Nokia wants to make sense out of the real world through Here

[protected-iframe id=”c28185304e63c8adcedee67a59e4441b-14960843-33105277″ info=”″ width=”640″ height=”360″ frameborder=”0″ scrolling=”no”]
Transcription details:
Input sound file:
10-16 pm Session 1_1006.MP3

Transcription results:
Session name: Making Sense Of The Real World
Kevin Tofel
Hans-Peter Brondmo

Announcer 00:04
It’s a little bit of a switch but it’s going to be covering the same topic. We’re going to be Making Sense Of The Real World, which we’re going to be covering all the seasons of The MTV Reality Show starting with New York moving on over anyway. Which is going to be moderated by Big Kevin Tofel, he is a senior writer for KOM and he is going to be talking with Hans-Peter Brondmo, he’s the VP and Head of Product Innovation for HERE a Nokia Business. Please welcome Kevin and Hans-Peter out to the stage.
Kevin Tofel 00:36
We were joking about that backstage but Hans-Peter thanks for being here.
Hans-Peter Brondmo 00:42
Thank you, sorry I am not Michael.
Kevin Tofel 00:45
I’m not Michael either, so it all works out. So location, let’s talk a little bit about location. We talk little bit also about how that’s going to impact future cities, the types of map will need and so on and so forth. I’ll making call for questions we have a few minutes left, I may not be able to see you but if we have questions we have the two microphones there so let’s get started. So, back in 2010 I went to Nokia World, London, had nice chat with some of the Nokia execs and one of them told me location for us is the next big user interface, and I had to think about that for second and say, “What you mean by that?” And I’m curious if that’s still a valid point at Nokia, and really what it means today?
Hans-Peter Brondmo 01:35
Yeah, absolutely it’s a good starting point. In the way I would frame it is, is going back to the beginning of maps, right the beginning of cartography and even the Babylonians might have been the first to chisel maps and to and representation into rocks and stones and then I think you have the first geological maps by William Smith do all the first geological maps and you had the London Tube which is a very abstract in the 20s. MacDonald Gill did these very abstracts representations that abstracted physical location and just drew maps and you can [inaudible] with the London Tube Maps, and then first forward to what we are familiar with today Nokia Maps, Google Maps and these are essentially digitized tiles of paper Cartesian maps that are just put up on a screen. And frankly they’re unimaginative and so I think when you think about what is location all about in a mobile world where we’re wearing things and caring things et cetera, it’s about connecting is or bridging us between the virtual world and the real world and I think a map is really just a way to guide you through your life, and so what we like to think about is creating maps for life. Maps that guide you, but not necessarily these flat, times unimaginative things with lines and squiggle sign. That’s one interface that works in some certain instances, but that’s not really what locations all about.
Kevin Tofel 03:05
And you guys have thought about location differently for a while and it’s evident in some of the products through here City Lens the augmented reality app for example is a perfect example of that I think. So, then thinking forward to where we have large organization, larger cities, what’s the impact to that user interface of location?
Hans-Peter Brondmo 03:29
If you look at the big patterns by in the middle of the century, something like 85% of all people are projected to be living urban or tightly denser environments and just the logistics of managing those cities, creating smarter cities is a huge challenge and behind that behind navigation again finding things behind operating those cities you need a model of the real world. So when you think about, the way we have a simple world view which is– we run a cloud which is a location cloud, and if you are going to have a cloud you have a couple of key pieces. You need an index, need data and need your platform to provide access to that data and then the user experiences on the front end. And so collecting and capturing an index of the real world is starting point and you do that for urban areas, for outdoors et cetera and with these patterns again you get changing ways of interacting with your local environment. Again if you step back you look it Google what do they do? They index the virtual world, right? They created a index, created the great oracle. You can ask questions to start with ‘what’. You got Facebook, Facebook index humanity, right? Over billion people and you can ask questions to start with ‘who’. And what we’re really setting out to do is answer all questions to start with ‘where’. So, you going to what’s that key question that you are asking the index, and so we’re creating the ‘where’ index.
Kevin Tofel 04:59
Let me ask you this, you mentioned Google, obviously Google Maps, Google Earth is two of their big products. Are they indexing it differently, in a different way that you guys already think and how so?
Hans-Peter Brondmo 05:13
I think what we’re doing is we’re looking at creating a platform, and specially know you probably heard there’s been occasional lot new stories about the fact that parts of our business was sold to Microsoft. It’s not the part that I’m in by the way so I’m still a part of the old Nokia, and the new Nokia now. And weekend sell and provide our platform, I index our data and our platform services to anyone. So the beauty of I think especially this world is as some of you might be aware, Amazon is using our maps in the Kindle. Microsoft will be one of our largest customers in this new way forward, and four out a five cars use our mapping platform for the in-car navigation. So if you unnecessarily interested in having a Gmail account as you way of authenticate yourself of the service, and being a Google customer in order to get into your car we can be a provider of a platform service and there really only are two today. One is the company I mention and the other one is Nokia here.
