 So our chat for the panel today is Arun Ram, Arun in Ramarathnoop, but we are going to discuss Arun Ram better. Arun and I have worked together on the back-end panel for many years and Arun is also now the director of the Wikimedia chapter in India. So anything that you hear about in terms of Wikipedia's activities in India, it's coming from the organization that Arun is ahead of. Arun is going to introduce others. Thanks. It's always difficult to have the final session of the day, so we will keep this as interactive as possible. So first of all, I would like each of the panelists to give a Twitter-like update about themselves so that the audience knows who they are. Today we have a little bit of a loaded panel with, you know, the game-changers, so to speak, the Microsoft developers. Yeah, that's what you call a CEO. Fine. Well, let me say what I want to say about you guys. Yeah, ground rule number one, the chair's, you know, decision is final. Okay. Then we have to balance it off, you know, a gentleman from the system integration side. So first of all, let me just request the panel to sort of brief the couple of sentences to introduce yourself, starting with Yenna. Hi, please. Good evening. I'm called as Yenna. As short as possible. I'm known as Narayan Ramanathan. I'm the global head for GIS and telematics practice from Mahendra and Mahendra Group. And finally, in this industry for 15 years, similar to Kiran, Eatsleep and Lewith, kind of, you know, programming chairs and so on. I've been in the Advisory Committee for Government of India for NSDI and I'm also the technical board of director for clients for the 4G initiative in India and the Euclidean Masters Award winner from the European Space Agency. So I govern them on the Euclidean implementation on LBS and navigation techniques. I'm Satya Gurd. I'm an engine manager with Google India here as part of the geo product area. We look after the mathematical product that some of you might be familiar with. I've been in the geo area for a minute or several years. Before that, Google, I was with Nokia looking on several products including Nokia Max for a few years. I work with being Microsoft. My area of focus is local and any application tools platform that have a local intent. Before this, I was with Microsoft U.S. and with MSN and online services. Before that, I was in Equifax as an enterprise software manager. Hi, I'm Srinil Pai. I work at Yahoo on maps.yaw.com. I'm a tech leader on the team. I'm involved in implementing mapping solutions across different properties inside Yahoo. Thanks so much. We have a really exciting set of panellists here. First of all, probably a bit of explanation of the format that we'll follow today. We've got an hour and the organization is willing and the interest will be a few minutes more. Essentially, the way we're going to structure this is we have about four different segments. Essentially, the way it will work is that we kind of introduce a certain segment and the panellists will go around. One of them will actually introduce that segment and share the thoughts and then some of the other panellists will chime in. At the end of that segment, we will open it up for questions. We typically will have five minutes for questions and maybe towards the end as well. We would request people asking questions to be very brief because even the folks in the audience are going to be very crisp in their responses. The idea is to get as many questions as we can so that we have a lot of perspectives from the audience as well. So that's pretty much the format. Well, first of all, maybe we have a pretty interesting panel, so to speak. The geospatial industry is right now at a very exciting moment. If you actually compare it to the larger IT industry, which is about 3.6 trillion in terms of its size, people are really struggling to actually size the piece, if you will. No one's really ever reported it. There's no reliable report today that actually can authoritatively say what the size of the geospatial industry is. There are a couple of reports that come in, especially in Massachusetts based company, which sort of has been trying to give indications of what the size of this industry is, but there are lots of debates on its size. For example, the GIS industry is estimated to be around 10 billion globally in a couple of years, whereas India's geospatial industry as per FICI is going to be about 23 billion in a couple of years, which is sort of the official FICI number that's going to be released. So I think the other thing that all of you know is that we have quite a few people from Yahoo! Google and Microsoft, and these companies have obviously made it quite exciting for the average consumer of the internet to sort of get closer to the geotechnology space, if you will. And their sense has also sort of shaken up the traditional GIS players who've kind of been pretty much started off client-based and are kind of moving rapidly to the cloud-based space as well. So given that, I think there's also a lot of exciting stuff happening around data, especially with the sort of acquisitions that happened on tele-attelers, like Nokia, et cetera, and now with OpenStreet maps sort of doing what it is everybody knows about it. So the challenging times in terms of how data would be, how is OS sample and transform the space, how are the current players in this space going to monetize things so to speak, and how successful they'd be, and how things would shape up, and how will the end user experience change as well. There are also other dimensions. Most of today's discussions happened around the APIs and a lot of what startups are doing and how applications on a mobile space, et cetera. The other aspect that we hope to touch upon in the panel is also with regards to applications of the geospatial industry, specifically industry verticals as well as in terms of how governments are actually trying to use that, which we believe are three different spaces, the consumer, the industry, as well as the governments, those are three different spaces. We also hope to sort of touch upon a bit of the standards debate and there's a lot of work happening in terms of standards with the OGC as well as the OSGO folks who are also trying to do a lot of work with the FOS4G events as well, and we'll probably wrap it off with a little bit of touch upon the trends and a few tips and a bit of crystal ball gazing. So that's sort of what we intend to cover. So there will be four segments for the day. The first one, given the fact that we have folks from three of the folks from the companies that do this, we will focus on the initial segment for about 20 minutes around mapping APIs and data and issues around it. And I will request, you know, so maybe I'll take it off. Sure. So I've been working at the house since December on maps.org.com and the developer story is pretty simple. Like, as much data as we can and as standardize the format as we can get. There's a data layer and there's an interface layer. Recent entrants into this are people like Tilewell and Eatslight and a whole bunch of open solutions. But these are our tools that on the developer side we try to get together and present to the user as an experience or as an interface. So yeah, there are some recent things that have made this a little more exciting for us, especially in the web field, stuff like geolocation and the fact that there are many, many of our users who are willing to check into places that they go and single intent. This is all data that is coming back to us. That's one thing that I wanted to really touch on. And the other fact that we are now getting other sources of data, for example, photos for a while now that are geotagged. So this leads us to have an interesting analysis and interesting interfaces to present back to the user. The Flickr map is essentially my favorite example. It's really nice and it makes sense. Even Google Maps has a layer for photos, the sort of thing that gives a clue of what the place is. And finally we get a lot of our data from Nokia if you didn't know, Yahoo Maps gets a lot of its location data from Nokia including stuff like driving directions and all. That I figured is still an unsolved problem. So yeah, this