 And I'm here from Nexra. Actually, the presentations that occurred before actually provided great introduction to what I'm hoping to speak a little bit about more today. So I came to India with the idea of trying to help our company locate our customers on a more efficient manner. We have about 25,000 customers at the moment, and we want to be able to locate them on a house-by-house basis. But right now, as we know, the app systems really aren't so efficient. And in the city that we're based out of Inhooli, this information is very inconsistent. And a lot of areas don't even have street names, just the basic information. So actually, when I arrived to India, I was interested in mapping, but I didn't really quite know how complicated or how interesting this whole project might be for me. I wasn't even told an address of where I was located because, well, it's actually not even useful. And I found out that after riding here, the auto driver, taxi driver that I was to give an address to would probably even have to ask five, six, seven different people before I even get to the location. So just having someone who knows that local knowledge is actually much more useful. It's getting you to that address than trying to crowdsource information on a constant basis. So what I'm hoping to provide here is some of the experiences that we've been facing on the field, as on a day to day basis, we're hoping to at least get to 500 customers a day. So what coming out of our report is an extra. A little background about us, we provide SMS messages to our customers of when their water supply is going to start. And we, people get water only every three to four days. So that leaves a lot of wait time for when your water is going to arrive. Some people may not be able to go to work because they're always waiting to collect the water when they need to. So what we're able to do is we can reduce that wait time to down to as little as 30 minutes so they can just go home, collect the water, and resume their daily business. At the moment, the way that we are collecting our customers and figuring out where they are is going house to house and gathering their address information. So what we've been doing is we go to our house, we ask them what their address is, they tell us all the relevant information that they feel might get us there in the future. And then we enter that data into our system, provide their SMS message through our address system. And then we actually go back to build them. But the problem with going back to build them is, well their address doesn't actually translate if we give it to someone else who's never been to their house before. So you may end up going to one house, two houses, three houses until you finally realize, oh, that's where they were the whole time. So what we're hoping to do is kind of understand better what this local knowledge is and find a way to use it to make more standardized system, which we all know is a little bit difficult, but somehow use that to eventually navigate to our customers in a more efficient manner. So as I was mentioning earlier, we want to locate 500 customers a day. We have over 25,000 customers right now in Boobly. In order to be efficient and to work as best with our resources and time, we want to get to 500 days to build them. And second, local knowledge. We know that the knowledge is there, so how can we get this, put it on a map and use some sort of navigation system to get there? Finally, it kind of seems at the moment that maps are geared towards more larger cities. I had asked earlier if Locazi might consider working with smaller cities, because we feel there's a lot of potential products, markets that could be available to people that are just a lot of cities and their address might not be the most useful, but if we can locate them on a house-by-house basis, then maybe we can provide more products in these areas. So starting with the pros, sure you all find mapping pretty useful, or else you probably wouldn't even be here. So as is to new markets, new products, e-commerce, helping municipal utilities bill our customers more efficiently. And as I was saying, work in efficiency. If we can place our customers onto the maps, we can navigate to them more efficiently. We can see where they are and just work towards if we will improve in general. And finally, because we can, people have some sort of address in their mind, and it's just a matter of collecting this information, figuring out how we can put that on a map and use it to our own advantage. All cons associated with mapping at the moment, we found the first was to be map or literacy. In Hoogley, not everyone is as tech savvy and interested in mapping, or they don't really see the need for it. But if they were placed on a map, then they could provide a lot of opportunities and different services that they could have. So finding the medium of how we can make our maps in a simplified manner, and in a way that others can read it as well and use it, is one of the key goals that we're trying to get to. And emphasize data, which I'll just elaborate a little bit more on. So this is Alston, it's a neighborhood in Boston. And as you can see, the area of boundaries and emphasize data is a problem everywhere. A great percentage of people know kind of roughly where they are, but once you get to the outskirts of the boundaries, it kind of becomes a little bit vague. So different residents, even though they might be located similar to each other, might not be associated with the same area name or colony name. Contradictive data, we've discussed this a little bit so far. We found that in Hoogley, you'd assume that most people might have a plot number or a house name, but there are often times people may have lived in the area for so long that this just isn't something that they even care about. They're able to get the supplies they need, they can live on a day-to-day basis without even having any sort of address. But for our purposes, to be able to provide service, like Neck Shop does, address information would be very useful. Second is a lack of city planning. What I'm trying to say with this is maybe in 2003, all the houses may have had some kind of chronological order, one through five. And just a couple of years