Kevin Tofel 06:24
Interesting, you mention the car and I want to talk about that for a second because earlier we were having some conversation about some different verticals where location is going to be important. Already is important, going to be more important in the future. You mentioned automobiles and a tie into the Internet of Things which is been a common theme that we’ve been talking about here. How so? You said something very interesting to me, so I try to share it then we can discuss. You said, “One of the biggest Internet of Things devices would be the car.” And we do already think of connected cars and such; would you have a different take on that? Someone if you want to talk about that for second.
Hans-Peter Brondmo 07:01
I think the Internet of Things, discussion is obviously very interesting. Is what I do for my day job. We build connected devices, next generation devices and services in cloud and all thing, but what I think is really interesting right is Internet of Things is not just little watches and wristbands, right? And nor is it particularly interesting to me to put my fridge on the Internet, but my car – my car is credibly and will increasingly become more powerful computer, right? It’s almost more about the electronics and everything else. It’s completely covered with sensors, all my bumpers, cameras. Soon my car can see in the dark, it can see, can pickup, it can have collision avoidance all vehicle stuff. So, the car needs to see, the car needs to think very fast, right? And there’s all interesting constraints around it, and it needs to connect with whether I have a Windows Phone and iPhone or an Android Phone in my pocket it needs to connect to that thing as well. So it’s a very smart device connected to my main mobile device but it has a ton of intelligence perhaps even more horsepower some level than the device carrying in my pocket. So when I look at the car as a Internet of Things device it’s certainly the most expensive one you going to have, but then different people will use it differently, right? Some people will share, see you’ll have to actually put this thing on the Net so that other people can figure out when it’s available et cetera with all the card sharing, right sharing stuff that’s happening. Again, driving conditions being able to understand road geometry understand that index of the real world so that you can navigate through it. Figure out where the closest parking is. The car is an incredibility interesting, large challenge in the Internet of Things, perhaps the largest and most interesting challenge. So we’re working with all the major car manufacturers now, or I should say not all but many of the leading car manufactures on autonomous driving solutions. Of course again on providing the cloud connectivity and the cloud data and the cloud information that allows the car to navigate the real world in a clever can be scenic, it can be safety first, it can be speed fist, it can be green first but whatever your preference might be or optimizing all of the above navigating the world [inaudible]. I scan through the program and I realize that the car was never brought up as a Internet of Things and it’s never frankly the most interesting one of them all.
Kevin Tofel 09:35
Sure, and we’ve already seen there’s different protocols and test being performed out. The cars actually talking to each other.
Hans-Peter Brondmo 09:42
Absolutely, they setup a machine that work, they have to understand the environment around them. Real autonomous driving won’t happen until you’ve got a networks on the roads, right? With real-time information sharing.
Kevin Tofel 09:53
And then getting back to that theme of rapid urbanization over the next ten years in the large cities. The benefits are almost obvious, it’s a shame we don’t have them sooner but better traffic or reduce traffic and such more safer roads because the cars know. So, I do see more potential there.
Hans-Peter Brondmo 10:12
For those of us who commute up and down, one on occasionally or too often we all know that the rubbernecking problem, right? Some guy in the other lane that has no impact on the traffic crashes, so everything starts slowing down just a tiny bit of delay upfront propagates back. Well, if the cars know where each others are right, and can drive and you have real information flow there. You don’t have to slow down.
Kevin Tofel 10:34
Exactly, it’s great. So, another way of looking at location. I look at from the consumer point of view, and there’s a couple things that come to mind balancing, privacy with my location and also incentive to use location, share location. So what you guys looking into to address those?
Hans-Peter Brondmo 11:00
I think the privacy balance is obviously critical and I have a long history in this…in terms of privacy and I always have the strong opinion. I think well speaking personally and speaking behalf of our organization that the feeling is that you own your own data. You own your own profile and that you should be able to have some control over that. At a same time there’s a value exchange, right? By providing that data into the system and it’s even a value exchange that goes further and then when we deal with an auto manufacture some of them might start by saying, “Well, we just build our own cloud services, we’ll just build our own maps.” And as you seen with some of the recent map quality issues for some of the platform providers it’s not easy to build a map. So, by having people contribute to the map and contribute at scale you build a better map. You build a better solution. You build a better index, right? So I think that there is that value for value exchange which needs to be very explicit and there’s needs to be that ability to say, “Look create a price is owner on my house. Never share my identity tied to my privacy.” Those thinks I think are just intuitively obvious. We should even debating them they’re just obvious. Now, how we implement them of course can be somewhat challenging and difficult. But the basic controls I need to be able to said should always be there.
Kevin Tofel 12:30
Okay, any questions yet from the group? I don’t see anybody– somebody walking up there? Yeah, got one here.
