Bojan Šavrič: Equal Earth Projection

Welcome to the GeoMap Podcast, where we discuss geo-innovation in any and all forms, whether for fun or profit. Welcome back to the GeoMap Podcast. In fact, I think this is probably going to be the first podcast of 2026 or very close to. I hope you had a good seasonal break and a happy new year. This afternoon or if you're in California this morning, my guest is Boyan Shaverick. Boyan is a senior software development engineer on the projection engine team at Esri, where he works with coordinate systems, transformations and map projections, which is all stuff that I do not understand properly. Boyan is the co-author of several map projections for world maps. We're going to get into a couple of those projections in a minute. He's also an enthusiastic lover of the mathematics behind maps. In other words, he's an absolute map geek. He's just the kind of person we want on the podcast. He's also a member of the International Cartographic Association Commission on Map projections. This is a real expert that we got on the podcast. Boyan, welcome to the GeoMap Podcast. We should have had you here years ago, but I'm glad you're here today. Thank you very much and thank you for, hello to every listeners, of course, and thank you for having here. It's great. It's awesome. Great. So, first up, the people listening to this podcast, some of them know everything there is to know about projections or at least they think they do, and some of them are like me on their map levels, but they're not quite that technical. So, can you explain what projections are and why they're important? I usually explain map projections with a process of like peeling the orange, so to speak. So you can peel the orange in many different ways, and you will always have basically hard time completely make the peel flat without kind of tearing or squishing it, so to speak. And each time you peel the orange, you peel it a little bit differently. So each time you're going to do it a little bit differently, not the same, so to speak. And every person is going to peel the orange differently as well. So every single case will be different. So yeah, every flat orange peel is therefore a little bit different. So the same happens with the surface of the earth when we peel it and we make it flat on a map. It's basically impossible to completely flatten the earth's surface without tearing, stretching, or basically squishing it. And basically every map projections somehow deforms the earth's surface, either deforms the distances, deforms the shapes, areas, directions. Basically everything is somehow differently distorted on a particular map. And therefore we have hundreds of map projections that we can use for our maps, but this is a good thing because with so many options that we have available, we can actually control this surface or as a surface deformation or as cryptographers calling it projection distortion. So selecting a map projection for our map is basically making sure that these deformations are the most optimal for our map that we are trying to make and the map basically matches the purpose of the map. And our goal is very simple, like representing the earth's surface on a map as closest to the reality as possible, but because every map projection does this a little bit differently, this is the reason why they are so important for the map. And just so I'm clear here, these differences are more pronounced in a world map than they are in a map of a small country or a city. They're correct, yes, on the world maps, like it's much more easy to see these differences because you are representing the whole world. You have a large area that you're mapping and you have a lot of material to compare with, like land masses from different areas and stuff like that. So yeah, there are more pronounced there, and therefore we have more projections that are appropriate for the world maps in general, right? With a large scale mapping where we're making a map of the smaller area, it's much more harder to compare because we have less things to compare, but because we are mapping small area that is way more, we can control that way more on the small maps. So it kind of is like, yeah, we have a lack of comparison, lack of material that we can compare to in the small areas, but at the same time we also have the very nice way to control that because it is a small area, it's much more easier to control than the entire world at the same time. So you said that projections can optimize for one of the distortions that occurs when you flatten the earth's surface. So it might be shape, it might be distance, it might be area. Just to give a couple of things up, when would you use different projections just generally, why would you use a different projection? Why wouldn't you use one projection and say, that's the best projection and I'm going to use it for everything? It depends, basically, depends on the purpose of the map. So you cannot just choose one projection and use it for absolutely everything that you're making, right? So first question is, like, what is my purpose of the map? Like, you made the example of area projections, equal area projections, like the projections that preserve, or I should actually say area preserving projections. And that particular projections, because they preserve relative areas, they are appropriate for thematic mappings and the mapping that kind of compares the differences between the stuff, between the features that we're trying to make. And because of this comparison, we have to make sure that the ground underneath is basically kind of relatively correct in this particular case. That would be area, right? If we're talking about other properties, like, for example, shapes, there is no map projection that correctly presents the shapes. Even on conformal map projections, the shapes are still distorted. And you can actually see that on the website, there's a website called the True Size Off, and you can actually plot different land features on that website. And if you go and grab, let's say, let's just a classical example, like, if I grab the Greenland on that particular map, a map is in web market or projection, and Greenland and Africa will roughly look the same size. And if you grab the Greenland and you drag it down, it's not the size is only changing, but also the shape of the Greenland is adjusting, and it's different. So therefore, in general, in conformal