is stuff that I know that Google and Yahoo Microsoft is putting money into. And yeah, that's what we're going to talk about. So just want to say a few thoughts on non-data and how it's kind of evolved in the last few years and maybe a couple of decades. From my perspective, I think the end goal that we all want is to have the most accurate description of the word and the most appropriate description of the word that's in a usable format. And this kind of got triggered by the two things that triggered kind of the revolution of advance. One is the internet where data from being kind of the domain of the geo-place became so influential for all the consumers to consume the internet. Second was the availability of cheap GPS devices. Again, consumers are going to feel first-hand how data kind of influences them and kind of data is from confidence to navigate in an environment. So those are the things that got triggered and that initially started out with digital marketing companies providing you as up-to-date as they can in terms of providing updates for data, let's say once a quarter. Now that has kind of evolved much faster. So now people want, people are saying that you have the facts of the world. There is the roads and localities and the ones that resolve that. But they want to have much more real time effects. Things like road closures, things like even perhaps restaurant menus which is overseeing all the time. So that's something that data is evolving towards. There has been efforts in GIS especially to make the data quite structured. So every bit of data is exactly described and what it means is perfectly specified. But I think as we kind of start investing in this more loosely defined data then people have to get used to finding some of the answer to that and we should create data as a valuable resource regardless of the structure around structure. So that's valuable in itself. The other rise in the last few years of course has been the contribution of user-generated content. So users themselves feel the need that there's data in the same thing helps in the world. And then using server products will not be the one where they contribute the local knowledge to the system by either drawing maps, identify points of interest and so on. So the domain of data generation which used to be in these multiple digital marketing companies is now actually moved into the user data. So we'll have to see how these two co-exist and how these two kind of interact with each other. And one interesting portion is all the data in the world will be generated by users. I think that's something that we can discuss and debate over. And the final point that I can make is that even though data was driven by these applications things like mapping on the internet or navigation on the road I think it's some value-proceed data as it's on set. So essentially collecting accurate data and making it available is a good answer. A pro question to you guys. On one side we seem to be trying different ways to monetize. On the other side you're having folks like OpenStreetMag getting data. How do you guys actually see tackling this? I mean obviously it's an area that you're working on quite actively. So I can talk about from a data standpoint there are two issues. When you talk about consuming data and that's where OpenStreetMag's OSM comes in is the restrictions to use data and the availability of data. A lot of people get touched upon tier 2 cities, tier 3 cities and lack of data available. And that's where we enter OSM where we want the crowd to come in and the local people to come in actually add those information and then it is available to everybody. So that is essentially one of the things about OpenStreetMaps is that get the data at one place and make it available to everyone and that was sort of the driving force behind the data. Regarding the monetization aspect I think it will come as the data consumption, paradigms and the applications evolve over time and the industry is evolving very fast. So just like today we saw a bunch of paradigms or applications so to speak but very soon we will see newer applications and things to come. So that's how I see proprietary data or curation effort which is focused towards a particular goal versus a crowd source and a map that is available to everybody, a geospatial craft that is available to the world which is basically OpenOSM. So how many of you have a take? Actually we have the DJScast mic. Yeah, it's just a couple of thoughts about monetization, right? So I think additional digital map providers were really, the business model was based on supply and data. I think that's going to get harder and harder as much of the data, if actual data becomes available. But then I think one thing to realize is that there are actually multiple tiers of data. Depending on the application, you have one kind of level of data that you want to have if you want to just display pretty maps. Another level is where you want to be able to local search. And a completely different level is when you want to have turn button navigation at your target. So there's different levels that you can play with even if you're just in the data market and say monetization can still be done in the higher levels. The second approach is what I think Google has taken is that Google does not try to make awareness of the data itself but actually our own experience around data. So essentially when you use APIs, the performance search and the data are performing directions. That's when third party developers will have to pay. And of course consumer experience is also like monetized by advertising around those experiences. So it's not the data that's monetized but the experience that's like monetized. All right, what about you? So at that point, that is the jumping point for any startup. For example, anybody who wants to get into the scene, the first thing they say is okay, I'll write this data, I'll talk to my users or how do I accept data and make my own database about it. Yahoo does the same thing. API is free for like a small scale at the moment. You actually want to make money after it, make a business after it, talk to our guys. So you're absolutely right. The data in itself isn't psychosatonic. Indra Nagar is Indra Nagar. But the kind of experience that you give to a developer or to a business that's consuming this data that is where the value lies in this industry, right now at least. If someone in the audience wants to add to this, we can take an audience opinion on this point. Anyone wants to take a photo? Yeah, just from this point. See one of the things is I think the broad spectrum of industries that can leverage GIS is something that I think it's still at the tip of the iceberg stage because I think GIS has been at least there for about 100 years as far as I know in terms of about 50 years maybe more than that. You are the expert on this. First the GIS map is drawn by a kill map. Okay, technically it's been there for ages. But unlike, for example, you know, I work with tools like SAS and SPS which are essentially the same. You are layering different types of data. However, I think in terms of layering, relevant data on top of this platform is something where I think there are a lot of applications No, absolutely. And OSM, is it OSM just by itself? It's a combination of a whole bunch of open source tools that are around it and give the developer experience to that person. I was talking to somebody during lunch and we were talking about the insurance industry, for instance, you know, if I'm paying for an automobile insurance in say Chennai versus say I'm preparing in Bhopal, say this is in Madhya Pradesh, I end up paying the same amount. Why should it be? Because it's one of the course and another is in the central part of the country, right? We haven't still brought in that intelligence into the insurance industry. So I think it's one of the problems. Yeah, so I think, yeah, I agree with you that third party is like a lot of value grad. I think an example that I want to highlight is like, there's an NGO called Transparenchai which was basically trying to figure out what are the developmental areas in Chennai and what needs to be addressed and so on. So