ridership in certain areas, I can adjust the filters and the parameters of my query in a way that shows that more than, okay, I'm just going straight. this is the number of people. So. That story that you want to tell with the map. And that's what that book of How to Lie With Maps is all about. Is everyone has a story to tell, but is there a agenda behind that story that you always have to kind of weed through as you do that? So those are just several topics of how you wanna take maps with a grain of salt because there can be a story behind it and you need to make sure you're getting your maps and your information from authoritative sources and things along that nature.
pj_1_02-01-2024_100239:Backing up, since we are, you know, a technical podcast, it's actually really useful to dive into how do we say where is where, like what is it that we utilize there? I mean, I can look on my phone and I can look at latitude and longitude. I've got coordinates, obviously there's GPS, but you know, you described earlier, you know, an address is some shorthand, but it sounds like, there's so much more that goes into knowing the address is the address. So could you dive deep a little bit into the technical side of this in terms of, how do I say where something is?
Track 1:Yes. That is a good question because there's, I don't wanna say conflicting views, but there are different ones along that line. I look at it as, as twofold, first off, as a mapping professional, I would prefer to use latitude, longitude for about anything. Because Apple, Google and who I work for with Esri do a great job of, okay, I would need directions to this latitude, longitude that it takes you along the street network. And it's not like climb over this mountain to get to it. I come at it from the very technical professional, aspect. Those people who, okay, I've got my phone and I just don't want to get to the nearest Starbucks. They're not gonna know a latitude, longitude. So it's really those who would utilize it as a consumer versus a professional. But, since I know, and I've worked on the creating the data that flows into those map programs, it's really based on latitude, longitude, and other projections, like we had mentioned earlier, let's use the example of my home here to map out the property lines. That's done through another type of mapping. So I guess there's another part of why maps are difficult is we call it, direction and bearing. So you would say, I find a brass cap, which is usually set up by county assessor's offices or whatever. And then I go, so this bearing, you know, 270 degrees, walk for 45 feet, find a point of beginning, and then you go from there and map out your actual parcel. I would say that's really how everything starts. So if you take the example of my home, it started with that, pointing bearing. To map out the actual property. And then as the streets come in, then an address is assigned to that from the city. From that, then all of the utilities come into this address. And so it kind of starts with that, pointing bearing is what we call it to an address to now my house has a latitude, longitude that Amazon or whoever is delivering can drop it off right there. So it's a process really beginning and starting when things are developed because if you're talking about raw land, you're either gonna be talking about that point in bearing or the latitude longitude.
rob_1_02-01-2024_100240:I have some of those point of bearing markers on my property here in, Boulder County. Aren't they placed like every degree or something like that, or some repetitive period.
Track 1:Yes, it's usually done by, quadrangles and, it depends on the county that's working in like Boulder County or wherever. One that I had the most experience with was Maricopa County down in Arizona. They had, basically divided the county into grids, and so every one of those grids had a point of beginning. And then subgrid, you know, there's Southwest corner of Grid two, and then there's the northeast corner of grid five. And anyway, that's how it usually starts. So that's really a surveying tool as much as anything.
rob_1_02-01-2024_100240:So that's how we get from an address back to a, a. Latitude, longitude. How does that vary across different countries? Because I assume all the countries are completely different, but ultimately there was a system for everybody. Maybe it's just literally you make your own grid And it's, it's you go from this address to this longitude latitude.
Track 1:so I've would love to get your take on this too, Rob, because I've heard ones in, especially these, property boundary descriptions in the United States is pretty straightforward because of this grid system. I had a friend who was also, he was from England, and he says the property descriptions are all over the place because they could go back to
rob_1_02-01-2024_100240:a, a thousand years old. And it's like goes to this tree and that tree's been gone for 500 years.
Track 1:Exactly. So maybe you'd start to talk on a global level. I would say at least in the GIS or the geographic information system world that, I work in is latitude, longitude is really kind of taken that over, but when you get down to the local municipality level, that's when you need this more detailed level because with, any kind of GPS device, there is a decree of error built into it. So when I take my phone out there, I can get within 15 to 20 feet of the actual location, where most of the time, hey, well that gets me close enough to the address that I'm going to. But we need to get down to the level of surveying quality. You need hyper accurate. We're talking sub centimeter type of GPS, and so that just requires more equipment, but that's when you crossover into that professional level.