maps, the shapes are still distorted, not as much as on the equal area projections, especially on the edges, but they are still distorted. So we cannot preserve shapes in general, but in the conformal projections, we preserve local angles. And when we're using conformal projections, they are appropriate for surveying, military, naval navigation, etc., and topographic mapping is another example where the conformal properties used. And those two properties, conformality and the equal area property are kind of like, I kind of explained that usually with the spectrum of all map projections. And if you take the spectrum, like, let's say you have a bar of map projections that are all map projections represented there, equal area and conformal are going to be on the opposite sides of that bar, because they cannot be combined this in time. And the reason for that is because they do that for the entire map. So entire map preserve local angles or entire map preserves areas, but you cannot mix them. But then you have other properties, like you mentioned distances, for example, where on the maps, distances are not preserved for the entirety. So they're not preserved for the entire map. So there's no map projection. As we have no map projections, preserved shapes, we have no map projection that preserves all distances on the map. But they can be preserved certain distances, like distances from the center of the map, the classical example for that is the asimuthal equidistant projection, or you can preserve it along certain lines, like a lot. the meridians or along the parallels. And because that property is kind of limited to certain distances, it can be mixed with other properties as well. So sometimes sometimes we can do more than just one particular property, which makes that. And the example for that, for me, it's like great examples, it's like a sinusoidal projection, which is technically equal area projection, but it also preserves distances along the parallels. Okay, so I think you've explained this, but I just realized that I have a brilliant quote. I love it when people give you a quote that is like and this is, you said, it's simple to apply a map projection, but it's tough to select the right one. And I'm guessing that's because you have to know what map you want to make and what you want to show on that map, the purpose of the map before you choose the projection. Yes, definitely. Like, you need to know every, like what is the purpose of the map, but this quote actually comes from something else, like it's not just the purpose of the map. So in the past, when the maps were still make manually, you probably remember, or some of the numbers. No, no, that's okay. I know people probably remember that. And I remember, I remember it from the times when I was in school, because that's what we were taught by professors and they were teaching us like how plotting the data is actually the process itself, right? And when that was done like this, it was pretty much like manually done applying a projection and plotting the data on a map was quite the complicated process. It was not so straightforward like it is today. And it took cartographers' significant amount of time to do all this process there. And because it was so complicated, it was also like it demanded from cartographers to properly understand the projection mathematics and the projections themselves. So like what are actually, what they are actually doing when they're plotting the data? And they had to know that and they had to sing very carefully, which map projection they were going to select. So they did not have to start over. Like in case they made a mistake, right at the beginning of the process and basically starting the entire mapping process completely from this scratch, right? But today we have a mapping software, which basically does the plotting of the data on a map just for us. And it basically requires just nothing but the few mouse clicks. And voila, you have a new map projection on the map. So it's super, super easy. And the result of that advancement in the cartography is basically that we also no longer require to know much about map projections per se, like in the deep mathematics, we don't have to know that anymore. So it became, because everything became also so easy, we kind of lose the knowledge of like how to correctly select a map projection for a map. And basically map projection selection has become one of the most challenging part in the map making today. So especially like if we look back like in the past two decades, we become comfortable of using one map projection for absolutely everything that we do, especially in the GIS for storage and sharing the data, performing spatial analysis, making maps, presenting, doing demos, like all the stuff we do in one projection. But the harsh reality is that basically most appropriate map projection, it really depends on what we're trying to do with our geospatial data. So this one size fits all approach that everybody likes so much today. It's actually wrong in many, many cases. For pretty much every map for every spatial analysis we're doing, for data storing and sharing, we always basically have to ask ourselves like what is the best map projection? Or in terms of GIS, what is the best coordinate system that I should use in this process? Like what is the best way to approach that? And like usually when I do the guest lecture for universities and I teach students about map projections, I tell the students that map projection selection is basically the key information that they have to take out of the class or the lecture we do. And basically it's a key information for every single cartographer. And if somebody wakes them up in the middle of the night and asks them which map projection they should use, this is the answer that they should actually know immediately. Not details, of course, they're not going to say like oh, these parameters and projection parameters and so on, but generically. So like oh yeah, I'm mapping this particular area so I should use this particular projection or this kind of projection, this kind of class or aspect or or that comes with that. And basically knowing and understanding what is appropriate where basically can help them master their cartographic skills, but what is most important, it basically makes them look like a true, due special professionals when they are actually entering the field of cartography and start doing