WEBVTT 1 00:00:06.000 --> 00:00:16.000 Welcome to the GeoMob podcast where we discuss geo innovation in any and all forms, whether for fun or profit. 2 00:00:16.000 --> 00:01:56.000 Hey, welcome Back to another GeoMob podcast. It's sweltering in London and I'm moaning about the heat, but my guest is phone is in Delhi at the moment and I'm guessing that if we think it's hot in the 30s here, he's laughing at me because that's not hot for him. Anyway, my guest today is Akash Wadhwani. Akash reached out to me completely cold a couple of weeks ago to share a link with me which we'll be talking about. And we got into a conversation and I thought, wow, I've got to get this guy onto the podcast. And the reason I wanted to get him on the podcast is because he's a data visualization specialist. He did an MA at the University of Arts in London and he now makes interactive data stories for all sorts of clients. And his maps have been featured in the New York Times and the Guardian and stuff like that, his work. So I mean he works with data all the time, but he's not a geo geek, he's not one of these sort of hardcore map makers and yet he started making some really fascinating maps and that's what we're going to talk about today. So Akash, welcome to the GeoMob podcast. It's a pleasure to have you here. Just start off by introducing yourself and telling our listeners what you do in your day job and how you got interested in maps and mapping. 3 00:01:56.000 --> 00:06:44.000 First of all, thanks a lot Stephen for replying to me. That was a cold email with few lines and a link and as you rightly mentioned, it can be fishing as well. But you took the leap of faith with my lint and thanks a lot for that. Myself, my name is Akash. I am the founder of Sheets Words. We made data drops, we do weekly data drops of data stories where I tell a story which I want to convey to the world with data. And the story can be anything which revolves around pop culture, sports or maps or which, whatever. Like I personally feel it's more like the creative outlook for me and that's how I do it professionally. I, I worked as a consultant for a few clients in the UK and UAE where I help them make their dashboards and their yearly reports. So that's what I do professionally. And regarding the map which we are discussing, it's called London drone and it's 278 large forte images stitched together making over 38, 000 pixel squares of that whole Zone 1 to Zone 4 of London. I have been fascinated by London because I lived there for a couple of years. I used to study, as Stephen mentioned in UAL at Elephant and Castle. Then I used to work at Rolls Royce. And I was there in the UK for a couple of years. And then I left that job, came back to India. And London is. Has always been really close to my heart. I used to live near Holborn and I just wanted to do something with London for a very long time. And then I realized that today with AI and with all the technology out there, you can do so much in field of. In any field. And this one I choose to be maps. And I am very fascinated about paintings. I did a piece called how to Read an Art Painting. Like for example, Mona Lisa and other paintings like the famous paintings. There is a. There's a whole method of reading those paintings which obviously I did not know before doing that piece. And that's where I got the inspiration. What if London was painted like a painting and you could read it? That was the whole idea behind it. And I just wanted to paint the whole London in a certain way where it looked like a painting and we could read it. So that's exactly what I did. It took me three weeks exactly to complete it. Two to three weeks to complete the whole project. Where I started with taking satellite photos of London from Booth Earth. I started with the tough bit of the Tower Bridge area, where the shard is and where the. The, as you call it, the Wati Tohi building. So that area I covered first because that was the tough bit. Took me three, four days to finalize the painting style which I wanted. And then once I finalized that style, after that, I started taking pictures of the whole London, like zone one to zone four. And from there I started stitching the pictures together like large pictures converted by AI and then you touch them into smaller pieces so that you have every postcode and every area with latitude and all longitude perfectly aligned, as you can see. Go to my website and you can search a postcode as well. Right. It works like the proper map as well. So that also took a bit of understanding. And once I was done with that, I wanted to make a tour of London. Like my friends who have never been to London, I wanted to take them with me. Like my wife has not been to London with me. And I wanted to take that alone. So that's why I made the buildings like places I used to go. St. Paul's Cathedral, like Big Ben and like Tower Bridge. And the shard and stuff like. So I made those buildings as well. And then you can walk through my map and visit those buildings and learn about those buildings. How many people are searching about those buildings, their history? What will you find inside those buildings? So that. How you. That's how completed the loop in my head. 4 00:06:44.000 --> 00:06:51.000 So let's just. I mean, it's difficult, you know, because we're talking about something that's so visual. 