rob_1_02-01-2024_100240:Yes. And then I guess the modern. Version of the 2D map projection problem is the whole GPS data problem as to how do we convert these, I think GPS reports in earth centric coordinates is it's bare basic system. It's based on the center of the earth, and it's a spherical coordinate system and converting that to latitude, longitude, given that the earth isn't perfectly spherical and altitude's incredibly difficult because GPS isn't great altitude to begin with. And then we have obviously varying ground altitude all over the world. So I think the one we use now is the GS 98 or something like that that everybody agreed on, on 94,
Track 1:the prominent one that, we utilize here at Esri is the, WGS 84 standard, which is one that, goes worldwide. Again, when you're using the devices, it can be a little, can have a degree of error in there. But for the most part that WGS 84 is the one that we utilize for all of our web-based mapping, products.
rob_1_02-01-2024_100240:I think that became more or less the standard. I I do know in the past people have been like, search and rescue a missions have missed people because they've been using a different data on the GPS than what the coordinate was reported as. And they went somewhere and maybe close by, but not close enough. So these projections still exist and we all have to agree on, like I says, on the datu of what we're gonna use for GPS. Otherwise your, lat long isn't my, lat long.
Track 1:Exactly. So there's ones in, I've worked in ones from 19 27, 19 84, I believe. You're right. There is one from 1998, but I don't think it is as, adopted as some of the ones from like 1984
rob_1_02-01-2024_100240:Yeah, I mean, mathematically, I guess long longitude latitude is exact for everybody. It's this exact position on the surface of,, the earth, but it's technically the surface of a sphere. So that's where the error comes from. And then obviously the, measuring error is due to GPS and how, if we projects that,
Track 1:Yeah. And how the satellites, coordinate and triangulate between multiple in there. So yeah, this is one of those rabbit holes that I could tell you we could go down because there's the built in discrepancy, if you wanna call it that. But then there's, one that personally, I haven't worked with it, but the, the military GPS is actually even higher accuracy than consumer. And so there's, to preserve the wavelengths, I can't remember the exact terminology for my classes that I took, but yes, it was because the military needs more accurate that the consumer version of the GPS that we all access has that degree of, error built in to preserve that frequency for the ability for the military to have even higher, accuracy with your GPS.
rob_1_02-01-2024_100240:Yeah, going back, I think that was called selective availability and it was, uh. It was on for a long time. The military one is still a completely different system, but selective availability was turned off by the Clinton administration. And since then, everybody has had the full accuracy of valuable and they, I think they can optionally turn it back on again for certain areas and things like that. But I don't think they ever have done, the military wanted its own thing. They have their own frequencies and their own encryption keys that they have, they need to access it. But the civilian one is now currently running a, the maximum resolution that it can, run out. And I think that's mostly competition too. It's like if Europe and China a release and Russia have obviously glassner and if they're releasing all of theirs, then why bother making GPS worse? And most receivers today will receive anybody's system. So that's interesting that you get multiple satellites in view and it, they can be different systems and it's all by the magic of software. All seems to work together.
Track 1:Exactly.
rob_1_02-01-2024_100240:So going back, we have these online maps now, and obviously all this is connected. We have maps that are from GIS data. We have maps that are from satellites. We have maps that are from high altitude, low altitude planes. We have maps that, or images that are from street view type cars. How is all that layered together to make that base map that you see when you open Google Maps, for example? And how do we know that the bits that are on the map are actually in the correct place?
Track 1:I've heard multiple people bring up, how do we get this information into the hands of consumers? Know that, from my time working in a local government, the city was the authoritative source for especially like the parcel information, the streets, the addresses. How does that get into these, consumer products of Apple Maps, Google, and so on? From the GIS side, again, that's the one that I know the best is these cities partner with the larger organizations. for instance, when I worked for the city, we were partnered with Esri to send them our authoritative data. So let's say a new subdivision comes into the city to take that information, we would put it into our system in GIS process that to where we'd send it out to other organizations, you know, the state, the county, or so on. But we also would send that to. In this case, Esri, and if I remember correctly, we sent that to Apple as well. So they worked with the local cities to get that authoritative data. So that's more of like the base map, like you were saying is, is getting the right information in there.
rob_1_02-01-2024_100240:I was thinking about the other way of what if the GI, I mean my land here was mapped wrong, so I have little faith that the GIS data is correct most of the time. And what happens when a satellite image is obviously correct? The road is there. I can see it pixel by pixel. And the GAS data says it isn't there? What happens there? Do you just move the road to where it visibly is of like how, how is that? I can just imagine this happening thousands of times every single time a day of like something's changed and whatever. And at least initially when we first mapped the world with satellites and gave it to Apple and Google and people like that, and people like yourself of just that massive correlation job of making sure we have a master base map that we can all kind of agree is correct.