later, with more development, all of a sudden, house three isn't located next to house six. House 34 is no longer next to 35, but 72. So how exactly do you find those missing houses? Do you survey an entire area before you can locate them or is there some sort of application that we can create out of this? So maybe it's a generation gap, maybe we need to think about this problem in a different way. People may have come to a city at different times and associate with different landmarks because of when they go to the area. It might be the older building that they associate as landmark or the new gym that just opened. So the next thing that it has done is we haven't exactly come to a solution to this problem, but we're just trying to tablet piece by piece as efficiently as we can. Six to nine months to locate all of our customers might be a little bit too long for us so we can keep our business going, but we kind of tested out a couple of different prototypes so far. So we're collecting address information from our 25,000 customers that most of them associate with the city. Some sort of hub area, parking system, if you give this piece of information to our builders, maybe eventually they can have some sort of systematic way of getting to these households and zoning in on them barely actually applies on the field, it was a little bit messier. Some areas and areas associated with different colonies, they had different names for the same place and it was just a big mess. We were trying to figure out what was going on. So we tried to de-organize our system. Everyone associated with some sort of landmark so we thought maybe if instead we group these households that pick the petro pump for example as their landmark then we can just say they're all located behind the petro pump. We don't need to know what their exact cross street is. Once you can find that landmark then that's enough for us. There were the cases where there just wasn't a landmark at all. Actually that sign up there, I was just happy that there wasn't a sign at all because usually there's just nothing. So we thought maybe we'll just create our own landmarks. So I can say that the houses located between landmarks one and three are to your right. Four and one, our landmarks just was very, very, maybe as we recruit our customers we can just do it in a different way. Instead of a little contact. You're a long life. The issue with that is we need to locate houses and so getting landscape and launching cornets for every single house is just a very time consuming. So we're trying to just protect very rapidly and that's why he kind of has a mission to add. So we had an area of all and we just assumed he'll be the main point of contact and he can tell us the boundaries of where we're trying to go. So for example, this is an area who gets water supplied between 7 p.m. and 7 p.m. So we put those boundaries out. We put this information on the smartphone and went out there and recruited our customers assuming that all of them in this boundary are gonna get water at the same time. But of course this didn't really happen. So these customers got water at 7 a.m. instead and they were just contradicting information. Who actually knows this information? Is it evolving? Is it the customers? Is it the city? And then we decided maybe mapping isn't the most useful tool for us. Maybe we need to think outside of the box. Being able to locate one customer in an area might be a little more useful than locating every single customer in the area. So I guess, well I'm sorry, I went through that a little. I just wanted to present some of the problems that we've been having on the field and see if there's any ideas that we can develop out of this and how we can get to the customers more efficiently. It's a very crude method that we're using at the moment to locate our customers. But hopefully this information's a little bit useful to what the complications are actually on the field. We found that unplanned development leads to a little confusing maps. And once we have these confusing maps, development and building our cities off of this is still a little complicated. So if we can somehow create maps that help development in the future, then I think we'll provide more opportunities in general. I just have a short. I just wanted to focus on this one. We couldn't do any of this without GIS. For the first time we have a 3D building model for the whole city. That's possible because the whole region now has LiDAR data. LiDAR systems, the satellite aircraft beams down to the earth. The beam bounces back up. They gather information based on the return of objects on the ground. So you end up with a very good terrain model. At laser point accuracy, literally, we can take our information and actually start to visualize things three-dimensionally. Instead of worrying about what if a proposed building threw a shadow all over the park. We did a shadow analysis using the GIS to calculate that it doesn't really impact the park. People went out and measured it and they were within two feet, I believe. The city says you can't build anything that's gonna block the view of the mountain from up on the hills. With the GIS, we're able to do satellite evaluation to improve what they've proposed. Buildings would not block the view of the mountain. And that made people happy. We get maps that actually work. We get influenced the way that development is done and eventually, hopefully these maps are even more useful and we're able to collect information that's actually just more useful to start off with. So I guess for that, I just wanted to put out some questions there and see if any ideas and thoughts might arise. Just to understand your need. Sure. Do you need the address and location only for building collection? Is it also for providing the information that water is coming out? It's for both. It's for both. And that's where we're kind of getting some sort of an issue because the people that actually get water at the same time aren't necessarily associated with one area. So there's a lot of hope in that that's going on. So being able to have different layers on a map is what we're trying to do. Because if it's only for building, you can probably bypass the problem. I mean, you can partner