the work in cartography. I realize that through most of my the time that I was working in the industry, we were working in the UK. And so all the maps that we generated were in OSGB, British National Grid. And on the on occasion that you opened up your GIS tools and somehow got data that was in web locator or something. And you didn't correct, all of a sudden you realized how big a difference there was between these two projections. It wasn't sort of the level of the UK, which is a relatively small country. In fact, it is a small country, you know. I mean, it was a massive difference, you know. But for us, we never probably, if I had a team of people and we were making hundreds of maps a month for different purposes with customers, they'd all, you know, they wouldn't even think about projection because everything was done in that one projection. In one projection. Yes. And it really depends like if you what you do, right. So it's not the same all the time. And yes, most cartographers and when you're working the industry in a certain country, right, you're using one or maybe two or maybe a few of them that you are dealing with, right. And then you can see comfortably, like I'm a Slovenian. So in Slovenia, we have one coordinate system or projected coordinate system for an entire country. We don't even have to think about the projections when you're working with this kind of things, right. Like you don't even think because like your data comes in the projection that you have, it plus it is appropriate projections. Therefore, you don't even have to address it. So data comes in pretty much everything is synchronized into one system. But as soon as you step out and let's say you're trying to show the Slovenia on the European stage or Slovenia in the world stage, then you have different different questions and then you have to start thinking like, okay, how I'm going to consolidate this data, how I'm going to show this data correctly and stuff like that, right. But yeah, in general, like if you're in small country, one coordinate system, I think in that particular case, yes, one size fits all approach works because you are not going out. But but it doesn't mean that like, you know, we have about Mercator projection today present absolutely everywhere that this is actually the projection that you should actually replace that with the with the national grid systems that are available. And I learned that in America, in North America, there are almost a projection for every state, aren't there? It's worse than that. So in the United States, like, okay, so if we compare Great Britain, Great Britain has one predefined projection called, oh, sorry, sorry, I lost the name. National grid system. National grid. National grid, right. Ordnance National grid system. And that's because this, like in the United States, because the country itself is so big and has like across multiple tectonic plates, even, because we have a Hawaii that is on the Pacific plate. Every single state has different zone systems. And basically today, as of today, we have 150 different projections that they can actually, the projected coordinate systems or projections that they can use. And they do not have officially defined projections for one state or entire contagious United States. We use the, there are a few of them that are kind of like known to be used, but not officially. Now, next year, there's a twist. Next year, they're moving towards the new system, which is going to, it's basically called NSRS 2022 modernization. And they're going to switch from 150 so to 900 and 990 something like 50, I think it's 52, 952 definitions. So for the entire entirety of the United States, they're going to have 950 something definitions. But besides that, there are two major changes that they're doing. And one is that they're going to have definitions per state. So they're going to have not just the different zones per state, but they're also going to have a definition per state for the entire state, which is definitely an improvement. And the second reason why there are so many of them, they are moving towards the concept of so-called low distortion projections, where projection is no longer designed based on the earth model, so like ellipses model, but instead the projection definition is basically adjusted to the actual surface of the map. So they're actually taking the heights, the elevation heights into the consideration when they're doing that, which is something that will basically help the surveyors on the ground. So the grit and ground correction is going to be much smaller or actually in most cases not even needed. So yeah, but they are going from 150, so to 950, so which is like plus they have different units, right? So they, so we have to duplicate the number. Wow. So let's talk about your projection. Well, it's not just yours because you've authored it in partnership with a couple of other people, but the Equal Earth projection. What is that and what's special about it? So the deal of the Equal Earth projection was, let's say born in 2017 when Boston Public Schools announced that they're going to switch their world maps into the Peter's projection for school purposes, right, for the schools. And we know from the books and history that like the Peter's projection is always a controversial topic, especially in the world of map projection experts, because we don't really like this particular projection. So back in the fall of 2017, Tom Patterson, at that time, he was still working at the US National Park Services. He reached out to me and my former professor Bernhard Tieni today at the Monash University in Australia with the basically proposal to like, okay, let's take a look at this problem and let's try to find a projection that would be much more appropriate for school purposes, like for the schools and for the world maps in this particular case. And in his professional opinion, he's a great designer and great photographer. Tom really did not find like a projection, both projection that would kind of fit his kind of eye, his graphical eye, so to speak. And he proposed that we basically come up with a new map projection that would kind of be more appropriate, basically correctly for the world maps correctly used. So yeah, we basically went into the works. Bernhard Tieni created the tool that uses the 3B morphing methods, so to speak. And Tom basically designed that using the tool to kind of come up with a little