5 00:06:51.000 --> 00:06:52.000 Yeah. 6 00:06:52.000 --> 00:08:19.000 And it's. You know, you look at really good aerial imagery and this is like another level up from it because it's all been. It's all been effectively redrawn from the imagery to have this sort of. And it's sort of that photorealistic style of drawing. It's. It's quite amazing. It's an amazing feat. And. But when you click on how many buildings have you got that you can. 183. Right. So I'm just clicking just because it's where I found myself on Somerset House. Right. And it's got this beautiful rendering of Somerset House, which is. Is not photographic, but it's very close to photorealistic. Is lovely. And then when you scroll down, you've got a building by William Changers in Georgian architecture. You've got a story about it, you've got how many visitors there are. You've got what it cost. I'm guessing at the time that it was built, not today, because it would be millions, billions of pounds. You've got some pictures from the inside and you've got a load of other information. Where's all that come from? 7 00:08:19.000 --> 00:08:36.000 So most of the information which is coming there is from Wikipedia. And then Wikipedia has this API where you can find the visitors of that building on Wikipedia. So those visitors are the people who are actually learning about the building on Wikipedia. 8 00:08:36.000 --> 00:08:37.000 Right? 9 00:08:37.000 --> 00:08:38.000 Yeah. 10 00:08:37.000 --> 00:08:38.000 Okay. 11 00:08:37.000 --> 00:08:55.000 So this is 70 Wikipedia and 30%, like for Somerset House, it was from Somerset Houses own website and other articles, what I've mentioned. Somerset House, not very old and not a lot of them because I had a lot of buildings to do. 12 00:08:55.000 --> 00:08:56.000 So. 13 00:08:55.000 --> 00:09:09.000 So I did not go very deep with every building, but enough for you to know that you might want to visit and go deep dive more into it. If you are into Somerset House, for example. 14 00:09:09.000 --> 00:09:13.000 So you didn't do this just with AI, you actually did this. 15 00:09:13.000 --> 00:09:54.000 Yeah. So not. So basically the drawn bit was AI. The coding was mine. So I code myself and then stitching and the themes you see on the map, that was all photoshopped by me because no matter how hard I tried, the AI was not getting the same color of water in every photo. So I had to sit down and patch each and every of that photo in Photoshop and made sure that every piece of themes looked exactly the same. You might find a few places, places there it's not. But most of the places now it will look a simple themes flowing in the middle. 16 00:09:54.000 --> 00:11:08.000 It's a magnificent work. I've got one complaint. I live in North London, just outside of the area that you've covered. So I'm hoping that at some time in the future Akesh are going to spread it out and capture the Emirates Stadium and then go a little bit further north to where I live. But I mean it is just fantastic. And I think it also shows to me a combination of this was hard work. This was yes, you had AI doing some of it and you've obviously got AI to generate a lot of the painted images for you, but you still had to do an enormous amount of work to it. And if you've got 180 buildings in here, gathering all that data is not just a case of pointing the AI and saying, fill this table for me. You've got to actually go and do the research and craft it all together. So well done to you because it is a fantastic thing. What would happen if you wanted to do another city? 17 00:11:08.000 --> 00:12:04.000 Oh, it will take another two weeks or three weeks maybe if I have to do another city. But I think this will be faster now because now I have a playbook ready and my idea for like this was the next step is what I wanted to do like, while, while we are discussing this, which I wanted to discuss with you outside the podcast. Anyway, is that what I have learned in London, especially if the city is X amount on the top of the ground, the city is at some bottom of the ground as well. London is very big underground. Like they have huge stations, some are very deep, some. So I just want to build other underground as well for this. That is my next step before any new city. I want to complete London above ground and underground as well. Because it is huge underground and it's very big. 18 00:12:04.000 --> 00:12:07.000 And can you get the find the data for that? 19 00:12:07.000 --> 00:12:45.000 That is because there is no aerial shots. Right. So now it all comes down to understanding and because I remember Russell street, where my, where my tube station was, that was 18 floors and I used to say South Kensington, that is around 12 floors deep. So there was a difference of six floors. So that means Russell Square is more deeper than South Kensington. So there's a line going up and down as well. So I need to figure all of that out, get the information from TfL and understand the whole dynamics underground. So next step is building the underground. 