with, for example, Department of the Post. I mean, they're related to the big powers. Yeah, yeah, we have to consider that. Or relatively, you can think of mobile-based payment solutions. Yeah, those are definitely things to think of. You could also look at some address-cleaning solutions. So it's kind of something like, you can pass an address with a comma or a separator and then standardize those addresses using some tools. So if you have one reference point, you can get some other addresses. And then, for instance, it could be some people write as NG road in Bangalore. Some others write as Mahatma Gandhi road. So there are ways of standardizing that and then you can do a geocoding on them. I think we probably have some things that we can follow the process. Yeah, let's hope you just keep going. Are these about this afterwards as well? And one of the things that Yaza Hort was trying to get the lat-long of the place, but he said that that's quite expensive. Yeah, I guess we're just trying to, as quick as a solution as we can at the moment. And that's the thing. Actually, I don't think it is so difficult. I mean, you just have to, you don't drive through it, probably have an audio tag with the proper, I mean, the audio tag, which is like geostag. And then you could go back and just do a transcription. So you don't really have to go there, stop at each and every house and say that. Just drive through it. Just drive through it. I can go there. So that's where timestang goes to probably. Instantly in geostag. Okay, yeah, thank you. And if you're already going to the houses, collect the address. You might just as soon get the lat-long. Instead of the address. Yeah. Long-the address. Okay, transport basically are all tools and most of the time many people will be using that. Probably we can use those facilities to get to address whatever we want. More than we frequently go in and out. But yes, part of our problem is even if we can collect this, you know, lat-long information put onto the maps, not many people know how to use these math information. So now we're facing a problem of how do you translate this and put it into a format that some people aren't familiar with maps and actually use to get to their location. So one thing we are considering is how may be printouts are using SMS to get to these locations. But this is still very basic thoughts and we haven't really developed anything out of it yet. One of the things that we have done, I mean, with respect to what we do is reverse geofort. So whenever somebody sends an SMS with a particular address, we try to pinpoint to a particular location that works quite well with our database. So what probably the good thing is if you have a pretty good map database in the back end and then you could provide an address in whatever fashion, whether it's MG Road or Mar-Malan or whatever, depending on the level of accuracy. And obviously it has to be custom designed because you have the extract and where the house number comes, where the road comes, where the EA comes, and then probably get a rough estimate of where exactly this lat-long should be. One interesting thing on that, see, we tried something like that with a place called Maruti Nagar, somewhere near VT. And then actually the reverse geofort in Maruti Nagar, geoforter at some of the Maruti Nagar features here actually. That's why I'm saying the search has to be intelligent enough. So when you say Maruti Nagar, you obviously say Maruti Nagar or Mar, let's say that area. So the more information that you give to, the more data that you give to the search engine with respect to the specificness, the better the result comes out to. For example, if you just- It also appears to be an IOS 6 problem, you know? You're asking for- Yeah, I think, yeah. We should ask them to get in touch with us. We're collecting information from customers. We ask, usually, what is your landmark? And oftentimes we'll get a building, or a room, or a school, or just maybe a tree. But there are oftentimes when people will give us a road also, and then you wonder, if this is your landmark, you could be located anywhere along this route. And it's just being able to gather information from people that is useful to us, is something that we're trying to figure out as a lot. I actually had a different question. You said, you know, you're on the other part of your business, so there's telling people when their water is there. How are you actually collecting that location information? Are you censored data? How do you- We're actually partnered with a water board. So we got this information from the development, the area development, and when they open a pipe, they call it to our IVR system, and that's what sends out the SMS message. Yeah, because really, you know, we correlate, you know, you need to know which pipe someone is on, not just their left, but along for that part, right? Yeah, and there are issues with having two pipe connections, or they tap into another pipe connection. So, yeah, I've also realized on day-by-day basis, I just try to just capture all of that here, and it's like- The size of smaller domes are mixed up as a solution. So, what's that? The size of smaller domes as a operating system What I have seen in villages is that typically this information that water is coming just flows, I mean, a lot of more. Yeah. I've collected a lot of information. So, there, I would think that every household will subscribe to this service to get this information. On the other hand, I think it will be more useful in cities where it's not much of a neighborhood in that sense, and of course it's not like a horizontal flow, but more like vertically, but laser. So, we don't know whose connection is there and what and so on, right? Yeah. So, I wonder like, how is the response and is it like every house subscribed to US service? Yeah, I'd say it's about, maybe about, 70% of the people that would try to recruit actually use our service. They do find it useful. In Cougli, there's quite a wide variety of middle income, high income, low income. And we get different responses from different areas. We're actually still trying to conduct more research to figure out why this is the case. And how much does the service cost? It's 10 rupees per month.