bit more aesthetically pleasing look of the world map. And it's also kind of tried to make a shape of the world and the land continent as close as possible to the Robinson projection. But with one big distinction, Robinson projection is a compromised projection, which means that the areas and basically balance that balances the distortion in general of the entire world. But with the equal earth projection, the deal was to have the equal area property of the map so that we can still preserve the areas. But with his good design capabilities or how should I say that? With basically his design eye, so to speak, he basically tried to make the shape of the land masses look more correct, more nicer, especially better than what we see in the Peter's projection. Because in the Peter's projection, you've got area is more accurately represented. But the map is stretched and particularly in the southern hemisphere, you've got these long thin consonants, is that right? Yes and no. So yes and no. It's like with the equal earth projection and the Peter's projection on both maps, the areas are correct. There's no difference in that in that sense, right? But the shape of the continents on the Peter's projections are kind of like drastically stretched along the equator and super compressed on the poles. And that is because of the shape of the world, because the shape in the Peter's projection is technically a cylindrical projection, basically the entire world is in the square. But on the equal earth projection, we had this rounded shape of the earth or the rounded shape of the entire world. And we kind of, basically adjust, there's adjustment for that. So we do get the different impression. And you basically plotting the land features a little bit differently than you do in the Peter's projection, right? And so yeah, and basically that was the idea to kind of come up with the projection that has a little nice features. Land features that are presented semi accurately, semi correctly, so to speak, more correctly, not completely correctly, but also preserve the areas, right? So the long map itself looks nice. And in advantage of that, which is the most important one, that we represent the earth in the rounded shape and not in the square, right? So that we don't have to get the impression of flat earth, but we get the impression of the rounded body. So talking about shape, you made me think, can I? We have, you said before, I think that none of the projections accurately present shape. But if you're as old as me and you've been looking at atlacies since you're a little boy, you've got a sense of shape that's burnt into your brain, right? When I look at the Peter's projection, it looks wrong. There's no, you know, I know that you can have different projections and I understand all of that intellectually. But when I look at that map, it looks wrong. And we all have a sense of the shape of the continents and like a mental map of the continents. Yes, exactly. I mean, and you can take one of those shapes, you know, just draw an outline and people will say that's Africa, that's Australia, right? They don't need any reference at all, you know, any other information apart from the outline to recognize those shapes, do they? And that's almost universal, I would think. Pretty much yes, because we are using the same pretty much similar maps or all the continents, more or less, are represented correctly, well, not correctly, but like the same way, right? Roughly we can say that the shapes are similar to each other. They're not the same, but they are similar to each other. So we can recognize the shapes, but it's very easy to like if on the world maps, if you are going away from the center of the projection or away from the origin of the projection, like the shapes are starting getting more and more and more distorted and it's cool, and they look differently on the maps, right? And we can get like if you put those maps, like if you put those shapes together and you start looking at them, they're kind of like you, some, if you're using the shape that is away, you kind of have to kind of take a look a little bit more and say like, oh yeah, this kind of looks like that, but this not quite there. So let's get to what prompt did me to get in touch with you, Boyan, and it's taken us a few months, which is probably a good thing because back in August, the African Union started a programmer campaign called Correct the Map, and I hate that phrase Correct the Map, and we can talk about that maybe, to persuade the United Nations and the World Bank to use the Equal Earth projection. And if I'd spoken to you in August, I think I would have been really critical of that and I would have called it, I would have said they didn't understand projections and I would have said that it was work bullshit or something like that, and I would have gone on. But five months later or four months later we're talking and I've had the opportunity to read more and to think more about it and realize that actually there is a story and there is some sense here and the fact that we're used to seeing in atlases that were published from the 1930s until 2000, most of the atlases of the world, or the world maps in those atlases were in a Mercator projection and we recognize it doesn't mean that it is the best projection that we should be using for those maps and so you were in the eye of the storm a little bit when that all took off, weren't you? Yeah, a little bit, not just me, but we're all three, it's all three of us, like also Tom and Bernie. Yeah, so what were they explain what they were talking about and why they were making this call for the equal earth projection to be used? So I think the campaign itself started earlier than August, I think it started in April and in August the African Union endorsed it and that basically put the entire campaign on the world stage. So everybody started talking about it. I think to my understanding the idea of the campaign is not to change the entire field of map projections and just use equal earth for absolutely everything. The idea here is to properly present the African, the world stage relative to the rest of the world. So like when you were trying to create the world maps, especially static world maps, especially for schools, for medias, for other outlets out there that are in massive use, so to speak. The idea is to kind of stop using the marketer or web marketer maps for these kind of