20 00:12:45.000 --> 00:14:19.000 Wow. Okay, I'm looking forward to that. And we'll have you back on the podcast when you've done that. I also think it would be fun to see to take a city like Delhi, your hometown now, which is massive. I mean, makes London look small, but to take the central part of Delhi with all the magnificent buildings and the tourist sites, and build the same thing there. Because I think if you could turn this into a workflow, then it becomes a really. Even if it's not completely automated, if there's a workflow that you can share with other people, then it becomes possible to. To do this for more and more cities. And it's beautiful. Okay, so that was the thing that got me wanting to talk to you. And then I went and looked at your website and looked at some of the other things you did. And there's a map related project that I want to talk about, which for all our listeners out there, most of you are either contributors to OpenStreetMap, supporters of OpenStreetMap, or users of OpenStreetMap. We all are, I think, who listen to this podcast in one way or another or many ways. So talk about the night the earth shook and strangers started. 21 00:14:19.000 --> 00:16:46.000 So I have this column in my newsletter where I made one wholesome story every Thursday, which, which basically is that there are kind people on the Internet and there is a good part of the Internet which we are missing, where kindness is equal to tech, how technology is wholesome and kind towards the world. So one of the Thursdays I stumbled upon OpenStreetMap and I realized the good work they were doing, like, it was like insane. Like, like what they did with Turkey, what you mentioned Haiti. And now I think they are doing Hawaii as well. When I made this, it was live. People were drawing maps for Hawaii. And when I realized my first ocean was like, why not Google Maps? Or why not Apple Maps? Why, why this? Why are people drawing? Then I understood. Then I realized the reason for it. And then I was like, oh, wait, I want to tell that to the world. I want to tell it to the world what these guys are doing, but I want to tell it in a way where they actually understand and go from doing what they're doing to understanding and then seeing the effect of it and how it is helping. And that's exactly what I did. It starts with explaining how to draw a building and then why no Apple maps? Why no Google Maps and why they're not able to do it with the imagery, because we all know that there are satellite images of the world. All like the satellites have taken it. But why we are not using those? Why are maps even important for that particular area? I cover that. And then how it actually impacts. And then it literally went viral on Reddit. People really loved it. People like, okay, we did not know about this, and this is so cool OpenStreetMaps are doing. And people signed up through me and I was like, just proud of it. I was not gaining anything out. I was just very, very happy that I got some people to actually draw some buildings in real life. And I was like, oh, my God, this is. That is all what I needed this to do. Just people knowing how big impact maps make when things like this happen. And this was just a gateway to it. That was the idea. 22 00:16:46.000 --> 00:19:10.000 And again, I'm struggling to explain to our listeners what goes on here, but you've got the map of the city, and you see at the beginning there's only 960 buildings and the rest is blank. And as you scroll through the story, you get to a map which, when you scroll over it, it starts taking you through the progression of the digitizing of that city in 2023. And it's quite. And it's just so beautifully done, and it just makes the point. And I remember after Haiti, you know, there were lots of visualizations and videos and everything, and of the instant response from the OpenStreetMap community. And this is just such a beautiful way of showing it. And I know that all of those people who are listening to this podcast, they know what OpenStreetMap does, but communicating it to the rest of the world who's not yet turned on to OpenStreetMap has always been a challenge. And I think you've done a really fantastic job here. I just love it. Then you go into, because you're a data visualization specialist, you show the graphs of who did what on what day, where they're coming from, and loads more statistics from OpenStreetMap. At the end of that, you push people to the OpenStreetMap or the hot OSM tasking manager, which is a great thing. And there's a call to action. And as you said, lots of people did it. So I just think it's a great, great exercise in data visualization and communicating a story. And I think communicating stories is what we try to do with maps, but sometimes we get wrapped up in the map technology and we can lose sight of the stories that we're trying to tell, and you've done a really great job. 23 00:19:10.000 --> 00:19:11.000 And so it's. 24 00:19:11.000 --> 00:19:22.000 Stephen, we've got just about time to talk about one more project. What do you want to talk about? 