Ian White, the CEO of Urban Mapping, makes his living collecting and selling geo data. For next week’s Where 2.0 has put together a panel of government mapping agencies (the UK’s Ordnance Survey and the US’s Census Department) and community-built mapping projects (Open Street Map and Waze). Crowdsourced projects like Waze and Open Street Map have forced civic agencies to reconsider their licensing. They have similarly encouraged larger companies like Google, NAVTEQ and Tele Atlas to implement their own crowdsourcing platforms (like Google Mapmaker and Tele Atlas’ MapShare). Ian and his panelists will discuss all of this in Base Map 2.0 on Thursday at Where 2.0 – you can consider their conversation as Part 1 of the panel.
Ian gathered the panelists (listed below) for a pre-conference phone call. Here’s Ian’s descriptions of the panelists:
- Peter ter Haar – the Head of Products for the Ordnance Survey which is the UK’s national mapping agency.
- Tim Trainor – the Head of Geography at the Bureau of Census within the Department of Commerce in the US Government. His group is responsible for TIGER which is a lot of the line geometry as the US as well as the master addressing file, a lot of improvements there in the 2010 Census.
- Steve Coast – the Founder of OpenStreetMap and also Cofounder of the commercial entity CloudMade whose role is to commercialize the OpenStreetMap data, although I believe he’s speaking primarily with the OpenStreetMap hat on today.
- Noam Bardin – the CEO of Waze, a company which effectively offers a crowd source navigable base map with an emphasis on real-time data.
The edited transcript is after the jump:
Ian White: Okay. Thank you all for joining us. This is a preview podcast or streaming or transcribed version of a session to take place at the O’Reilly Where 2.0 Conference in San Jose on April 1st at 2:00 p.m.. The panel is Base Map 2.0 or Base Map Revisited. My name is Ian White. I am founder of Urban Mapping, moderating the panel. And I’m very excited to have gathered these four very well-regarded individuals together virtually today and then in the flesh a few weeks from now. And we’re looking forward to having part two of this in the session on April 1st.
Peter, the Survey is going through quite, let’s say, a relative tumultuous period. There’s been political pressure on the Survey to rethink its business model. By any estimation, there are three possible courses of direction that the Survey might take. We have just concluded as of yesterday what is called the consultation period, which is a period of time for the public to provide input. And it seems like there are three general directions the Survey could go. One, stay the course; change nothing. Two, make all Survey maps and digital data available for free reuse under some sort of Creative Commons-style license. Or a third direction could be make a subset of the OS’s digital maps available for free reuse on some sort of Creative Commons-style license. Now the Survey is a bit of an interesting entity in that it is known as a trading fund so it is able to generate revenue from public sector and private sector and has a bit of a reputation for having a stranglehold on development of commercial-based map products in the UK. Peter, so the first question for you is can you comment on the state of affairs relative to the consultation period and what might shake out from that and how what you say on this recording might be different than April 1st when a few more weeks have elapsed?
Peter ter Haar: Yes, because it’s good to [inaudible] the question. It’s good to understand that we’re recording this on the 18th, which is one day after the consultation. And obviously, Survey is not the organization that is running the consultation themselves. It is the UK government that is doing that. And, therefore, I don’t think I should anticipate or I should, yeah, anticipate on the outcome of the consultation. There’s one thing that I would like to qualify what you said. You stated Survey as a trading fund is able to generate revenue from the base map, from the IP. I would like to qualify that as Survey has to generate revenue from the IP because it is the only source of income that Survey has. So if we would not be generating revenues from the base map, then we would not be able to maintain the base map. So I think that is an important qualification to make. So, therefore, everything that we do also as part of the outcome of the consultation will have to be constructed in such a way that the creation and maintenance of the base mapping of Great Britain is a sustainable activity in some way or form because we do underpin quite significant parts of government but also of private enterprise.
Ian White: Okay. Noam, Waze as a crowd-sourced product is effectively based on the goodwill of the community. We know that from the experience in Wikipedia you have a very small number of users responsible for a large percentage of the contributions. So the Wiki approach can effectively give you a lot of hopes and fears all baked into one. Can you comment a little bit on how Waze sees these users? And I understand that there are kind of more active and passive users in terms of the edits and contributions that they’re making. And how has this growth evolved over time? And specifically, I’m interested in the US versus Israel which is your home country and the of the world?