Audience Memeber 1 12:38
Hi there, can you talk about a 3D imagery layer and how that’s contribute to some of the high value scenarios and I would appreciate some examples, specifically for consumers.
Hans-Peter Brondmo 12:47
Yeah, I see it’s good question. I think 3D is one of those super exciting areas because are we were touching on it earlier the whole idea of the map as interface starts– we start, little bell start going off. So when we drive our funny little cars with the rig on the top and we made an acquisition over in Berkeley recently, for a company called earthmine and they’re building our next generation cars. It’s very cool rigs that we mount on top of a car and then they run what we lidars, these laser radars, they collect about 1.7 million datapoints per second, and the reason they collect those datapoints is so that again we can build our real world model, while we’re driving down the road. We also of course image, we image street signs, anything’s we can do automatic street sign detection, but the real objective there is to create that 3D model. We are also doing some aerial lidar and then we’re overlaying on that both with photogrammetry in terms of collecting the photography which the rig also collects of course and it does literally a full spherical photographic capture, huge amounts of data but all of that is then combined in order to create the 3D model world. Now why is that interesting? You do mention live sight. I like to be able to see through buildings. I like to be able to around corners. Why shouldn’t I, right? They did it in latter on a while back. So, my view is that what ultimately should be able to happen is I should be able to hold a device up– I’m wearing it on my face, got in my pocket. It something I flip up from my wrist I don’t know, but I should be able to look through that and see through buildings. Now I can only see through buildings if there I know there’s a building there, meaning the device knows there’s a building there and has a model of the world and then can basically remove that model and see through the building because it’s knows what’s behind it and has an image rendering and a full on physical rendering of the world behind it or the world around it. So, our objective is not only to create an index in the sense, a flat index but is truly to build a model of the real world so that you can navigate and be in that world. So if you’re looking for something, say you’re sitting in your car and you looking for a particular building; what if you could see heads-up a model of the world in front of you and then just have buildings disappear and you could see a very simple road with a couple key kit turn right at the McDonald’s. Suppose to that woman telling you to turn right after 300 yards, well 300 yards how long is that again? What if she just said, “Turn right at the McDonald’s.” And then you see the McDonald’s, you see it right there, you see it out there. So, in words interfaces and the whole experience you’re creating can change because again you got this amazing ability to create a model of the real world. We’re starting of course with urban environments and with cities but we’re increasingly also going for both again photogrammetry to build models from imagery but also then lidar to build these very detailed real world models. It’s very exciting stuff actually.
Kevin Tofel 15:56
And that experience you just mention can get rid of a lot of noise that I don’t need to see on a map. When we look at the maps they’re filled with information. We think it’s a good thing but yet it can be overwhelming I think at times.
Hans-Peter Brondmo 16:09
That’s exactly right. And in there parts of the world where there are no street names. Right, large parts of the world. So you navigate by landmarks, you navigate very differently than in our world where things are very laid out here in the US and structured. So yes, absolutely you need to be able to remove things and then focus on the key elements that you want to take the left turn the first.
Hans-Peter Brondmo 16:35
Question any?
Audience Memeber 2 16:37
Hi, yes. I think here and the Nokia have done a fantastic job of reaching out to a large companies in the car manufacture stuff like that. And one of the [inaudible] I’ve hear is things like The Raspberry Pi and all these small Internet of Things start-up companies that are looking to corporate location and stuff and I think it seems nine out of ten if not 11 out of 10 use Google Maps, is there go to for developing things. Do we have any [chuckles] plans to improve our outreach to smaller developers the big giant companies of tomorrow?
Hans-Peter Brondmo 17:16
Yeah, it’s a very good question and so the API is the platform, API et cetera are accessible and available. We have not done a great job with outreach in Silicon Valley and to the next big thing in terms of location services when they’re not yet the next big thing. But I think it’s very important and I think where I’m seeing some really interesting work going on right now is actually more in the Crowdmapping piece, very quick story there in Belarus we opened up the Crowdmapping platform, because we don’t people to change freeways and highways here in the Bay Area, they’re pretty good the way they’re. And if they change then we validate that very quickly with little – true validations mechanisms and we go back and modify the map, but in Belarus there were no roads and a relatively small Wikipedia type structure, very small group of people but 200 total within a few months created essentially a complete map of Belarus with high, high detail and I forget the exact numbers of kilometres added but it was phenomenal and so whether it’s Myanmar, Belarus or lot of places where there was no infrastructure for bringing the crowd into do that and that’s using our tools and creating capabilities on top of them both on the data and technology side.
Kevin Tofel 18:34
Well, I know we have another question, we’re actually a little bit overtime so perhaps we can knew with this gentleman with question outside that we can refer the stage and the show can continue here. Hans-Peter thank you so much.
Hans-Peter Brondmo 18:46
Thank you.
Announcer 18:52
Thank you guys up.