Track 1:Mm-Hmm. So I guess it would depend, like you said this happened, most likely happens thousands of times a day, how it gets fed back into those. Now I could see if one of the product projects that I worked on specifically was change detection of seeing aerial imagery and being able to see what has changed over time. So you look at historical images and, I've even worked with a, imagery company to where they would do daily flyover satellites of North America and they had that data served up. So doing that change detection and bringing that back in., in my current position working with Esri, I work with customers across the country and I could tell you ones that would say, oh, you found an error, let's fix it, and would be happy to do so. But I can also see ones of like, oh, that's not our problem, not our issue. So that, that goes back to, I would almost say the varying complexities and complacency of government agencies. Because really if you think about it, like your, the road, if it's a public road that should go to and work through the streets department or the dots or whoever that might be, and some would be happy to bring in those kind of changes. And I could tell you some would be saying, well, not really. Uh, we'll send someone out to get to it. And at the speed of government, you know, it takes six to eight months to get someone to go out and verify that that road is in the wrong place.
pj_1_02-01-2024_100239:So this is an interesting point because it, it hammers home that the governments are the authoritative source of this data. So even like in Rob's case where there is this clear discrepancy between the government data and the maps, and you can point it out, you guys aren't able to correct that yourselves. You have to feed it back into the authorities in order to close the loop. Is that a fair way of putting that?
Track 1:I would say yes, partially, I can't speak to the exact way that say Apple or Google maintain their street network and everything, but at Esri. We rely specifically on a program of partnering with the local cities to get their street network data. So we're not gonna go in and change Rob's road, even though it's blatantly wrong, until the county or whoever is the authoritative source for the data send sends us an update with those
rob_1_02-01-2024_100240:think on the flip side of that, I think Google and Apple do actually change it in the data that they get. I believe a lot of it's done through machine learning and AI type processes these days. But you'd see areas like you say, like you could look at upon the map and the satellite road is over here on the left, and the actual rendered road from GIS data is not overlaid on the visible road. That very rarely happens today with Apple or with Google Maps, so I can only assume that they are fixing it from the data they get, so they may not be authoritative with the local governments.
Track 1:Mm-Hmm.
rob_1_02-01-2024_100240:Which I think is what you're saying you are, but they are correct from the roads actually here.'cause the satellite said it was, and it's hard to say that the image is wrong and I think they do all that through just computational photography type operations and machine learning. And maybe there's a bit of human intervention in there when it can't be done automatically. But I do believe that that is done quite a bit.'cause I, I know of a few roads around here, if they were roads that were wrong and now they haven't been wrong for years.
Track 1:It is very true. And. That's, again, I think the close relationship that Esri has with these local governments because they provide the actual mechanisms for maintaining those roads in the GIS system. So they rely on the authoritative source for those before pushing it out. Because if, you know, Esri does have base maps as well that utilize points of interest and all these others that, um, is used in multiple aspects. So yeah, there, there's, there's different parts to those. Again, more of on the side of the professional users of I'm maintaining that road, I am putting that into the correct location versus someone who's needing to drive to Rob's house to get, uh, to drop off a package or something along those lines.
pj_1_02-01-2024_100239:What.
rob_1_02-01-2024_100240:I, I do think the current map systems that we've brought mapping into the consumer space maps are now a consumer product where previously, other than printed maps, they were very much a professional product, especially in electronic form. So with all of the resources and assets of Apple and Googles of the world that are making these. Electronic maps. I think the customer consumer pointing side of maps is these big mapping applications, which have progressed a long way from giving me directions from your house to my house. And now maps are an entire ecosystem of
Track 1:Mm-Hmm
rob_1_02-01-2024_100240:backend, data. There's all the backend processes that Google, apple have to do because like I said, there's Apple and Google. Both know where traffic lights are and where speed cameras are, and they probably weren't entered by hand. They're probably all done via image detection. And image recognition image changes across time. So there's a lot of technology at the back end, at the front end. We have these now incredibly complicated applications that render the world in 3D and you can see buildings and you can see, even individual, like may only be a square for a house where a skyscraper might have the actual mesh of the skyscraper. So it's the correct shape, but just the sheer amount of data that these maps now contain for the consumer space and for consumer presentation is insane and feed all that back in. Now it's not even a static system, like maps are always static. They print them, they'd last a hundred years and then you print them in a book and they'd last a year. Now they're truly dynamic. It's like traffic maps are live, updated from anonymized data, from how many iPhones are slowing down at this part of the road, or how many Android phones are slowing down from this part of the, road and speed sensors and cameras watching the road that are then machine processed to come up with,, flow maps and current live state maps, and then directions, route around the slow parts So. We've come a long way from having a paper map in the back of a pickup truck.