I think it's something like clean public toilets in Chennai, right? And I try to do pure correlation between that and let's say the residents of the nearby council or something. So it's like interesting analysis like that. It's possible now that you have a base that is available and then third party is like Transparenchai holding their data on to that. The next question I'd like to pose to the panel is really with regards to data quality and we talked about it earlier in different sessions. So what I'd like you guys to reflect on is what are the different techniques that you're using to tackle this aspect and how you're engaging communities as well in terms of trying to address this and make it as easy as possible for any of us. So this is, I think in terms of when it comes to Jewish spatial, I think data is the most downstream of the fundamental thing that everything builds upon and if you have the data which is of lower quality, I think everything upstream is going to go back or south from that point on. So cleaning the data, getting the right data, ingesting the data, normalizing it, making sure that we are able to see entities as they are or see businesses as they are because there will be multiple sources that may be flowing in the same data or same entities and then be able to mark them and rotate them properly and serve them. This is essentially the data life cycle, right? And we are from an industry standpoint, we're looking at touch points at each level. So OSM is think about as a starting point and then when you come to tools that are on the periphery of OSM are the next level of things where people start to say Orbecky mafia for example, right? Where people can come and start to collect listings and that gives us another signal. So that is how we are looking at how every step of the way we can get the community involved and get the collective wisdom from the people who are there on the ground versus us who are sitting, let's say I'm sitting in Hyderabad so I don't know anything about Bhopal or Chennai for that matter but the person who is in Chennai for years and years, they can actually, potentially even in sleep can say this is the right address and this address actually is the red house next to the pink house or green house or whatever which I can't. So that is how we are tackling that problem. To compliment RFR with the community effort as and when we can actually instrument and provide interfaces. Just to add to that, another major problem that at least I see amongst a couple of teams in Yahoo is nothing in India is spread the same way across this thing. I don't know the correct spelling of dipsundra, I don't know, I don't think I have decided whether we are Bengaluru or Bangalore yet and this is the part where the data starts. Some parts of it makes and some part don't and we need to, a lot of it is done by a lot of things of course but we actually have to have human intervention where people keep telling us, you sent me down the wrong road, you sent me wrong way and you are pointing me a city and you are telling me Surat instead of Surat Kalp. These problems are huge in India and of course we are trying to put a ban on it but really tough so far. So I don't know, this is one place where Hyderabad is ahead of Bangalore. We thought so. Yeah, so this one, yeah this affirming the data quality is a hard problem where I mean there are multiple dimensions that you can look at. One is the coverage, so it's how much of the data you capture when you compare it to the viewer and that's by definition of no, because you don't kind of don't know what you don't know and you can use estimates like what a total number of roads to kind of capture that but it's a very rough, right? The other aspect I think on the fine-grained level, I think user input has been a very good mechanism because they didn't report a problem or they didn't actually go and leave it at all. They actually try to make the data quality better so that's where we can drop our data quality. But it's a hard problem. One interesting point to add, basically what we are trying to do is instead of attacking a problem it's the source of the problem. If we can be able to define one source of truth of the data then that will oversee almost 80% of the problem. So what we are trying to do instead of a top-down approach, we are trying to have a bottom-up approach because we are providing to the, in fact what are the maps we see from Naftai, Karnokia, is definitely back in Hyderabad in Baalupadi. So we know that what kind of quality of data we generate and Sardar Vallabhai Patel, probably like he's not alive, his name is spelt in Saudi to different forms in maps. That is how he has been represented. So what we are trying to say is is it any way to have a tolerance defined? Vallabhai, I can generalize the data. At the best we are able to arrive on the Naftai dataset is 48.2%. So that is what the tolerance that we are able to bring in and we pass on our bug to Google and Microsoft and then on. All right, I think with that... I have a question for the balance sheet. Yeah, we are just coming there. With that we can close this segment and then we'd like to take about five minutes on questions around the consumer experience, data quality, APIs, et cetera. And I'd be very keen if there's anybody in the audience who is actually very familiar with the open GIS tools. Actually, I've got to get a bit of OSGO, see if you have experiences on that. I'd love you to talk about it. I've sort of been working in this space for a bit. I started volunteering for the OpenTreatMap Foundation. Started adding one to data and also consulting for a lot of nonprofits. Helping them set up their own geospatial structure. So what I started with is that collecting data on the ground and using the OpenSource, OpenTreatMap tool ecosystems and trying to upload this kind of data. So one of the things that I've experienced and when I share these kind of stuff with other people who would want to contribute data is that the user experience is not that great. And that's one of the places where Yahoo! or Microsoft or Google is ahead of what it is right now. So it's really about the user experience when it's mostly... So one of the things that works quite well with MapMaker is... So that was about my question in the morning. How do you think MapMaker is quite well adopted in India than OpenStreetMap in the US? Because more or less it's the same features that's happening, but just that the data going to Google and the data gets into an open repository. So user experience is something which we found difficult. Apart from that, most of the tools around OpenStreetMap with the open geospatial data ecosystem is quite brilliant. It starts from collection, editing. They have their own ways of tagging and the entire system of tagging your own. You can apply for a tag it gets voted up and it gets selected into database. And then that's the tagging scheme. And then you have ways of storing the data. You have tools and ecosystems around the Postgres and PostGIS database systems. And then it comes to representation how you represent this map on paper or on the web. So you have a set of tools which does that like renderers. You can write themes for these renderers so that's something which I found interesting. I've been able to generate thematic maps quickly on OpenStreetMap than with Google Maps because I can write my own customer style sheets and say that I don't want these roads on the map. I don't want these stores on the map. I just want these schools. So that's something which is quite interesting. Apart from that, the data quality is still a problem like everyone is trying to tackle. It again depends on the number of people who are computing the data into the repository and things like that. So yeah, that's why. Yeah. Can you pass it to me? Raise your hands please. I'll pass it to you. So I just wanted to differ with one of the panellists because one of you said that in terms of user generated content I feel that, you know, it's more of a botheration for me than encouraging. One of the things that I usually do is switch off all the layers where there is user generated content because typically our mindset is go to that authentic