Noam Bardin: All right. So at Waze, we have a free mobile app which the average user just uses in their daily driving. You get free turn-by-turn navigation. And they get real-time traffic and a lot of community and social features. And 99 percent of our users just drive around, the contribution they would do from the client would be their house numbers to mark and problems you can map, things like that. The one percent who are really active are those who actually go online and edit the map and record new roads and help maintain it. But every user who just uses the map and 99 percent are actually contributing, so collecting their data. This is a little different than a Wikipedia model where 99 percent just consume the data and one percent actually create it. In our case, 99 percent are contributing data; one percent edit that data or manage it or help control it. And so in that sense, I think it’s a little different. At the same time, having a consumer-facing application really makes it appealing to a much larger subset. And part of our challenge going forward is to engage a larger and larger percentage of them in contributing to the application. So I think when you look at our growth rate where we just passed 650,000 users. We just passed the point where Israel was the leading country. So today, almost 60 percent of our users are not from Israel. We began in Israel in the beginning of ’09. And we opened up to the world in November [inaudible] in August. So I think in that sense, when we look at the community challenge here, the community is very involved. But different parts of the community do different things. And allowing each part to do something else really keeps them engaged and doing what is right for them.
Ian White: Okay. Tim, from the perspective of US Census and just by way of a ten-second overview, US Census produces TIGER which is essentially the base map of the United States at a street level detail, primarily used for enumeration but also has many derivative uses. For instance, forming the basis of other commercial base maps in the case of possibly NAVTEQ or possibly OpenStreetMap, of course much more annotated by these parties. Tim, can you comment just for this is kind of a public service message in about 30 seconds on with the upcoming census how the master address file and TIGER’s line work will be different or I’d love to say actually improved and made available to any interested parties and what rough timeline that will follow?
Tim Trainor: Well, thanks. I guess the first thing if I could just put a plug in and say for all of those in the US that get a form, we’re hoping you fill it out and send it back and save us money. And it’s good for your communities. Okay. That said, we actually maintain a master address file on an ongoing basis to support our surveys. And we use it to support the taking of the census for the decennial census. Along with that, in addition to needing the addresses to do mail out of questionnaires for surveys and censuses, we also need to maintain the street network so that we can do things like geo coding, so that we can get enumerators to locations where they have to go out and conduct work on the ground. And so in order for us to do that, we’ve employed the use of newer technologies for the 2010 Census that we’ve never done before. We’re using GPS or we have used GPS technology for the 2010 Census. And in order to use the GPS, it really required that we realign the road network in TIGER so that it was more positionally accurate, so that we could assure that when we collected those GPS locations of housing units that they actually fell within the correct census block which then allowed for the correct allocation of that population within that collection block. So it was the need for from an operational perspective and the use of technology that drove the need to improve the basic road content and the road accuracy.
Ian White: Okay. With the use of GPS and other technologies, do you anticipate the availability of data to be more rapid than in years past or is it simply a matter of maintaining quality and, therefore, that doesn’t necessarily change the new release dates?
Tim Trainor: Over the couple of years leading up to 2010, we were actually doing two releases of our TIGER data a year. And that was really to get back to the partners, those who had provided data to us. We wanted to get back data to them so that they could —
Ian White: And this is primarily other government agencies or departments?
Tim Trainor: That’s right. That’s right. These would be local and state governments, tribal governments and so forth. That’s right. Since we’re now in the census realm and we’re trying to manage the volume of work we have, we’re now back to a once a year delivery of the TIGER data. But I think that is largely going to be driven by the need from the use of the surveys as well as the need for the use of the data itself. We can increase that number if we had to.
Ian White: Steve Coast, OpenStreetMap, can you give some let’s call it “advice” to know I mean how to manage a community of users in a geographic domain where you’re looking to generally support the initiative of the commercial product but recognizing also in the case of OpenStreetMap that part of the success has been because of your hands-off approach?
Steve Coast: Yeah, I mean you just nailed it. It’s the more freedom, the better. The more tools you give to the users, the more open your API, the more open the licenses, the more contributions you’re going to get; the more contributions you’re going to get from unexpected places. And that’s true across any crowd source project pretty much. The more you try to constrain your users, the less interest they’re going to have, the less data and community and so on you’re going to get out of it. I think that’s really the key.
Ian White: Noam, could you comment a bit? When we’ve discussed it previously, you’ve said that the opportunity to work with some national mapping providers has only emerged with the advent of kind of the gorilla not in the room. So this is Google with its kind of go it alone base map approach and commercial entities or national mapping entities looking to you as a viable alternative. So how has Google helped or hindered in terms of ways of ultimately offering a free navigation-based handheld product?