Track 1:To keep that, you know, annoying little kid in the back, entertained on a four hour drive. Oh wait, that was me.
rob_1_02-01-2024_100240:So I remember when Google Maps first came out, well before Google Maps, obviously we had, we had MapQuest and anyone who's remembers MapQuest in the early two thousands. It was great. It was basically an online version of the older products, which were like order route, auto map. they were text-based DOS apps, I think made in the UK originally. you could type an address to an address and it was fairly mundane. It was technically just like a simple graph. Solve the graph from A to B, uh, add up the distances as you go, or add up time or whatever you wanted to solve for.
Track 1:And then print that off and take you
rob_1_02-01-2024_100240:And then.
Track 1:And
rob_1_02-01-2024_100240:Print it off, take up within the car. And then, but this was like the eighties and this was magical. When you first got this app, I think someone gave me a, uh, floppy disc with it on or a couple of floppy disks and it was like, wow, that's amazing. I can get directions to anywhere. And all those questions in my mind came to be of like, how did you update this? How did they get this? And it was fairly crude back then'cause it didn't actually show you a map. They just abstracted the concept of a map into a, like a abstract graph in computer terms. And then they solved the A to B problem on that graph and they assigned that back to rows, things like that. And that's how they got the directions. Fairly crude directions. It would be, wouldn't it be wrong? It, you would get there, but it, you had little faith like it was the best route or the fastest route or the road even existed anymore.'cause sometimes it was like, go down this little dirt road. things like that. But it was just, that was just the connection of the digital data to the real world. Data in the eighties wasn't very good. But then in the two, early two thousands we got. MapQuest? Well, probably the late nineties. Actually. A MapQuest was great because it would, show you the map. It would still give you directions, and I remember it would give you a little map, a static map for each turn that you had to make. So when you printed out the directions, it was like 15 pages and that was what everyone used for. Everyone would print the directions for, they left the house, they wanted those smartphones, and it was great. You've got the actual maps so you could see what the road looked like. And then I remember when I first saw Google Maps, my mind was completely blown because of a few things. One, you could scroll it with the mouse, you could just grab the map and move it. And I think back then Google just used pre-rendered like 64 by 64 tiles, and there was literally just a stream of PNGs coming down to your machine as you scroll the map around and they had them all on the server and they'd update them and things like that. Today it's not that way. They're doing a lot of actual rendering from. vector type data in the web browser, but back then it was just there. But the other amazing thing that Google had from day one was satellite maps. And I guarantee you the absolute first thing everybody did was to go look at the house from a satellite and view it from high altitude, view it from low altitude. How close can you get? Can I see my car in the, driveway? And then everyone played the game of trying to guess when that photo was taken because of like, oh, I don't have that car anymore, so this must be a year old. And things like
Track 1:cans are out there.'cause it must have been a Wednesday that they took it on that day. Yep. Mm-Hmm.