source which will provide me the right data. For instance, if I were to look at palace grounds here, all around palace grounds including Mekli Circle is marked as palace grounds, right? So that is something that's a disturbance for me than using it as any authentic source. Yeah. Can you pass it to us? So I said user generated content is important. It's not necessarily important and it's wrong, but when you actually take it and curate it and have the processes to make it sure that you have attributes like canonicalization, which means that you have one copy of the same everything that you have there, then you actually make a really good use. And I think the value in at least the fact that the India map is most predominantly contented by users and that's now kind of used in Google Maps for navigation, which means it's moving at a very high bar of content code. I just wanted to add a small point to what Sajith said about the map model versus the focus sheet map. I think it also has to apart from real stability, it also has to do with people who continue being the end consumers themselves. Google Maps is a public facing platform. Of course, open sheet map can be used for it, but the primary use has been by developers for open sheet map. On the other hand, let's say my house is there or my office is there and I want to put a map of my office in our contact page, then I would like that to be added to map market and any root mismatch will report and then it is in our interest in some sense to do that. So, basically having a mechanism for users, like a large number of users to consume this data in a simple way, that's like the biggest incentive I think for user generated content. If you don't have that mechanism in place, it's going to be hard work to get users to control it. And to just add to the comment about user generated content, I think in a lot of scenarios, that's the only choice we have. There is no authoritative source as such. And even if you want to think like some provider is an authoritative source, they are also basically doing the same thing, right? They're just having people walk on the street with GPS devices. So, and they're just making the best estimate of what street or what palace province is. So, there's no such thing as authoritative and user generated content that's what we have, which is the best thing in many parts of the world. A request to the audience, when you make a question, could you just call it your name please so we could refer to you when we respond? Thanks. I'm Vipul. So, my question was actually leaving out from the discussion around user generated content. I agree completely that, you know, it is a kind of good source of updated information, right? Because things keep changing and the users report that enough. But I am sure that there are accurate sources which are, so, for example, when you plan a layout, right? You know which number, down the street, second main, third cross, right? Which number is 275? Those things don't change, right? And surprisingly, I remember that those things used to be on paper maps, which were available in electronic form. And I clearly remember that the booklets of iShares city maps being circulated were those things have they been lost, have they been deleted, are they done, what happened to them? Those data, for example, house street house number information, right? It's pretty much unavailable. So, my question in a summary is there are definitely more accurate sources out there which are static, which don't change. Why aren't they in any form or why aren't they available? Yeah. Yeah, I have a question related to government things. How a government organization collects data? Can we park that for the next segment? We'll be having a section of it. Okay. So, yeah, so the question that I have is, and the aspect of this user-generated data most of the time the formal name is never used. For example, if I go to an auto driver and ask him where is I wrote, he won't know, but if I ask him where is the double road, he knows. And nobody calls it KGPS or whatever, everybody calls it majestic. So, that is another problem which I know it's very hard when I use all these maps. How do you get stuck to that? Location area is similar. Yeah, so, one way is to have multiple names for the future. So, your road can be called multiple names. Obviously, on map rendering, you can do only one, so you can actually search by different names. So, you have like old and obsolete names also along with the primordial and official names. As a user myself, I haven't really seen a way where you can actually see multiple names you hover over a place or something. Is that something that you guys have any problem with? Yeah. This is a question that should we show multiple names when the user is that what is the question? Most often, I don't get the most commonly used name. I get the official name which nobody uses. That's the problem. And I think something they talk to how we tackle that problem. It's from a concept standpoint that's pretty much the same. But I think your question was on the user interface should we actually show multiple names and my thinking is that we should pivot to what the user has searched for rather than giving 13 different names which could again be a confusion. It's a little more complicated than that. Say for example, I'm looking at the map of Bangalore and I'm looking at some of the sub-open areas. I hover over a map so I wouldn't even know. I wouldn't have searched for it to try. It's a complex issue. I don't know if there are answers on the finalized part but that's a practical issue. It's a valid point really. There's one more point on the question about having standard sources of data. We can start thinking about it then you actually see that when you go to the government they actually have multiple views on the same type of data. If you ask what are the sub-localities of localities in the city, what department will have a different set of data and your administration will be different. In terms of user visible data that contributes to experiences then it seems better from our experience to actually let the users define what the divisions are. Even though the official division actually doesn't really quite influence what the users want. There is no clear, there is no one truth. For example, we talk about locality but the government doesn't ever talk about locality. It's more municipality. It's district, it's taluk, it's municipalities. Basically one locality could be one municipality multiple or two or four. That representation isn't there. Those are some of the things that are surrounding. The example was house numbers. They never changed and they're not there. House numbers I think my parents also have been reading about like two other times in the last ten years. They don't date as much as like there's a shopping cart problem. Also, I'm not sure how comfortable the government would be giving you the house addresses and names of everybody. There are no real documents. This is because what I said, these were on paper maps. They are still on paper maps but they are not electronic maps. I request you take that offline. The gentleman there and we'll probably do that in a minute. Thank you. With that, we'll move on to the last question. I just wanted to follow up on the question of displaying multiple names. You talked about a trend toward more user-generated content rather than from single data sources in older days of GIS. Do you think that the actual visual representation, maps are very visual, hasn't kept up with the fact that now we have multiple diffuse sources? One of the things that frustrates me in a lot of times of mapping is Google will have done some reverse geocoding and they say, okay, here's this plan on a map. But there's actually a large uncertainty about it. They only really looked up the street and they're giving me the middle of the street. And yet the marker doesn't really visually indicate the data uncertainty. I was wondering if any of you are looking at specifically your visual representation and your UI layers of how you represent the messiness data instead of hiding it from the user. There are two very good questions in there. One is the visual