In episode 12 of the Let’s Connect! Podcast, Ed Freyfogle, Head of Product at OpenCage, joins host Ken Briodagh to talk about Location Tracking, Geocoding, IoT Standards and Open Data networks for IoT.
Ed Freyfogle is Head of product for OpenCage. Prior to OpenCage, Ed was co-founder of Lokku, the company that spun out OpenCage. He started his career in the early days of the internet as the first developer at yahoo.de. He has an MBA from MIT, a degree in engineering from Duke University, and was a Fulbright Scholar in Weimar, Germany. Connect with Ed on Twitter and LinkedIn and listen to his GeoMob podcast!
OpenCage provides a geocoding API used by many IoT service providers at high volume (millions of queries per day) to convert device location (longitude, latitude) into useful location information. And they do it using open data from OpenStreetmap. OpenCage started in 2013 as a project within Lokku, a company with a long history of using and supporting OpenStreetMap. They implemented the first ever commercial use of OpenStreetMap (all the way back in the internet dark ages of 2006!), and were one of the first companies to use OpenStreetMap maps instead of Google maps. In May 2015 Lokku was acquired by Mitula Group. The OpenCage project was spun out into an independent brand, and has run independently ever since. Follow OpenCage on Twitter!
0:00 Show Introduction
1:04 Ed Freyfogle Introduction
1:30 What is Geocoding?
2:28 OpenStreetMap, or Google Maps: What’s the Difference?
5:18 How Much Will it Cost?
6:50 Security and Privacy in Open Location Data: Is it Good Enough?
9:00 Remote Locations, Away from Roads
10:35 Location tracking at Sea, and on Rails
13:00 Data Protection and the Unintended Consequences Law
13:30 Organizations for IoT Standards and Open Data
14:55 Final Thoughts
- [Ken] This is the IoT For All Media Network. Hello friends in IoT, welcome to Let's Connect!, the newest podcast in the IoT For All Media Network. I am Ken Briodagh Editorial Director for IoT For All, and your host. If you enjoy this episode, please remember to like subscribe, rate, review and comment at all your favorite podcasting platforms, and to keep up with all the IoT insights you need, visit iotforall.com. Before we get into our episode, the IoT market will surpass $1 trillion in the next few years. Is your business ready to capitalize on this new and growing trend? Use Leverage's powerful IoT solutions development platform to efficiently create turnkey IoT products that you can white-label and resell under your own brand. Help your customers increase operational efficiency improve customer experience or even unlock new revenue streams with IoT. To learn more, go to iotchangeseverything.com. That's iotchangeseverything.com. Now, Let's Connect!. My guest today is Ed Freyfogle, head of product at OpenCage, and we're gonna talk a little bit about open data, geography and whether or not they're making more land or just renaming it. Ed, welcome to the show. - [Ed] Thanks. Great to be here. Good to be here - [Ken] In case folks aren't familiar with you Ed or with OpenCage, can you tell us a little bit about sort of what you guys do and how you fit into IoT? - [Ed] Right. So we do one thing and one thing only and that's we have an API for geocoding. So, geocoding is the process whereby you take geographic coordinates and you turn them into address data or the opposite. So, the coordinates to address is known as forward geo-coding and sorry, coordinates to addresses is reverse geocoding, address to coordinates as forward geo-coding. And we have APIs that do both. We do that at very high volume for many different industries, but IoT is one of the major ones, primarily in the subset of asset tracking. So, people trying to answer the question of "where are my things", and I guess what makes us different than some of the major providers that people might be familiar with, the biggest of which is pre Google maps, is that we do this only with open data. So, that means data that doesn't have kind of restrictions around how you can utilize that data. And so, the biggest data set there is something called OpenStreetMap. That hopefully people are familiar with. It's become quite a big project now over the years. But also several smaller data sets usually on a country by country basis. - [Ken] Is OpenStreetMap the one that Google has been building or do they have their own proprietary one? - [Ed] Okay. Google has definitely not been building OpenStreetMap. OpenStreetMap, you can think of this as kind of like a Wikipedia of geographic data. So it's crowdsourced, it was a project that started about 15 years ago in the UK and London, and basically the reason it started was the guy who started it, he had a project where he needed to have a map of an area and so he went to go get some data to do this from the official government map service and it would have cost him quite a lot of money. Okay? It's not freely available. And, so he said, "Well this sucks. "I don't have a lot of money." And the area he actually needed the map of wasn't that big, so he said, "I'll just walk around with the GPS device "and make my own map." And he kinda