Noam Bardin: One of the advantages we have because we keep the data under license and control the terms of use of license is that we can partner with commercial entities. So we can preserve their licensing model and we can create revenue opportunities for them; they can create for us. And also, we can update their data. But to make this happen, they need to have an incentive to work with us. And overall, up until the point when Google came out and announced their initiative, we were seen as the enemy by local map providers because we were giving it away for free navigation-wise to consumers and updating it. And they saw us as competitive. When Google announced what they were doing, suddenly we became the ally. And so all over the world, we’re getting approached by map providers and owners of base maps on different quality levels who basically have the challenge of how do I keep updating my map fast enough, add real-time layers when the requirement of the navigation center keeps going up in terms of how fast you’re updating your map. Well, the cost structure is definitely not going down. And community is really the way to do it and crowd sourcing, too. And so this has allowed us to have many commercial partnerships [inaudible] Latin America. And if you look into our map in Latin America, most countries out there [inaudible] Columbia, Peru, Argentina, Chile, probably have Brazil and Mexico by the end of the year, this commercial-grade maps in there which we’ve got base maps from our partner there, Location World. And then we update these maps and can both share in the data that’s being created as well as in revenue that’s being derived from this data. And so this has been a great opportunity for us. And all over the world, we’re partnering with the local players. And at the end of the day, maps are an extremely local activity. And we are a global player that provides a slice of the map infrastructure.
Ian White: Any idea why the countries you just mentioned I think most or all of them are Latin/South America? Is it a coincidence or do they share something?
Noam Bardin: Our first partner is Latin America. But we’re negotiating similar relationships in Asia, in Europe and really around the world. If you own a base map today, you’re basically threatened with how do I lower my cost structure dramatically. And this is sort of what we do come in and help out.
Ian White: Steve, with OpenStreetMap, do you see something similar in other parts of the world where you could help kind of or a national mapping agency could kind of defray or spread their costs across kind of a wiki kind of geo user base to help maintain that?
Steve Coast: Yeah. I mean I wouldn’t say that Waze is the only sort of company and entity in the world that’s able to partner with commercial entities, right? I mean it’s not like OpenStreetMap hasn’t done that or CloudMade hasn’t done that. And everywhere we do, we see massive increases in the amounts of data, both commercial data but then obviously places like Haiti which is what I think you’re trying to get at where straight after an earthquake everybody contributes and uses OpenStreetMap as the central repository of all map data. And it leads to just incredible quality of stuff because we just get out of the way and let people add exactly what they want. And today, you see OpenStreetMap in the disaster areas. It’s not only the default map; it’s usually also the only map that people can use because there simply isn’t map data.
Ian White: Right. So just by way of editorial clarification for those listening or reading, in the wake of the Haiti earthquake and, I think, the Chilean earthquake, weeks back there was a groundswell of support of individuals and NGOs and others contributing base map a la OpenStreetMap as opposed to other map platforms that might be out there and could’ve supported these initiatives and so Steve was commenting on that. Steve, do you have anything else you could comment on relative to kind of response time, other map platforms? I know Google has Map Maker which has a similar kind of aim as OpenStreetMap. Although, clearly there are terms of service and there is a clear business model behind it, meaning commercial use is quite restrictive. Any perspective there?
Steve Coast: I think people like Google and others, they have a clear company goal of getting out of revenue fees being paid to people like [inaudible]. So their primary goal is that they want to replace those guys. So that means road maps with rich points of interest and a few other things as the first step. So helping out places like Haiti, while nice, is incidental and a bit outside the scope of exactly what they’re trying to do with Map Maker. And also slightly outside of the scope because Map Maker is a lot more constrained than OpenStreetMap; it would be hard for people to use it in the same way, both because the tools are very different and also because the license just doesn’t allow a lot of people to use the data. So I think there’s clear advantages.
Ian White: To continue on the doom and gloom topic, Peter, could you comment on the Survey and the role that they play in any kind of rapid response? I don’t necessarily mean production of a paper map based on data that’s maintained, but if there’s some kind of incident that’s called a natural disaster and I, pardon, can’t think of any in the UK in recent memory, meaning the last few months, but the role that your organization or other kind of mapping entities or then the UK Government might support in terms of public interest and helping to very quickly respond in kind of a geographic sense?