rob_1_02-01-2024_100240:yeah, ev, ev, everyone played that. Everyone played that same game in like 2004 whenever it was, when Google Maps first appeared. And of course it could do directions too, and it would draw on the map and you could see it and you could scroll the map. While it had directions on it, it was an absolute game changer. And then, I guess this was the first I. Consumer facing like map product, as we, as, as we'd recognize it today, it's, we'd recognize MapQuest is the old internet where nothing was interactive. Um, I think Google Maps is the first one we'd recognize a, a current map product. You can see its history. And I also think then Google then took a huge step forward with street View, where we're like, let's just go and drive on every single street and get GPS data. And that's the start of the goal today of where we'd like to have a, some sort of 3D app that shows you exactly what it would be like to drive that way out. I'm surprised that there's no VR apps yet that do that. And. I think AR has a huge place in maps too, because if we could get wearable AR we could use in a car, then we could do things like markup turns. I always thought the only grail for an AR app would be to literally have something like Google Apple Maps that literally draws on the road, like follow the yellow brick road, like the blue line that's drawn on the map, is drawn on the road where you have to go. That's incredibly hard to do. We talked about this in a AR episode. lots of understanding of what you're looking at. It becomes an AI rendering problem, ar problem more than a maps,
Track 1:so going off of that, the ar part, seeing that yellow brick road like you were mentioning, one place that I see in, like they said, the GIS world is okay, you have your directions. This is again, more on the professional side, but if you were to use your augmented reality without the issues like you were mentioning in the episode of being outdoors and all of that, but think of for the road crews being able to stand on the road and see exactly where the underground utilities
rob_1_02-01-2024_100240:Yeah, I think that's all in incredibly, useful, but I'm surprised even for VR that doesn't, exist. I think we're heading there with the 3D views in Apple View, apple Maps, and we have the same in Google and Google do have some AR component where you can have the camera and it will basically point, like, go left here. But it's, it's just icons that are put on top of the map. It isn't really like integrated into this understanding of the scene that it's looking at. So we're definitely heading there and it's just amazing how far we've come from what we originally did for maps. We print them out. And today we have this device that will just tell you, it's like, turn here, turn here. As as you're going along, what we don't do anymore is ask for directions. Very rarely does somebody ask you for directions anymore.
Track 1:I actually had to do that recently. I was outta the country and I didn't have data service, so I had to do the old ask for directions, and it felt so nineties to me. You know, I could almost hear the grunge music playing as I did it. But yeah, that's just something we just don't do anymore because we either have it in our cars or on our phone or some way to be able to find the directions for that.
rob_1_02-01-2024_100240:So I guess another question we have today is these maps are now full ecosystems. They're full platforms and they gather a lot of data about our movement and our location and. That's used to fund ads and which fund the platform, and is that ultimately worth what we're giving up to get what we're getting back? I mean, that's a question of technology and society as a whole, but maps are incredibly sensitive due to location. And you can even like visually mine map data so easily. If I know you were at this house three nights a week and then here for three nights a week, you can imply a lot about that just based on static location data that you just looking at with your own eyeballs. If you start mining this data with machine learning algorithms and supercomputers. The amount of information you could derive about somebody solely based on their location and the time of that location is ridiculous. And we know Google already do some of this because they have the, your his, your location history, which is kind of like, well, I went to all these places and it knows you've been here before. And they say some of it is anonymized, but that's more like the traffic flow data, though.
Track 1:that's what they sell to companies like Esri, because we get that type of data to like traffic flow. one that I use specifically is flow around, subway and rail stations, is using that data to which platform for this particular train is more utilized from a choice perspective rather than just going to the first one. So that's one of some of that anonymized data, but you're, you're right, that. It tracks that information I looked at my phone the other day. We were, it was a last Friday, and my wife and I usually go out for drinks on a Friday night and I was finishing up work, opened up my phone. It's like, oh, are you headed to this location at, uh, for, I mean, it knew where we go at a certain time and probably knows exactly what we're doing there because it has the, the business information. So it knows that I go to the same place every Friday for drinks. So, and you're spot on. I'm sure the, the tech companies are gathering that information and helping sell ads for me
rob_1_02-01-2024_100240:Yeah, absolutely. It's probably not gonna give you the ad'cause it knows you go there, but it's gonna give someone like you the ad because it wants, yeah.
Track 1:cause you know, we, I go to the same place as a lot of
rob_1_02-01-2024_100240:But I don't think they sell that information. I think that information that they have privately on you, I don't believe, I don't believe they sell it as is. They may anonymize it and sell it in other ways, but I don't think they sell. This is PJ's information. This is where we went. But then there's a question of law enforcement and subpoenas and all of that at the back end, which is also an a very interesting thing. And just the amount of information that they have based on literally you were here at this time
pj_1_02-01-2024_100239:so Rob, I think you brought up a really good. sphere here in terms of the consumer space and it's really interesting to see, Google and Apple able to make these changes because they have consumer clients. Corey, the idea that the enterprise slash professional space with these municipalities where it's, this is, you're going back to this as an authority to whatever you can speak of. What are some of the major use cases that you're seeing for new avenues for the municipal space, that enterprise and professional space? So, like Esri obviously is, is occupying something different than a B2C? Obviously there may be some B2C components in there, but like can you talk us through a little bit about where the latest, greatest sort of advances in avenues are happening for municipalities or for, you know, the clients you guys typically say.