purposes and start to using something else where Africa will relatively be correctly represented on the world stage and on the world maps. So they're not, I do not think that they are campaigning on the for every single map of Africa. So like even continental maps, which are just showing Africa, they are just campaigning for the fact that they want to make sure that Africa looks correctly on the world stage. And just to be, just to add another twist to that, right? On marketer projection, Africa actually looks relatively, relatively correct. But the problem is that everything else around the Africa is largely distorted and basically because further away from the quater you are, the land masses are becoming bigger and bigger and bigger. And basically gives them pressure that Africa is actually smaller compared to relatively of the world. And I think that's the direction they're trying to get. So basically correct the view of the Africa on the world stage with world maps. So I just remember that back in 2018, which is before the color of projection was released, isn't it? We did release in 2018. Right. But before we, it wasn't in common use, then, I don't think. So Kenfield and I did a, I don't know whether it was a keynote or a big talk at phosphogen, dara salam. And we were talking about the UN's development goals. And we were talking about, it was a combination of the things that people didn't know about sort of world geographical statistics and the assumptions that they made about sort of economic and social and cultural statistics of the world. And so we produced a series of maps and it was also an opportunity to have a little bit of a workshop to style talk about mapping sustainable development goals on a global basis. And we did that. And I'm pretty certain that if I go back and look at the maps, they were all in web locator or locator. And they shouldn't have been right. And I can see that now. And certainly for thematic mapping, which is what we were doing. Basically, we were coloring in country borders to represent the intensity, the number of goals in secondary education, the number of people who had access to water, those sorts of things. And of course, a lot of the focus in that whole thing when we were in Africa and a lot of the focus was in Africa. And we were certainly I was. I'm not, you know, I'm not going to speak for Ken because he's a great cartographer and he may say, oh, you're an idiot, you didn't know what we were doing. But I wasn't aware of these things. And I think a lot of us are unaware of these things and a projection that provides equal area so that all countries are shown relative to each other as the correct size. It's a great idea. Yeah. And it shows how tiny the UK is. Oh, yeah, absolutely. And basically shows the relatives, but equal earth is not a projection that is the first equal area projection we have. Like even before, right, even before the equal earth, we had a equal area projections. Actually, in the fact we have many of them that are equal area projections. We have many of them that are around the shape. So they're pseudo cylindrical projections like the equal earth projection. And none of them stick so to speak none of them like got the attention from the cartographic field only, right, not speaking on the world stage, but the graphic fields like none of them got the big attention to it that would be basically largely used or in use or something on that. I think the only projection that I can say that like really got the more used on the world stage is the Robinson projection that the, but it is not the equal area one, right. But on the size in terms of equal area projections, like none of them that but equal earth back in 2018, we went through the same process we did this year. So it was also like as soon as we published the projection, basically the whole thing become nuts and the basically I'm saying like it went viral at that time. Everybody was talking about equal earth projection for months and sorry equal area equal earth projection for for an entire month and we went through that process there. So this was this year, it was like the second time that's basically we got the attention and I think that like the whole idea here of the while the whole idea of the equal area projection is world projection is not new in sense. The name itself, because we're saying equal earth projection actually that brought the projection equal area world projection into the into the big stage. And I think that was the key decision that Tom suggested that basically brought that up into the into the. Whether you're using Esri's arc products or you're using QG's it doesn't matter you start making a map and one of the first things you do is you have to choose the projection. It's very near the beginning if not the beginning and you click on a little down arrow and you have a list that goes on and on and on. Yeah, and you've got I don't know how many projections there are in all just hundreds of them though on there, you know, maybe even a thousand of them. I can give you exact numbers go and go and tell me so when we talk about software right we have to distinguish between two concepts and one concept is equal. So it's a projection and the other concept is projected coordinate systems right and the projected coordinate systems can use the same projection. So we have to kind of distinguish between the two so if you open arc pro for example or arc map or art map is not anymore. But you open that you see the projected coordinate systems and currently with the latest release we have only 6300 definitions to use. Wow 63 hundred and I'm basically simplifying and we're talking just projected right we have another 600 geographic coordinate systems and so on and 450 I think it's a vertical coordinate systems. So we're talking we're just talking about projections is 6300 right about the projections when if you want to really look at projection themselves when you are looking to the projection algorithms that we have. We are down to over a hundred different projections. That also includes the variations of it, because for example, Transverse Mercator is also called Gauss Kruger. There's also Transverse Mercator complex. There's Transverse Complex NGA definition as well, Transverse Mercator. For any kind of particular projections, we can also have different variations of it, which basically just means that mathematics is a little bit different and doesn't