did that and then he uploaded it in the internet and he said, "well, here's my map. "Maybe someone else could benefit from this." And then other people started contributing as well. And it kind of snowballed from there. And now it's become really quite a massive thing. Kind of on the same scale as Wikipedia. And I'm sure everyone listening here today has interacted with things built on top of OpenStreetMap. Many companies do use it, including my own, but also, of course, big brands like Apple, Facebook, Microsoft, are built on top of it. But people like Google build their own data set, so they've got of course their cars driving around and obviously they're tracking every Android device out there. So, that's kind of the two extremes. You have people building their own database, someone like Google, and there are others as well, it's very high quality, but if you want to use that data you need to obey Google's rules. So you have to pay their costs and they have all kinds of restrictions around: Can you store the data? How long can you store it? What happens after you stop being a customer? Do you have to delete the data And can you use the data behind the firewall? Can you display it on any map? So, in theory if you geocode something with Google's geocoder, you can only display it on a Google map, for example. So, yeah. It's a typical thing that when a developer signs up to get an API key for Google, they have to click through some terms and conditions that are hundreds of pages long that no one's ever read but the reality is if you are using it at a volume Google is gonna contact you and you're gonna have to pay a pretty steep price. - [Ken] I was gonna say, and the expense alone, leaving aside the sort of rights and uses and management stuff, all of which is probably overcome-able, the cost alone is gonna be at scale significant, I would expect. Maybe not insane, but significant. - [Ed] Well, for varying the definitions of insane. It depends really on what your use case is. So, there are use cases where you might say, "Well, Google's data is the best data there is, and I need to use that and absolutely have to have it, and I'm gonna have to pay that price," right? But, for many use cases it would be completely overkill. So, a lot of customers that we work with the geographic data that they need is at a much more coarse level, and so it's complete overkill to pay Google's price. But, I should mention the restrictions are real. Like so for example, the fact that imagine you've been a customer of Google for many years and then all of a sudden you decide, "I'm tired of paying this price," now you have to go and delete your own database. Again, that's no joke. So-- - [Ken] I would love to see compliance numbers on that as whether whether or not people actually do that. - [Ed] Well, it depends. Obviously if it's you and your little hobby project it's probably no big issue but if Google knew you were a big time customer and now you're going to switch to their competitor... - [Ken] It's murky to say the least I think that's fair to say. - [Ed] Correct. - [Ken] So let's sort of get down into some brass tacks and stuff. I talked with a lot of folks about sort of the difference between open source versus proprietary data sets, or connectivity, or any number of other types of things within IoT, and one of the main sort of arguments is about which is more secure and which one is more respective of privacy and, sort of, I think there's a lot of philosophies on both sides but generally speaking, I think that folks tend to agree that privacy is better in the open source world. And that seems to matter a lot more when you're dealing with geographic and location data. - Right. - [Ken] Security is one thing but privacy really matters with location data, I feel like. - [Ed] Yeah. So this is a very big issue and kind of put on the map significantly by the GDPR two years ago, but even before that, of course. Yeah, so, if you know where someone is and you're recording their location every 30 seconds or whatever, or the device or whatever, you can create a pretty precise profile of that person. So, location is deemed to be personally identifiable information, PII, and this is frankly another big reason that a lot of our customers don't wanna send their location information to a player like Google. A lot of European companies don't really welcome the idea of giving even more data not to Google specifically, but also just big American internet provider generally. And so they prefer to use things like OpenStreetMap or even just to work with a small player 'cause then they know we do geocoding and only geocoding, and so obviously we're not tracking their users anywhere across the internet. But in general, the other big advantage of open data is similar to open source. They say if the software doesn't exactly solve your specific problem, then you have the source code and you can change it and adopt it. And it's similar with OpenStreetMap in that if the area that you need to know the geographic information about isn't in the database, you can add it. And it truly is crowdsourced. And so this gets to another point. A lot of