Track 1:So we've already kind of talked about a little bit this topology. If everything's connected, you know, there's a way that traffic is connected to how I do my day-to-Day job or so on. One of those that we're seeing a major growth in not only municipalities, but state governments. the customers that I work specifically in public transit of real time tracking of vehicles or. Anything that you need to track and analytics done in real time? Lemme give an example again, my specialty is public transit, so taking where that bus is, and one of the things that every transit agency has to report to the federal government is what is the on time percentage of that bus? So across the system, we're at 85% on time and so many minutes late or so on, but how do you track that? Before it was the bus would gather its information, on its on the bus itself, and get back to the bus yard, download that data, it would be processed and help with your on time performance. Tracking these new technologies of being able to stream the bus locations of being able to see that, map of the bus's movie. Now with this real time, not just tracking, but analytics, I have the ability to take where that bus is, compare it to the schedule that this bus should be at, this location at this time, do the analytics and give me a number in near real time. We're talking 15 seconds and being able to track that over time. So that's one that we're seeing a lot of use for that. Another good example is especially in cold weather climates tracking where the snowplows are, being able to see where every snowplow is in the state of Colorado and see where they've plowed, not only where they plowed, but what materials they've put down and showing that to the public in a way that, okay. yeah, we just had a big snowfall of, you know, 18 inches, whatever it might be, and you're needing to get out of your neighborhood. Well, the snowplow is scheduled to be there today, and it's six blocks away and it's putting, whether it's gravel or salt down or whatever to help with that. But that's being able to track those kind of locations in real time. So that's one thing that we are seeing a lot of use for, and it's growing. I talked about vehicle tracking, so snow plows, buses, I've worked with a customer who use that for the collars they put on wildlife. So the example I want to give for that is salt Lake City Airport. You know, there's the Great Salt Lake right next to the airport, but they also have a good amount of pelicans in that lake. I wouldn't think so, but they are. In fact, I here in the summer here in Colorado, I see pelicans in the lakes around here. But those pelicans can cause a disruption to the traffic coming out of the Salt Lake airport. So they actually, the Department of Wildlife put trackers on these pelicans so they can see which way that the pelicans are flying and direct traffic of the, airlines to avoid the pelicans to avoid bird strikes.'cause you know, we saw what happened in there in the Hudson, in, uh, in New York, I can't remember how many years ago, where they had a bird strike and had the plane flew into the Hudson. So these are examples of using that real time analytics. I mean, tracking a bird. We got 50 birds outside the airport, seeing where they are. Okay, we need to make sure the planes go the other direction. That's tracking and analytics happening as. The situation happens. So that's a big one that we're seeing with that.
rob_1_02-01-2024_100240:So going back to the real time data you get from people like Google, is that truly real time?
Track 1:From Google? Well, it depends on the source. Like the ones that I work in, uh, especially in transit, we're getting them within five to 10 seconds after the device is, is ped,
rob_1_02-01-2024_100240:so, are you getting that from like Google doing, its anonymized Android tracking, or are you getting it from the various little RFID type antennas that has placed along the freeways or radar sensors?
Track 1:so are we talking Traffic
rob_1_02-01-2024_100240:Traffic flow? I'm thinking about like where's the where's the bottlenecks and why is this bit here slowing down?
Track 1:that's the anonymized data that we would get for those that I, I actually utilize in my transit models to show that at a certain time of day, the traffic slows down at this location. And we can see the impact it does on the schedule. The other ones that I see, like I mentioned, the utilization of tracking the phones, that's also an anonymized, you know, which part of the rail station are they going to or anything like that. That's the anonymized, we don't get anything specific, but there is a push in. Uh, that was going to be the next part that I talk about of where we seeing this going is not just outdoor mapping, indoor mapping, so utilizing, the various indoor positioning systems or IPS A good example that we use for this is, a convention center. So you've got a huge convention center, say like San Diego, one of the biggest ones that, that was utilized. Being able to put these indoor positioning trackers is think, okay, I'm going, I want to go to a session in room B 15. Where is that at and how do I get there? So the same thing of indoor routing, not just along the streets, but actually routing you within a building is something we're seeing a lot of use at in convention centers. Airports is another one of helping people get from one part of the airport to another. I was just in Houston and that airport, you know, I was going from one terminal to another and trying to get through there. And the amount of time that I had from make my transfer was crazy. But that's this thing of, we think of maps as the whole wide world right outside our window, but really it can be indoors too. I mentioned airports, convention centers, hospitals is another one. a. Good utilization that some local municipalities are doing is fire departments utilizing this indoor mapping to help with the evacuation of a building or if they're searching for it. Police departments can use this too, to, alright, we're having to clear a building because of a police situation. Where are the little cubbies and Heidi playing that we would need to be able to look out for any kind of thing? So mapping isn't just, like I said, not just the world. It's also indoors that we're really seeing a lot of, growth
rob_1_02-01-2024_100240:What is the technology behind those indoors? I assume it's just like static. Known position, some sort of node, which is wifi, Bluetooth, ultra wideband, whatever technology it may have, but it has to be something that your phone has'cause there's no
Track 1:correct,
rob_1_02-01-2024_100240:for this.