produce the same results. But we have more than hundreds of them. And that's just a small, really small part of all the possible projections. There's a call map-desh projections that I think is a dot-math, where it's like, to be as young, basically, it's a enthusiasm of the map projections, and he basically collects, like, there's a large library of hundreds and hundreds of projections that he was able to find in the literature. And there you can actually see how much the amount of the projections that we can have. And if you're a map maker choosing from all of those, the moment you call it equal Earth, for a certain use case, it becomes obvious what that projection is going to be useful for. I mean, I think when we talk about Peters and Gauss and Gaul and all of these, they're named after the people who originated the projections. But unless you're a projections expert, that doesn't help you in working out what you should be using. So, we haven't really talked about Gerhard as Makator, and we can't talk about projections, because the projection is constantly criticised for things that those of us who are map makers think, what are you talking about? You know, why are you criticising it for this? What's wrong with Makator? I don't think nothing is wrong with Makator. Right. Why don't people like it? Exactly. That's the question. So, every map projection, equal Earth, Makator projection, they have their own purposes, and they have their own properties. And the same way that equal Earth projection is appropriate for world maps, and the thematic maps, Makator projection has its own purpose. So, and because Makator projection is a conformal, and because it shows the length of compass bearings as straight lines, projection is fantastic for scene navigation, for mapping, basically using it for to navigate over the sea, and that's what the projection was used originally when people were going from Europe, going basically travelling to America and discovered America. And that's why the projection is, even today, still the requirement for the navigational charts, and basically is actively used for that purpose. Navigational scene navigation and navigational charts, right? Not for, in general, for navigating. And today, we don't need it for navigating per se, so to speak, the projection itself, because like we have GPS, GPS can be plot points, can be plotted on any projection in any time. It's super easy, super fast. The criticism of the Mercato projection comes with from its misuse, because projection itself is not used for the purposes that it's supposed to be used for. It is used for all other purposes, like you were talking, back back in 2018, when you were at the UN, you used the Mercato projection for the purpose that the projection should not be used. And that is the reason why the projection is criticized, especially on the world stage. When we are showing the entire world, that projection should not be used in that particular purpose. But also like today's world, like Mark Mimir said very nicely in his book, that misuse of Mercato projection used to amuse cartographers and geographers, because it was kind of like, you did the basic mistake here, you should not be using it, right? But today, with all the web mapping we have, and all the images that are there, and everything that is, let's say in the media, and beyond the media, the internet, right? Everything we see is Mercato map. And very rarely we see alternative projections, or projections that are appropriate for the world stage, or for any other purposes. And I think that's the, well, I don't think, but I believe this is the reason, basically, that this is the reason for the criticism here, of the projection itself, that misuse and not, that it's not bad, that it's bad projections, so to speak, but misuse is the problem, because it is so largely used and misused. But in defense of web locator, I mean, I was going to say 90%, but it might be even more than 90% of the time. We are looking at small areas. You know, we're looking at city level. We may be looking at state level, but, you know, most of the time we're looking at a level where the distortions are almost invisible, and where the ease with which it can be rendered makes it very, very attractive. I mean, we don't use web maps. I mean, if we talk about Google maps for a Google map or an open street map, we're not using them very often at a world or even a continental level. We're using them at a sub-country level and, very possibly, a sub-city level, at which point I doubt that the distortions matter at all. They do. Okay, go on, correct me. I walked into that. Yes, yes, you did. This is an example I use for students as well. For example, if you have a web map, two web maps left and right, you use the exactly the same scale, so zoom level. So you zoom it exactly the same to the same level. And on the left map, you have Oslo, capital of Norway, and on the right side, you have Singapore. And if you look at those two images, let's say that you have one by one, Oslo on that will appear the largest city, largest than Singapore. And Singapore is going to look smaller. But in reality, it is actually opposite. Singapore is much significantly bigger than Oslo itself. So it really comes down to like how do you use it at that point. And you also have to correct the computations even though you're using the small area, like let's say Oslo or UK or London, maybe getting through the London, you even have to correct for the projection distortion because it does and largest the whole area and the whole distances and everything at that even at that level. So if you're just kind of zooming to market or projection, you're putting the, you just don't take the distortion into account and you just plot the scale on the map, it's still going to be wrong. But it's still not going to be correct. So you have to correct the scale of the scale of the distances and everything like that on that particular map. But even though like, for example, if you're, what is another good example I use is like Switzerland. If I use web marketer and I basically create the hexagon bins across the entire Switzerland. And then I go in a calculate that area there. Let's say that you're using it for spatial analysis or something like that. And you go and you calculate the areas of those hex bins that are generated in web marketer even though it's a small scale. You create the areas like it's going to be