the big data providers, Google or the others, they've traditionally focused more on the high value markets. So North America, Japan, but also traditionally focused much more on areas that vehicles go. - [Ken] Alright. Higher population centers, right? - [Ed] And so, vehicle routing is such a high value use case that it's all about the roads, but there are a lot of use cases where the roads aren't... If you want hiking trails, you want bike paths, things like that. And so if, if that's what you need you can go out and map it yourself and put it into OpenStreetMap quite easily actually. - [Ken] The IoT implications of that are pretty big, I would think, because so much of IoT is remote. So much of the things that need to be tracked are not on a road. Where they're going, they don't need roads. - [Ed] Well, another very real use case is construction sites. In that it might be just the location that's just not on the map yet, right? Even if it's in a major city but because it's a new development or whatever, the kind of formal bureaucracy takes its time until it actually gets officially registered and things like that. - [Ken] Well I wanted to ask you about that because you mentioned earlier sort of the difference between address to geocode or GPS coordinates and then vice versa, I wanted to ask about so what if it doesn't have an address? What if it's wilderness? What if it's whatever, right? - [Ed] Alright. So if you send us coordinates, we'll return something for that in the limit that... We just kinda zoom out, if it has a precise address, we'll give you the address. If it doesn't, then we look for the road. If we don't have that, we look for the kind of the postal code. It depends on the country and the town, the region, the village, the municipality, all the way out. Of course, most of the world doesn't have addresses in the sense that most of the world is covered by water, right? - Well, right. Yeah. - [Ed] So on oceans, bodies of water and things like that, which is also a major use case of course with shipping. - [Ken] Yeah. In the US trucking is a sort of from sea to shining sea, I think a major sort of asset tracking issue, but that's also somewhat in doubt, I think, over the next 10 years. I expect to see a lot more train use. It seems like there's a lot of development happening in terms of rail. Eventually, maybe even high-speed rail here in the US like the rest of the world. So, I think that roads are going to become... Although probably at least in the US always gonna be important, probably also in Europe, but the rest of the world might never have as much focus on roads for asset tracking. It might pretty much always be rail or whatever comes after rail in terms of mass movement. And so, I'm really interested in, especially in the use case of asset tracking, because that's probably, maybe I'm wrong here, probably the most common sort of use case for geo-tracking right now anyway. - [Ed] Yeah, certainly. I think there are two pieces of that. So one piece is that the devices that do the tracking, of course, are continually getting cheaper, smaller, longer battery life, and on the other hand the connectivity is of course getting better. So, whereas previously you would have the guy in the army with the huge backpack-sized device that would track where they were, now everything will have a device and even if you're in the middle of the desert or wherever you'll have super precise location. But what that means is that things that previously weren't being tracked are now being tracked. For example, pet tracking. We work with several customers who do pet tracking. We work with a lot of things that where it makes sense to track a cargo container or a vehicle you're, now it makes sense to track... If you're buying an electric bike it'll have a tracking device built in, any device, anything that's over a couple of hundred dollars in value we'll have tracking. - [Ken] Yeah. Why not? It's so cheap now that the cost benefit is impossible to ignore at a lower and lower threshold, I think. - [Ed] Right. So then these things, are not an interior electric mountain bike or whatever, you're gonna be riding it out on the trails and things like that, so you're gonna wanna know where that is. I think in all aspects of the digital realm privacy and data protection is gonna become more and more of an issue as everything goes digital, but location particularly. - [Ken] Well, I've been using location as an example of something that security concern that I've been concerned about for a while in IoT which is the unintended consequences law of data collection. The idea of we don't know what we're going to find out as we collect larger and more diverse data sets and then run smarter algorithms on them to look for correlations and things like that. We don't know what we're going to find out. And some of that stuff could be really revealing and/or sensitive information. And I think location is extremely sensitive as people are starting to realize. 