representation. You're right, maps aren't really kept up in terms of giving individual experiences to users. That's definitely an interesting area. You can look at what Google has done with search. It's more about trying to make the search possible. I think that's a good area to explore. To make the visual map really something that the user can relate to. The second part you might about geocoding, yes. I think the geocoding has to get a lot better in terms of going, I would say, at the extreme going towards point geocoding. You start labeling all the points that you know. So the old method of interpolating from localities and streets and so on will end up in that scenario. Do users are actually able to quantify addresses in many cases and that will help the geocoding aspect? Great. We'll now move on to the second segment. So far we've covered the consumer side of things and the APIs and stuff like that. The second segment we want to talk about what this means or what geospatial technology means from how can we apply for multiple industry protocols as well as in terms of how governments globally try and use it. I'll probably kick it off with a bit of a context here. Typically enterprises have quite a challenge. They're very good at handling applications and databases and stuff like that. But the big challenge that you see in enterprises is that the folks who can actually apply the geospatial technology are actually sometimes the engineering teams. They're completely digital team and these are siloed within enterprises. So that's a huge challenge in terms of the applications understanding the geospatial technology and vice versa. So in a sense these are in a sense converging but IT departments and organizations really have been shaped. So that's one challenge. When you look at governments really I think the big challenge is really not budgets but more about prioritization, decision making and you know things go round and round. So first say if you look at it globally the spend on geospatial technology has been around the the difference in government and you also have the telecom guys as well as if you look at utility companies globally are beginning to use geospatial technology. I'm not sure how many of you noticed this in the last 48 hours. We actually had this startup called Spacetime Insight which actually was in a $40 million was sort of developed funding for essentially the intersection of geospatial and the smart grid so they really go to map energy and smart grid data in the US. So that's what has been recent development as well. In that sense there's a lot of huge potential. It's not that enterprises haven't really been using geospatial stuff but it's sort of a lot of them have used some of the earlier client based applications and you know the kind of leading vendors there have been like VSI's and the internals have been kind of servicing them quite a bit but now with the cloud based services coming in so it's quite an interesting challenge for them to sort of figure out how to leverage the assets that you have and also kind of leverage the move to the cloud that's really happening. So I would actually request NR to probably who's actually done a lot of work around this space to touch on what are different industry what it is doing with geospatial technology, what are the kind of applications they're using and some subsequently maybe we'll do the government one so maybe you'll like to touch upon the industry particular. Thanks. Guys probably I'll give up a big perspective. Let me start with some good news. I've been asking about the open source perhaps to be used and so what the news for you is Indian defense government has backerized for the make program so Indian defense want one day open source. So probably our friends may not like here but let's say that's all it goes for. I'd like to say we are getting into a window of seed. Its small GS has become more basic so that is one of the things which is driving all industry verticals that means which are just brought in. The space time ink is just for smart grid GIs and anybody has got a guess how much is there growing year on year. There are so many budding progress so I've been waiting it. Any wild guess? Let me make it a little easier. They are growing 300% year on year. That's the growth that GIs is currently pushing in. Probably I would like to say that the recent research report which published is that close to 3.7 billion dollars of business is that 2017 for GIs. Which I talked about data and application. Nothing to do with hardware or software licensing. With that probably some good news for the government here is the Bureau of Labor Statistics from the US because that is one of the primary focus for us where we have the records. They say that GIs is the one industry which is helping the US government to trade off the unemployment problems. GIs industry is giving them 35% year on year employment for their citizens. That's how they are trying to tackle it up. If you are having a segment I have a break up also here. If you are a cartographer you are more into map creation and so on. The business opportunity year on year which has been generated is 22%. If you are in the survey and mapping kind of a thing it gives you 16% year on year additional employment. If you are into other generally not rated cartographer GIs it is growing at 14% year on year in terms of employment. That's a good news. When they move in for cars and arbitrage and your pool obviously it will be taken up to the different countries which comes into India and China as well. Having said that if I come to the Indian context from global to US to the Indian context all government agencies are making GIs as one of the primary policy and action decision in every department. The NSDI has been put up and in parliamentary commission where we have been pushing hard to the existing government. National geospatial authority will be passed. That means now we will define how the standards of GAs data should be. If anybody wants to publish data what are the standards they want to fall in? What things should be in place? That generated more data and most of your source of the problem and the data will presume to be used and the largest spend in terms of mapping Indian railways has recently approved 40 crores of budgetary amount just to build their GAs data base. It is going to be interoperably used by the telecom players to run their optical fiber cable all across their rails. You know that Indian railways is one of the biggest network in terms of the media assets. In terms of the budding engineers who have got some brilliant ideas from real estate to using it on the LBS and so on. In terms of business revenue that we are targeting is 10 billion per year is the pie that is available in the market. So whoever wants to be in the next building gets can Ion GAs. That is the kind of message it comes in and when we talk about India IT and so on obviously our redirect also raises up so I will touch up on China how they are putting it up we keep an Ioni and they are actually in the GAs growing at RMB 500 million here on here and when I talk about the apps business the RMB 3 billion yuan has been marked in GAs if you put in a statistical graph or say I would say that they are growing at 50 times what were they 10 years ago. So that means I don't know about the construction I don't know about the cheap labor I don't know about the mill products but I am sure that GAs is growing faster than us okay that's in China and if you come to the much more see if you see that I am also having a layer approach I went to global, US, China, India and I come to Karnataka so if I come to Karnataka one of the early adopters of GAs is probably Rajesh may be happy Andhra and Karnataka not Hyderabad and Bangalore so Karnataka has come down many of the new measures and how much of you know the road information systems which they called as exactly Web RIS and now Karnataka schools which under the they are going to map almost 74,000 Karnataka schools to be mapped and that will be monitored. The first one to go into this technology is Gujarat Government followed by Andhra and Karnataka so what I am basically trying to say is it all comes under the E-governance