like roughly about 50% wrong. You're making half of the mistake on the area calculations in Switzerland if you're not taking the projection distortion into account. And today when you're just clicking through GIS and doing something, especially spatial analysis, where areas are a huge part of the computations. Like you still have this mistake. Even though you zoom in, you did not correct projection distortion. You still have that present. And you have to correct it manually if you wanted to do it correctly. If you're using the same time using the marketer projection. So even in small scales, sorry, large scales, it is important that you take the projection distortion into account. And if you're using web marketer for that, it's pretty much the same. same thing. You still have the mistake. So almost even at very, very large scale, if area is important, you need to be using an equal area projection projection. Yes. Or it is okay that you go with the official projected coordinate system that they have, right, in this particular area, like state plane or, uh, or ordinary map and national grade and stuff like that. So because those projections are kind of like, uh, designed to be true, so to speak. The angles that through areas are not quite true, but they are close enough that we can actually disregard, uh, on the analysis that we're doing. But I would agree that like, if you zoom in and you're just looking a map, right, just the plane map, uh, it's very hard to kind of have a problem with it. Uh, but today we're zooming into the map to the small area. Um, and we, we're not using it just to display. We're using it for a lot of other purposes, uh, including distance, uh, area computations and stuff like that. And there the market of projection still, uh, matters because the projection selection basically still matters in that purposes. So you've solved all the projection problems of the world with the projection. Um, I thought you'd say that. What do you want? Maybe I'll get the world piece, uh, noble piece price for that. Maybe you will. Um, don't, don't hold your breath by it. Um, so what do you, what are you working on now? What, um, I, as working on it. I mean, uh, yeah, go. I was going to say, haven't you solved all the projection problems? And no, I haven't solved all the projection problems. I still have a problem. So I still have to work on projections. Okay. So I work, uh, uh, basically I work for us and projection team. So, uh, we are still, uh, actively developing the software, um, adding new functionalities and capabilities. Um, besides the regular adding more stuff into the software, uh, I'm kind of right now exploring two, um, two areas. So to speak. One is, um, uh, low distortion projections. Um, I'm actually playing a little bit, uh, uh, with examples and try to see like how this low distortion projection method is actually working and, uh, how you can actually create a little bit more, um, let's say accurate, uh, two projections for topographic mapping so that you don't have so much differences between the grid and the ground. Um, that was just part of my curiosity, uh, pretty much everything is already developed, uh, and researched, uh, right now I'm just kind of like looking into it and see how this, um, behaves. Uh, and the second part is like, uh, not just selecting a map projection, but actually selecting an existing predefined projected coordinate system. So I'm looking into the possible filtering algorithm, uh, to kind of like pay, I'm making this particular area of, uh, here. Can you find me, uh, the most current projected coordinate system, uh, that I can actually use for that and I'm kind of exploring the samples abilities of like how to do that when we have a predefined definitions into the, uh, in, in place. Right. So that's kind of like the current, current, uh, so sort of smart projection selection. Okay. So I've projected coordinate system selection AI without the AI. Right. But based on, I want to do this. What's the projected coordinate system for this area that would best? Yeah. Yeah. Yeah. Because this is the only thing you can in the software development. This is the only thing that you can actually queue on, uh, the area of interest, um, and not so much about the purpose because purpose can be mixed and we don't have like a solid rule of like one to one based on the projection properties and, uh, purpose of the map because purpose purpose can be like very different. Also, it depends on like, you know, photography aesthetic choice, like what do they like? Like one is theory, but then they also like it is a practice, like what is in the practice looks good and what doesn't and this cannot be controlled by, um, by the software, um, itself. So yeah. What looks good is very important, isn't it? Uh, it's an important part. Yes. Because you get a lot of resistance, you know, which is why gopeters never really got massive uptake, I don't think. Um, Boston school, yeah, probably like, uh, Peter's projection probably did not really, um, keep basically got because of the way how they changed the world to you in that particular case. And it did not follow the basic rules of the, like my projection selection and that is probably the reason why the resistance on the photography side was, uh, so basically the Peter's projection. So, when I was just thinking about the Peters, we've had people asking for Peters projection. Um, there's certainly, I've seen people suggesting that we should have, um, different meridians, you know, the map centered on different meridians, you know, I think if you're in North Korea, for example, the world is centered on North Korea, which is fine, but looks a bit strange to most of us. Um, and we've had people talking about south up rather than north up. Um, and we've also had, in biblical times, we had east up, um, a lot of the, you know, the maps from the sort of, from 150 in the common era to about 1,1200 were east up when they were mapping the, um, the biblical lands. Um, what's the strangest request that you've had as a projections expert that someone's come to you and said, could you do this or why don't you do this? Strangest, wow. Uh, ooh. I think the most strangest thing is, was that like, I want to have a projection that preserves all the distances. I think that's the, that's the most. Right. Strange thing that comes in just out of the, not understanding uh, uh, requires like, I wonder what that, um, I think it's