'Cause that just tells you so much about an individual person. Is there a standards organization or standards work around location services yet? Is there a sort of like an association working on that? - [Ed] Yeah. Yes. There are quite a few actually. There's something called the OGC, the Open Geospatial Consortium, and it's a big issue because more things are moving than ever before. So we're free to, say, you had a few cars or whatever going around now you're gonna have drones flying around, you're gonna have all kinds of things that are moving and need to be coordinated, so yeah, We need standards at all level. But first of all from a technical perspective of kinda data coordination, but then I think as a society we also need to decide like what's okay and what isn't. - [Ken] So we're getting near the end of our time, unfortunately. This is such a big topic. There's so much that that can be covered. I wanna make sure we leave the listeners with sort of something concrete. We've talked about a lot of sort of things to think about with location, but if I'm a company that's looking to implement or enhance my existing IoT with some say supply chain management, which is as I said before, I think probably the most common use case in the the B2B IoT world-- - Yep. - [Ken] What should I be thinking about at the beginning? What should I be looking for as a general rule in an asset tracking, mapping system? - [Ed] Well, I think it's about understanding your use case in terms of where are disasters gonna be, what type of connectivity will you have there, Is it all in permanent fixed facilities so you'll have wifi or things like that or is it gonna be traveling in territories where they might not have good reception and things like that. So basically you have devices that from the connectivity, established location, via GPS, via wifi or whatever and that will have different degrees of accuracy. And then think about, now you have coordinates, right, which have some degree of accuracy, but then at what point do you need to turn that into useful information? Which country is it in, which tax jurisdiction, which address is it at? Things like that. Or even you can of course go... There's a whole universe of people doing indoor mapping, where in the warehouse exactly is this device or things like that. So, location has many, many aspects, be that technical, but also be it as we started talking about the legal aspects and the potential security issues. So, it's big in this complex and it's probably not something you should just add on as an afterthought. You should think through it from the beginning and obviously if anyone wants to do that, we're here to help. So if they wanna get in touch both... By that, I mean of course my business, of course, but also I have my own podcast which is weekly. It's called Geomob and it's a together with a friend and every week we just talk about different aspects of location-based services, interview different startups, different people from the industry map world. There's a lot happening. I know there's a lot happening in IoT but there's just as much happening in the location as service fields. And it's really interesting. So, get involved and yeah, get in touch if you have any questions. - [Ken] Yeah, for sure. And we'll make sure we get those links into the show notes for you folks out there listening, but after you finished subscribing to Let's Connect!, make sure you subscribe to Ed. Ed, thank you so much for being my guest, really appreciate you sort of talking us through some of these logistics and mapping issues. It's really complex and interesting, but there's just so much to wrap our head around. I might have to have you back on again. In a couple of months we'll talk about it some more. - [Ed] Anytime. Anytime. Happy to come back. Thanks for having me. - [Ken] Thanks again to all of you listening out there. I hope you've enjoyed our discussion and if you have, please make sure you like and subscribe so you don't miss out on any of our episodes. We post every week and I hope you'll leave us a rating review and comment on your favorite podcasting platform. If you'd like to suggest a guest please click on the link in the description. And we also have a great sister podcast on our network called the IoT for all podcasts, so make sure you check that out. - [Ryan] Hey, Ken, let me jump in real quick and introduce your audience to another awesome show on the IoT for all media network. The show that started all the IoT for all podcasts. Where I bring on experts from around the world to showcase successful digital transformation across industries. We talk about use cases and IoT solutions available in the market and provide an opportunity for those companies to share advice to help the world better understand and adopt IoT. So if you're out there listening and haven't checked it out be sure to go check out the IoT for all podcasts available everywhere. - [Ken] Thank you, Ryan. Now get back to your show. - [Ken] And thank you all for joining us on this episode of Let's Connect!. I've been Ken Briodagh, editorial director of IoT for all, and your host. Our music is sneaking on September by Otis McDonald. And this has been a production of the IoT for all media network. Take care of yourselves. - You are listening to the IoT for all media network.