and try to put it in. To touch upon the verticals now having given all the statistics some questions on the government side which have been made to departments in the central and state governments in India which are sort of pushing the geospatial thrust so to speak. That's a good question basically any GAs primary funding comes from the land department that is one of the things properly creating a lot of trouble. If I say the reason example, Municipal Corporation of Delhi has come up with the automation of the building so the first department to go to you is the land department or the revenue department followed by the utilities that comes up. Now what the government has also come up little smarter is here is they are trying to get the central funding how many of you are aware of our APDRP program it's for the energy and utility program that they are trying to get the smart meters so that the revenue leakage from the electricity distribution is cut out. Nothing to do with Andhra as I have here. But basically what we are trying to put in here is where all we can do the plug-ins and check-ins so that we can get the revenue to the government. So the third one is the utilities framework which is coming to the picture. What does the NSDI do? How many of you are aware of NSDI? Very good. It's considered to be a Godfather for defining the standards. It's called national spatial data infrastructure. It's kind of a governing body fund under it if you want some things like that. It was first voted by department of space in 30th October 2000. They said something has to be done because that's what how many of you are aware of our Gagan program? Yeah, fairly good. So Gagan basically you know GPS you know GLONASS is coming to Gagan we want to try out our own individual GPS system because Uncle Sam shut down anytime the signal and we are black. So we don't want to get into that so that is one thing. And it was under the survey of India. So the general of survey of India has made us a chairman and then the head of NSDI was created he becomes a member secretary. Primarily cabinet approved the NSDI in 2005. Three things that they want to try is to bring the standards. They have about 16 nodal agencies, primarily government departments. So one of the gentlemen asked that where is the source of you getting the government data? So earlier what happened government departments were very conservative. You won't believe that when I asked for a utility map to plan for Hyderabad metro rail it's a hand scribble the diagram we say this is what the enterprise see where is that. I cannot test it on the ground so they just scribble the paper and say this is where the manhole is. So that is where so what NSDI tried to bring is they brought the survey of India map. They digitized all the survey of India map in different scale 50k, 100k, 50k, 25k, and they try to form a cluster out of it. And this data is shared across all the government agencies within India. So that is what NSDI first ensured. And to be precise the 2002 July strategy and action plan was endorsed and how many of you are aware that NSDI is the first government agencies which is OGC component put up the metastandards onto the web okay so not US, not UK, not any other country. So we can be very proud of. Probably we may be last in the game is what we thought but it's like a total you know end of the race kind of thing just leading them out and we also have OGC compliant web services with the 128 bit introduction and NIC National Informatics Centre has been endorsed to operate and give it up. So many of the people are asking at how to use the OIS to take it up it's a very simple procedure. If you have generated a map where you want to publish and take it to OGC just write a mail to ministry of defence and external affairs, get their stamp don't put the military establishment or some 52 you know things which you should not suppose to publish. Get it stamped and direct to you on the IP or you can publish it to NSDI. See the service settlement and land records department the service settlement and land records department in Karnataka the major digitalization exercise about 10 years or so before. So I just wanted to know what are the updates is that exercise complete? It's still complete see what happens that is what the difference between the private and public you know. Private for example any of the map provider want to publish they can publish they will not face the litigation but when you publish it from a government body you are an engineer or a missionary and you own that commitment you know if I say you have a land by 60 meter by 40 meter and suddenly you have a land of 45 meter then you will file a case of litigation. So where this process being delayed is one and second thing what happens there are many duplications which are into the system that means for the same pata for the same survey number there may be three original documents. So those are things which are delayed in the process. Is the gentleman from ISC there yeah I think he's done some work on the Karnataka Geoportal. So do you want to talk about that? Thank you sir myself Umesh and we have been working on Karnataka Geoportal. As Yannas sir told we have been implementing that survey of India portal framework into Karnataka Geoportal and it will be in 50k scale and soon we have been promised from survey of India that we will be putting one is to 10k maps also. Even I would like you to please check the Karnataka Geoportal.in website. You will be getting a lot of information there. Even 2001 to 2011 census data we are updating the 40% of the data has been updated in by the coming December by the end of this year we will be getting the 2011 data census data in that Geoportal website. So I would like you to please register it it's a pre-registration you can check complete data and you can give us a feedback so that we can make it much better in the future days. Thank you What is it? One is to 50k census data it's containing from GP level, village level and giving a village level. It contains 170 attribute level and functions. Thank you We take one hundred questions on this segment and go on to the next one. So does NSDI plan on standardizing Indian addresses very much? Anything you can share about that? There are two tough policies already has been made on the standards the idea here is first of all the mapping the base mapping is not ready so what they have taken us, they have taken up at the district level or a Mandel level of the mapping done. Now the next level is the physical survey that has to be carried over here so survey of India help has been solicited to generate the map so as in survey of India makes up on the 5000 scale actually it's talking about 1 is to 10,000 but survey of India is initiated at 1 is to 5000 scale at part of 1 is to 5000 scale so we are capturing the street numbers and road numbers which are going back to the data list. Is there a plan to standardize them? Yes. One last question we will go on to the next segment. So a couple of weeks ago I was trying to help my friend who works for Greenpeace so what they wanted was they wanted to set up geo-coded field codes to run campaigns so what they did was they approached the India Post and they signed up on the website and bought a CD which was shipped to this friend of mine with the CD to work they tried writing to them calling them the data doesn't work and it's not properly geo-coded thanks for it totally broken and then the same day I found this interesting company called geo-pincodes.org which sells this data just for India it's 65 euros for a 30-day license so you actually have this level of you can market this data and there are people who are willing to buy the data and Greenpeace bought a year-long license from the other organization so I mean I'm just wondering why the Government of National Mapping Agency is not coming up to like market this data there's no problem with this we pay a lot of tax we need more money to give a data tax it's quite unfortunate but I'll give a solution and then we'll go to the other side if there is any problem please reach out to your regional national informatics center they have every data backup in their place so you want