Peter ter Haar: Well, one of the examples was obviously when two-and-a-half-years ago some of the rivers decided not to stay within their normal course. In the summer of 2007, in Carlisle, we had the problem that the emergency center of the local authority was housed in the basement in a building in one of the lowest areas of town. Obviously, that flooded and all of the emergency response almost came to a halt because people didn’t have the mapping anymore. It didn’t really feel like Haiti, but it was quite grim. We have something called mapping for emergencies. And we have some of our staff then go in and do the analysis on behalf of the local authority and of the emergency services there to make sure that we knew which houses to evacuate; which roads actually were safe to drive on and also to look at what potentially other areas could flood if specific dams would break, et cetera. It’s that kind of fairly low-level response that we do quite regularly for natural disasters, but also sometimes for let’s say to support security services or the police in all kinds of incidents.
Ian White: Okay. Tim, from the perspective of US Census or possibly other agencies within government Katrina was obviously a pretty well-known national incident in terms of government response. Can you discuss briefly the role that your organization was able to play? And I’m most interested in the time to respond and what was kind of custom as opposed to a traditional months to weeks development cycle of turning out data to reducing that to much shorter intervals?
Tim Trainor: Yeah. The fact that we have the data sitting here, it’s a very valuable resource. So when we got the call from the administration to try to get some information out from a demographic perspective, you know, what’s the condition on the ground for the type of population, the age, the sex, where the senior citizens were, the children, those kinds of things. We were able to generate data from TIGER, maps from TIGER to show the characteristics of the population, get them posted to our website very quickly and get them out there. And then that generated more interest. And we then produced more products as a result of those questions coming in. And so it was a very quick, timely response.
Ian White: Okay. Noam, can you comment on — I don’t think this is something that Waze has done, but to what degree do you see something like Waze potentially being used as a tool with potentially hundreds of thousands of probes out in the field?
Noam Bardin: Well, we haven’t really had an incident where it made sense to use us. But we’ve had similar response on the commercial side, meaning commercial map customers who decided to purchase our map. And one of the compelling features for them was the fact that they could go online whenever they wanted, update things that are important to their customers and the next data that we compiled into the map. That’s something obviously you can’t get from a Tele Atlas or a NAVTEQ or anyone else. And that goes back to the real-time side. Now if you have a customer who has an entrance for their truck into their loading dock, it’s critical for them. There’s no way they’re Tele Atlas to add that to their base map. But for us, [inaudible]. They went in online, added information; the next day, we had it. And so I think it’s similar in that sense of how fast you actually build the map. So a crisis situation is the same thing, right? It’s how fast can you build and share this data? So we haven’t had a crisis situation where either a community or data could help. Of course, we would make it available if it could. But we have seen it in commercial entities.
Ian White: Okay. We’re coming up on time. I wanted to keep this to 25 minutes, so I do want to conclude.
Peter ter Haar: Absolutely. Go to Mapaction.org and see what you can do to support emergency mapping. That’s all I wanted to say, Map Action. Very good.
Ian White: Okay. Thank you, Peter. And actually, that Map Action’s not to be confused with the patented Map Action technology that Urban Mapping uses in its printed map product. Final question: Which of the entities in the room would you be interested in partnering with and why, in one sentence?
Peter ter Haar: I would love to be partnering with Waze to understand what we can do, how they can help us keep our mapping up-to-date even more.
Tim Trainor: Well, I’d like to partner with Steve. I think that the time is here where not only have we relied on local governments and other sources for data, but with the volunteer geographic information activity going on, it’s a great source for getting what seems to be very high quality data from a large labor source. And I think that’s a great opportunity to contribute to a national good.
Steve Coast: I’d reflect exactly the same thing. I mean OpenStreetMap, US Census, USGS and other people should be working together to create these free and open base maps that are eventually going to just do away with every other proprietary map on the planet.
Noam Bardin: I think we have a bingo here because I’m actually in the process of writing an email to Peter about how I think we should cooperate in the UK and so we’ll take that offline.
Ian White: Okay. Good. And I’ll take only five percent, which is saving two percent off the regular investment banking fees. So thank you all for participating in this “podcast” or “remote cast.” Again joining us today, we had Peter ter Haar who is Head of Products for Survey, the UK’s national mapping agency; Tim Trainor, Head of Geography at US Census, responsible for TIGER and the master addressing file; Steve Coast, Founder of OpenStreetMap; and Noam Bardin who is CEO of Waze. My name is Ian White. I am Founder of Urban Mapping. We look forward to having all of you listening and many more. Join us at the O’Reilly Where 2.4 Mapping Conference in San Jose this April 1st — this is not a joke — on April 1st at 2:00 p.m.. Thank you very much. And please do contact us with any other questions or things you’d like to hear at the panel. [End of recording]
During Where 2.0 NAVTEQ will be awarding the winners of its LBS Challenge in North America. The Awards will be given out Tuesday evening (March 30th) during Ignite.