like, uh, often like commonly we have this, um, uh, because of the vocabulary, uh, we have a so called equidistant projections, which basically kind of gives the impression that like all the distances are preserved on the, on the map and all the distances are correct. Um, and then the distances, it's the day they, they, they expect that it's basically possible. Like, okay, you can preserve all local angles, you can preserve all local, uh, air like, oh, sorry, all areas, right? Why not to preserve all the distances on the map? Um, but mathematically that is, uh, impossible, but they got the impression that like, oh, yeah, I'm just going to use equidistant projection and measure the distance in some weird strange way and why this, this distance is different. Uh, I think that, that probably is one of the weirdest things, not to weird this, but more like coming out of the lack of understanding, um, stuff like that. But today changing the view like you were describing with, uh, central meridians, north op, east op, and stuff like that, it is not difficult. So to speak, especially central meridian, right? Um, and if you want to create a projection that is like, uh, north op, right? Sorry, east op, uh, no, sorry, south op. It is basically just a question of like rotating the entire map, um, and just grabbing the cursor and turn around the map around it, and you will have a north op, uh, sorry, south op, south op, uh, south op, um, east op is a little bit more challenging because you have to manually rotate the entire globe, um, to different view. So to speak, uh, in order to kind of minimize for the distortion so that the things doesn't look weird, um, basically creating a transverse case, uh, and certain projections actually can do that, uh, is certain projections, like in certain software, we can actually do that, no problems, and project the software is a little bit restricted to that. Um, yeah. So, um, outside the London School of Economics in London, there's a giant globe, um, I don't know how high it is, 20 feet, maybe 30 feet in height and diameter, a globe, just sitting on our concrete plinth, except that it's south up, super great, and it's got all of the countries of the world on it, you know, and you can stand under it looking up for it and you can walk around it and it is absolutely I went there with a group of map lovers. We happened to be walking somewhere together and we stopped and we started staring at this map. And Denise McKenzie, who's an Australian, was with us. And her view of this map was so different to those of us who grew up in England and who were so used to seeing the North Up map. And all of a sudden you discover the most amazing things about the world when you look at it from a different perspective, like that. - Perspective, yeah, yeah. - And it was really amazing. So I think the idea of occasionally, certainly flipping the map and having South Up is a really good thing for sort of just changing your perspective on how the world hangs together. And also seeing how unpopulated some areas of the globe, you know, there's no land. - How different they look, right? - Yeah. - How different they are and changing the perspective. It's also like you can also change your decision. I remember the times where I was at Oregon State with Bernard Dianne, we did the adaptive composite map projections and we created that video map of the CO2 gas over the year and how it swirls around it. And you NASA produced that video with the geographic projection. You could see certain patterns. But when you put it on the adaptive composite map projections and you basically wrote data that you could completely see completely different perspective of the data itself, like how the data was like actually, the data was actually swirling around the poles, like how the winds are actually going around the poles. They're actually not going crossing directly through the poles, like through the poles. You could see the pattern, like that actually, the wind has the rotational axis, how the rotational axis actually infects the wind of the earth as well at the same time, which was like completely changing the perspective of the data and stuff like that. And probably a lot of times in today's world, we need to change our perspective. Like because we're what we're used to, we're used to looking the world one way, either being cryptographically looking or looking it in other thematics. But a lot of times we need the different perspectives just to get the new ideas in and get the new inspiration or something like that from that. And just not seeing it the way that you've seen it for the last 30 years, and just having to think again, rather than automatically recognizing the shapes and everything, brings different-- - From the beginning, from the beginning, right? And just kind of use different perspectives. Like how many times I have a problem when we're doing something mathematical and I'm stuck and I cannot do anything, I cannot find a solution or I cannot do something. And then I just walk in and I start making myself dinner and I change the perspective itself. And suddenly I got like this crazy ideas into my hat that I would never get them like half an hour ago. Just because like changing the scenery, changing the perspective of the whole thing can actually inspire you to get more ideas. - And that's probably why we need 6,300 coordinate projections. - Free, free, I have a projected coordinate system. - Projecting. - The United States, where we're going to cross probably 2,000 very soon. - Okay, boy, it's been an absolute pleasure talking to you. I've learned a lot. I've displayed my ignorance. And also, you've made, even though I thought I understood why the Equal Earth Projection was important. You've made me really recognize it. So thank you very much. Thank you for everything. - Thank you very much, John. It's been great having you. - Yeah, thank you very much for having me. I'm happy to talk about the projections all the time. It's my time. - I know. - Time. Take care, bye. - Thank you. Thank you. - Thanks for joining us today on the GLMAR podcast. We hope you enjoyed the discussion. You can get the show notes over on the website, which is dot GLMARB.com. And while you're there, you can sign up for our monthly newsletter and find links to our accounts on Mastodon and LinkedIn. So you can stay up to date on all our upcoming events. We hope to see you at an event soon. (upbeat music)