anything, sensors election data, electoral water they have a copy of it even if you go to Saudi of India and take a digital map that may be one of the cheapest things that you can ever get in the government form so now the government will never look of making any monetary benefits selling that data you can buy this particular data you can use geo data for the postal codes whatever form it is it's not accurate to come right now the process is very cumbersome they sell it to the CD but why can't we use it on the website that's a separate debate in itself a separate topic honestly so we'll move on to the next segment this was really focused around the intersection of technology as we speak the geospatial space is quite evolving with a lot of changes there is so much of impact with mobility, there is so much of impact with the cloud aspect as well and then there's lots of people doing work trying to integrate you know data that is coming from telemetry and SCADA systems there's a huge big data piece as well that's sort of coming into play so I'd like to run this session with the panelists and take on you know the trends that is essentially the intersection of technology and the geospatial aspect of it as well so over to you guys, do you want to start so like a wishlist what is happening in the how are you seeing the convergence of different pieces of technology actually this goes back to I didn't catch your name to what visual you were saying you were talking about visual representations of data not just multiple things my current browsers and phones are really nice but I'm looking forward to to the future I'm looking forward to Google glasses I'm looking forward to having my card anywhere I have to go in what direction and which words to avoid that is interesting there's a whole bunch of people obviously happening in that not just Google but a whole bunch of people in the US and so on beyond that I keep talking to my friends at Flipkart and Mitra and they keep telling me about how they need better ways to get their telemetry guys faster to where they're actually delivering their t-shirts or watches or shoes and so on and they've done everything they could they've found the best people who know the place really well and so on but now they need a way to algorithmically with math figure out what the best way to get a guy to the destination is and that results in huge savings for them and in such a thin margin market solutions like this are what I think is going to start doing the competitive edge so I'll come to the I'll first start with a day in life for GS special right let's say the scenario is and this is more of you know few years old scenario and we're just starting to look at GS special as a platform under a layer downstream right and build apps on top of it so think about typical scenario is a family with let's say one or two kids right and morning you get a view wondering whether the milkman is going to come and arrive with the milk and what time they're going to arrive think about you can actually go click on your TV turn it on and you can actually track where the milk truck is right and then you can know it is going to arrive here in 32 minutes or whatever X minutes next up you have to go to a parent's teacher meeting you're running late you again check on your laptop how far is it which what is the traffic and then you jump in your car and the moment you turn on the GPS you turn on the car it automatically knows that this is where I need to go next right this is your next destination is this school now you've done the parents teacher meeting now you're just about the iPhone 5 or 6 or whenever you know the scenario comes alive and you say hey I'm going to go to UBSC and go to the Apple store and buy it right you buy it don't use the map here yet on the iPhone still keep using what you are using till now and you say hey I'm going to go eat some good food I go to coconut grove I want to eat some opum and paratha and you basically you punch in that address or the name it gives you not just the address and the direction but also gives you the reviews it also gives you recommendations or what so think about the data and application and layers that you can put on the top of this data right you got good food I want to get home take a good nice hot shower by the time you get home your car is intelligent enough that based on the GIS data based on the geospatial data it knows how much time it will take for you to now get home and it can actually turn on the heater at home so when you get home can actually get hot water right and by the way you are now in a mood for there is a my friends have been talking about a black label limited edition that is just arrived so that's a double black so now you want to pick that up you punch that in it will actually take you on the way these are the four shops this shop has this deal going on and you can go pick that up so think about this is kind of that application that you can build on top of GIS now what does it from a technology to strip over technology what did we talked about mobility cross device connectedness cloud big data and most of all coming back geospatial is the data trustworthiness because you don't want to end up planning your day or running through your day on the bad data so that's essentially you know my issue on technology thanks for that wonderful look ahead scenarios of how things could play out you talked about switching on the TV and knowing that your milk man is 42 minutes away if only there was technology to get my milk man to wake up an hour earlier that would be nice yeah I think a lot of what I just said on the intent I think we need to kind of move from trying to figure out what the user asked us to go to model where we understand the user so much that we are able to that's really something that GIS can help the second thing I'd like to comment is that the cloudification of the whole GIS space data moving out of the cloud will make it much more easier for people to kind of share it across platforms and then you can actually do some really interesting joins across different types of data and maybe that actually business model where people actually make a living selling the appropriate data on the cloud one good thing being a system integrator when looking at this product companies you know it's like having a wife and a girlfriend they can have a taste of all Microsoft is but you cannot own anything so coming back to the technology things what we are looking at is coming like a digital convergence kind of a concept you know for example the camera business of what is a company is killed by a mobile business because the moment the mobile games put their camera in their digital Polaroid camera business got out so the competition didn't come from the same industry came from totally tangential way so the same way this is driving the technology going forward nowadays how many people are owning an in-car navigation device that is going to become obsolete because your own smartphones is going to act like an in-car navigation devices so all in that platform is going to either switch or going to be killed over okay so when I look at it the cloud and the big data convergence is making a lot of sense for example recently they implemented with the Singapore government land authority initially each of the government departments say that suppose you are a public works department you ask for 10 million utility department you ask for 15 million what these people end up doing is out of this 10 million 15 million Christine was mentioning about the data things you end up in creating data for your own business so what we went and told to the government they say look Singapore roads are not going to change overnight tomorrow morning remains the same so why you as a government agencies cannot pull out and bring a base layers and expose that as a web service government departments including citizens can use for own apps and applications that data very well so the convergence of data happen so of course we had a challenge of a big data hosted in the cloud 632 layers have been hosted and till now government of Singapore has reaped 11.2 million dollars of savings in one year okay and this is with a