Track 1:yes. The tools that we're putting together utilize those indoor positioning systems. And with those nodes throughout the building themselves, it doubles as the GPS that you would use outdoors. And so you can actually get very accurate information. At my company's headquarters in California, if I wanted to say I've got an issue at a certain room, we've mapped all those out to where I can have the, IT people come in and know exactly what projector is having an issue. Because you have the indoor location, you have your assets in there, you can actually go through and make a digital twin of your entire building.
pj_1_02-01-2024_100239:This is pretty awesome actually. and you got under the edge really of where I wanted to go next, which is, okay. So indoor mapping looks like this. Giant frontier. That is, we're on the verge of for maps. What are some of the other exciting areas that, you know, your work maps, where's this stuff going?
Track 1:I think, one of the big one is AI or machine learning. We're seeing that in a lot of spaces. given example of that is a colleague of mine built a model. He works specifically with commercial railroads, so you know, union Pacific, BNSF and so on. But they came to him with a problem of, okay, we need to track, how many types of rail cars are leaving a port or a depot or whatever like that. So he built a machine learning model that watched a video feed. And actually it's like, okay, as this train goes by, that's a rail, that's a, container car. That's a. one for liquids or, and that sort of for cars. So it actually gave the train location and the time and actually counted the types of cars through there. So, the machine learning can utilize so many different things like that. We'd already talked about image detection. You know, that was one thing that you brought up, Rob, was being able to find where those roads are not, in the right location. An example that I've helped a customer with was actually oil wells in, the Permian Basin, you know, New Mexico and all those to where the, regulatory agency was using that, these change detection. I had used that phrase earlier, okay. These satellites are going over daily, maybe every other day, to where they were using that detection to locate drilling without permits. So it was actually a revenue maker for the, state itself because it was able to say, all right, you say we've got a permit for a well at this location, but from the aerial imagery, we're able to see that you're also drilling here at this other location. One of the biggest ones then this came, first, came out, is also, um, around swimming pools. And I know that sounds, why do we need to look at swimming pools? Well, in Southern California, they had a lot of problems with, green swimming pools or pools that weren't maintained. But if you think of how many pools are within Southern California could take, all right, let's zoom into each house and check those. Now with this machine learning and this ai. It scans the entirety of the aerial images you train to say, this is what a green pool looks like. All right there, you know, gives you the list, the addresses, and all the owners of these pools that are not being maintained for, you know, that could have mosquitoes or West Nile virus or anything like that. So that's really where another one of those uses of machine learning and AI come in. Helpful.
rob_1_02-01-2024_100240:This sounds to me like a somewhat of a dystopian future. It's like, I mean, how, and this is way off the subject of tech, but how far can municipalities go with this? I mean, if you put a shed up and they go, oh, you built a shed without, telling us we found it by aerial imagery. It's like, where's, where's the line as to what they can. And force based on this technology. I, I think that's something that's gonna be up for debate in the future when machines are looking for things that they can look for instantly that would take people years to do.
Track 1:Your point is valid of when, what is the ramifications? Okay. What are they gonna look for next? Because of this ability to have these images of my backyard, of your backyard, of where everyone else's. And how does that play into regulatory surfaces?
rob_1_02-01-2024_100240:Absolutely, because I mean, you can say to Google like, this is private property. I don't want you to look at it. And they'll, they'll honor your request and they'll fool you outta street view and things like that. Why do, why is the municipality allowed to do it wholesale, regardless of whether you want it to be private or if you don't?
pj_1_02-01-2024_100239:Rob, I think this is why we moved to places with lots of trees. Unless Corey's gonna tell us, basically we, they've got some other, you know, penetrating radar that could be used to, get through that.