25 00:19:22.000 --> 00:20:33.000 See, as we have all the map people listening who actually know about maps and are like, really want to, like, have been following. So then I want to talk about Belts of Tokyo that I. That was a personal project more than anything. I didn't. There was no aim for that project, like, for this, like, as for the OpenStreetMap was to teach people about OpenStreetMap. This was also to teach about the belts of Tokyo, but it was more like a personal project because I went to Japan with my wife and we actually listened to a few of them and that was really, really cool. So I can tell about the project from the start. Bells of Tokyo is a map which I have made where you can hear the map, hear the bells of Tokyo, which means every station in Tokyo has a different sound for a train coming from a different line. So you hear those stations and those lines and you can go through the whole map and listen all of those on the website and actually play through them. 26 00:20:33.000 --> 00:21:54.000 So it's a journey around the Tokyo Metro, listening to the bells that announce the arrival of a train, and you can travel around Tokyo. And I mean, it's just fantastic, you know, I mean, it's really. I've never been to Tokyo and probably and it's on the list, but whether I get there, I don't know. But this is just a, you know, if you want to lose quarter of an hour just going around this site and listening to the sounds, it's just an amazingly relaxing and tranquil experience. I really, really enjoyed it. Akash. It's been a lot of fun this, and I'm really impressed. And I think the thing that's. You say you're not a geographer and your training is not in geography, but you're doing amazing things. And we, the people who say or think we are geographers, we can actually look at this work and say, oh, that's an interesting way of doing it. And we can learn some things from you about telling stories with maps. And I think you've done great work. So thank you for coming on the podcast. 27 00:21:54.000 --> 00:21:55.000 Thank you for having me, Stephen. 28 00:21:55.000 --> 00:22:59.000 It's a pleasure. If you go in the links at the bottom of the show notes for the podcast, there's a link to the Data Drop, which is Akash's weekly data newsletter. I tell you what, it's well worth signing up to this, even if after a few weeks you unsign up yourself or whatever it's called. But they're fantastic. And he's just released, rather than a data visualization, a little game which, which is a lot of fun. It's a lot of fun. It's very clever and simple, not too challenging. It's a five minute effort in every morning when it arrives. My wife and I are quite hooked on it. And I'm not going to describe the game to our listeners, but go on to Akash's website, sign up and you can get this game every morning and it's a lot of fun. 29 00:22:59.000 --> 00:23:01.000 Thanks a lot, Steven, for playing it. I didn't know you were playing. 30 00:23:01.000 --> 00:23:25.000 So. Yeah, I mean it's about being concise with words and it's a really, I mean for those of us who write and know about having to fit into word limits, this is a game about fitting into word limits. It's a lot of fun. Akesh, thanks very much for being on the podcast. 31 00:23:25.000 --> 00:24:19.000 Thank you for having me. Stephen and I just have one request for all the listeners. Just if you have any ideas for a map or any wildest project you have ever thought, send Stephen an email. He'll send me an email and we can definitely just build it. So any ideas you guys have which you thought was not possible before or anything of that sort, just send Stephen an email and we'll do it. And while I have only one more minute left, I want to share the wild idea which I had and I wanted to take Stephen's short opinion on it on the London drawn bit. Steven, would you want, if I do underground or would you rather as a geographer, would you like, if I do it different eras of London, like 2000s, then 1900s, then 1800s. So you pass through the eras and you will reach today. 32 00:24:19.000 --> 00:24:25.000 So that's a no brainer for me. That's a no brainer for me. I want the eras. 33 00:24:25.000 --> 00:24:26.000 Oh dear. 34 00:24:26.000 --> 00:24:57.000 I mean if you could give me London at the beginning of the 20th century and London in the 19th century and if you could do that, that would be amazing. And I guess. Well, go and do it, go and do it and we'll get you back on the podcast. Okay? Okay. So listeners, think of a challenge for Akash. He's put himself out there. Let's see what he can do. Okay, take care. 35 00:24:57.000 --> 00:25:31.000 Thanks for joining us today on the Geomob podcast. We hope you enjoyed the discussion. You can get the show notes over on the website, which is@thegeomob.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.