thousand concurrent users okay and with the refresh rate of 630 layers coming up in less than 20 seconds Google search appliance for application use to article special this is the first government NSDI private cloud hybrid cloud practicing has put up on the web so this is where I feel that trend is going on and the convergence is going to make our day okay so we probably take a couple of questions at this stage before we close our final text so about convergence right yeah so about convergence there are so many more attributes that can be exposed to the map it need not be all about shopping and iPhone 5 and all that for example groundwater levels or water salinity or addedness or fertility of soil or weather conditions and all these data from actual data from earth that keeps on changing and we can see such kind of data be available through maps the real time data is already available for agricultural, aspatati, maindra, industrial and forming equipment whoever buys the swaraj tractor we are loading this application and giving to them this is for information sake other things you can go to public utilities board.cov.sg where www.pub.gov.sg you can go to the check water level, flood level through this camera system and real time CCTV cameras it's 24x7 monitor and what we can most time to show that we have a real time streaming coming in over the cloud and it's totally free we are calling it that's right yeah essentially it was an example where it's actually pretty cool it was a specific instance I am Nikhil I am here so I will try to add to what Kashmir just mentioned lifestyle and the other things are ok but when we talk in India's context we are still dealing with what used to be called as third world problems and right now there are two, I can count at least two more things which have been added to the basic necessities other than food, clothing etc there is electricity and internet and that is actually available in most parts of India so it's more of an opinion than a question that recently what motivated me with this thought was the recent project which the government has taken up was Aadhaar which is the world's largest biometric identity platform and it can be coupled that data with the geospatial data to suggest urban planning to suggest where we could use this data to probably plan the traffic so right now we saw that there is a live stream available which takes the pictures of traffic signals can we make some image processing on top of it get the traffic density at particular squares and then suggest predictive analysis on that to actually install more traffic signals at particular checkpoints stuff like that so one curious question is something like that is already being planned at the government level is the Aadhaar data just being stored or is it going to be consumed in a fashion which actually encourages people to come up with some recommendations or some predictions based on time series analysis of that data or something like that so I would really like to know something about it as part of Aadhaar we are also involved from Andhra Satyan and you showed the cards in the backcoding I would like to take this question in two parts one is asking can you use a predictive analysis and use it up let me take it up before Aadhaar of the EMRI emergency management we are using this predictive analysis already input into it suppose there is a fire which has happened in a slum area typically you don't know what is the damage that has occurred so you have to dispatch and the paramedics team immediately to this part to rescue people and do a first aid so what it does is it immediately takes the census data of the slum which is of course has been given projection in the last year in 2001 so we are separated to be nearest close by maybe some of the other plus or minus that will determine the number of ambulances that has to reach this part and not only ambulances suppose there are households one household may have 5 people say for example and how many first aid equipments that has to be taken for example number of cotton band ages injections is that and everything that is predictively done and dispatched so to answer the question first of all yes we are doing it but not the scale that what we are looking at but I am always happy that the first aid has been taken and scaled up. The second question is other what we have more try to look it up is Allah is used for data validation and consolidation purposes more than video analysis we see the basic problem in India is that one source of truth what we have been discussing is nowhere available and there is no single owner who will say that yes these are the data accurate up to 60 percent 70 percent so the intention of Allah is something different but once the primary objective is fulfilled the government initiative are at the ministerial level the objectives to go on are just there. Okay I think that would be the last question we need to wrap up I would request the panelists to sort of give one futuristic take about 10 years down what do you see fundamentally changing or what do you believe might change that might change the experience of either consumers or enterprise in the geospatial space. Shall we start with you? Interesting phenomenon over the last couple of years has been this second phenomenon there are also niche subcultures across the world which are also involved in body modification stuff I see some subcultures of convergence happening there I see people walking around on this day I see people walking around where the phone is probably a little tiny or fits in your wallet and is recording your data on a non-stop basis private something so on and so on but yeah I don't see Star Trek in the future that's what we're going for and yeah it's going to be exciting I'm going to love being a part of it I'll be happy if my phone battery lasts more than 10 years 10 years 10 major problems yeah absolutely a replicator for it yeah so starting my future scenario I think Google last is only faster but I think perhaps 10 years from now we'll have so much data and we'll know so much about the real world that we'll actually have a completely immersive experience that will actually be there 10 years I will be Google eyes or something I think the convergence of these different industries that we think of at some point GPS was a different industry now GPS become part of the phone and eventually every phone including this very generic starter level phones would have those and I agree with something that he said there will be so much data that it's just processing that data and making sense out of it would be amazing amount of work but once we have that the mapping of all that is going to be great so I think the convergence is really what is going to get to the next level of experience and where can actually pretty much go settings go to a business sitting at home and simulate the entire experience is what I think would happen I would say that more of a predictive modeling and digital convergence is a theme is going to be one of the project you're working on is a digital soldier that means how you can use it for health care and how you reduce the casualty more on this we are also working on a humanoid robot who can do a complex operations kind of a thing so mechatronics and telematics which we are trying out with Intel and other R&D labs so there is no more data equipment analysis it's going to be cut together as a future debt I was just thinking what might I take with me I was just laughing as I was thinking taking on the battery the way I visualized 10 years from now technology will be so sophisticated that you will be struggling to figure out how to use it and when to use it more importantly imagine 10 years down and then your gadget would be nice screens and you know that everything in the geospatial stuff will be on your mobile it already is everything else you could dream of would be there and then it will start heating up and explode so the service providers will actually have been shown as a ghost that your phones can be changed as fast as they can but I'm just kidding thanks so much for making this an interesting evening and the panelists around if you have any questions and thanks for your patience