Track 1:There is, with lidar, you can actually remove the. Foliage. So if you wanted to see like underneath the trees, there's certain types of lidar and radar remote sensing is what we call it to where you can actually take that ground cover and be able to see that. I, I don't want to make, you know, you're gonna move now. PGII realize that, but,
pj_1_02-01-2024_100239:No, no, no. I just, it just justifies now covering everything in aluminum foil, uh, to bounce back the lidar. It's, you know, it's fine.
rob_1_02-01-2024_100240:Yeah, it's, it's very interesting have all of this, all of these disjoint technologies all feed back into one big database of, like you said, where, and you factor in. This with like topology and everything else that these databases have, and we end up with a pretty interesting map of the world that we've built on top of the world that was already there.
Track 1:That where? Component that wear variable you could go on for days. Tracking all the little nuances of our lives to that wear.
pj_1_02-01-2024_100239:Even with all of this data and technologies, it goes back to a really interesting point you made really early on in this, is that there's still some human I. That can draw arbitrary lines through neighborhoods to tell a different story. but It seems fascinating that there's still this arbitrariness that Im like, impacts all of this objective
Track 1:Mm-Hmm. I wouldn't be surprised if it utilizes AI for those kind of things now, but I would hope that there's always a check and balance put into that. Being able to have a impartial view. I, I know I'm reaching here for the stars, but hopefully to this ability to always have an impartial view of it, but the world is how it's
pj_1_02-01-2024_100239:Any, other thoughts in terms of where things are going from a technological standpoint? Are there any bits of technology? I mean, we, we just talked about, I mean, obviously satellite data as imagery, layering in lidar. What else is on the forefront here in terms of how we improve maps
Track 1:So one that I am,
pj_1_02-01-2024_100239:that you can talk about?
Track 1:I could talk about, well, there's one that I'm hopeful I would love to see it take off. I, I am a little cautious on it, is utilizing game engines where, where I'm going with this is take a forest fire perhaps, but you want to be able to see what it's like on the ground there. But you have all this great data and authoritative and accurate data, but you like the episodes that you had on AR and VR managing to be able to put on some goggles and actually see what's happening on the ground with a forest fire in virtual reality, but utilizing real data. I think that's one that's has a lot of potential. I've seen something similar to this with, we call it bim or Building Information Model. So this is like the 3D buildings we had talked about, a little bit earlier, but being able to utilize and stand in a building that hasn't been built yet in virtual reality, that's one that I see. But also in places that, maybe we don't have access to or for safety reasons, like a forest fire. Being able to see the extent of it, what it's looking like at a, in a location that you can't visit. Being able to take the person who would be seeing the 2D map, take them to that location, show them real data, but within a game engine, you know, unreal or whatever it might be, to really take that immersiveness of taking the data, you know, your maps. I I don't want to call it four D, but whatever. But you have your 3D maps. Let's take it to another level. And I think that's where this possibility of utilizing game engines can go as well.
pj_1_02-01-2024_100239:Do we have the data to do that today?
Track 1:That example of a forest fire, I have seen a demonstration of it. We have the topography. We have the elevation information. Let's say that there's buildings there. We have the footprints. One of the things that I did a lot of work on, especially for my, graduate degree, was forest fire data. We have satellites could tell you the exact temperature, how intense that the fire is burning from satellite data. So you can actually combine all those two, all those together and put it in a, in a virtual reality view and see what it's like to be in a forest fire. Or wherever you're needing to go. So really, we have the data to do it. It's just combining it into a medium and a platform that we can see it all together.
pj_1_02-01-2024_100239:Rob, maybe we should make that one of our next projects.
rob_1_02-01-2024_100240:Something there for sure. But that would be great for the people. Like say it's like the, the firefighters and all that. I think the AR wanted be even better'cause they could look at the real world and see it marked up. But I think we're a long way from the.
pj_1_02-01-2024_100239:well, Corey, this has been amazing. thank you so much for coming on. I know we've only just scratched the surface on so much of this stuff, so I think we'd love to have you back on and do like some deeper dives into a lot of the, I mean, we covered a lot of ground here from the past to the current state of things, indoor mapping potentials for the future. So we would love to deep dive a little bit more on any of those things and also any of the new and exciting things that are yet to come in maps.
Track 1:Well, thank you both so much for having me on here and as a map nerd. And I say that with pride is that I, from the kids sitting in the back reading the Ran McNally to today, being able to think of maps in a game engine and indoors and all of this I'm happy to talk about this anytime
rob_1_02-01-2024_100240:Awesome. Thank you for coming on, Corey. That was a great conversation.