Last year, I described OpenStreetMap (OSM) as the “open source of maps”. On the occasion of the project’s tenth anniversary, I’d like to explore this important example of open collaboration in a little more detail, and explain why I think it is destined to become the next absolutely key open project.
First, some history. To celebrate OSM’s anniversary, TechCrunch has an excellent interview with the project’s founder, Steve Coast. Here’s how things began:
the original idea was way simple. I had a GPS unit attached to my laptop and there wasn’t very much you could do with it because there wasn’t any data. You could download a picture of a map, but if you wanted to do anything like have the computer figure out what roads you were on or do turn-by-turn navigation or anything that’s kind of useful, you couldn’t do it because there wasn’t any map data. So I just thought: why don’t we create some map data? When you have a GPS, you can just drive or bike or walk around all the streets, all the roads, all the footpaths and use that information to create a map. I build a little bit of the map where I live and you build a little bit of where you live and we build this thing like a jigsaw puzzle — and incidentally give it away for free. Just like Wikipedia was building an encyclopedia in a very similar way, I copied many of the ways Wikipedia was built with — like open licensing, the ability for anybody to contribute and so on.
Although Coast invokes Wikipedia there, GNU/Linux also played a crucial role in the project’s development:
I gave a lot of talks. Linux user groups used to be very popular — people getting together on a Saturday afternoon to talk about Linux. They already knew half the story because they knew about open source and they knew about computers and data. So it wasn’t too hard to explain what OpenStreetMap was doing. From there I gave talks at mapping conferences. I stopped counting at something like 500 talks. I used to use the number on my first slide.
The pre-existence of Linux communities allowed Coast to piggy-back off them. As he points out, the basic ideas of collaboration and openness were the same, so he didn’t really have a hard sell for his application of them. This is another good example of how crucial the success of GNU/Linux has been to all the other “opens” - open content, open data, open access and OpenStreetMap.
The other interesting parallel with the rise of GNU/Linux is how Coast now works for Telenav, a navigation software company (he also worked for Microsoft for a while, intriguingly....) Telenav has apparently hired some key OSM people, just as companies like Red Hat hired key kernel hackers in the early days of Linux. In the TechCrunch interview, Coast explains what the key challenges are currently:
OpenStreetMap lacks a couple of things. Navigable information like one-way streets, time restrictions or speed limits. It also lacks address data. Telenav has a lot of GPS traces. We process all of those into navigable information. If you take all the people traveling on the freeway and they all go 55, you know that’s probably the speed limit. If nobody is turning left at an intersection, there is probably a turn restriction there.
So you can fix that with GPS traces. But the address data is harder. In the U.S. you can license this data, but in Europe and other places it’s much harder. In the U.S.. the federal government is mostly a public domain organization and that trickles down to the local governments. You can get data from that. Every other country tries to own all the mapping data. In the U.S., most of the address data is also very predictable. In the rest of the world it’s not that simple. In Japan, the street number is often based on the block and the age of the house. So the first house is number one, the second number two and so on. So it’s much harder to figure out where things are by inference.
It’s something that a lot of people want to solve, so I expect it will get fixed somehow and there are a bunch of interesting way to do this. You can crowdsource the data, you can pay people to go collect it, but I expect it’ll be a mixture of a number of things. Every time you use a check-in app, for example, you are signalling that this restaurant with this address is in this location.
That brings us neatly on to the future of OSM: what could it add to its basic mapping information – and what should it add? Should it, for example, seek to match Google Maps in every respect, including navigation (something that Coast obviously thinks is important) and even satellite imagery? The former seems eminently achievable, and maybe low-cost drones could be used to crowd-source the latter. But what else might be added?
As I wrote last year, I think that OSM should aim to become the open source infrastructure for the mobile world, just as GNU/Linux has for the preceding generation of computers (and also underlies the dominant mobile platform, Android.) Quite what that might mean can be seen from the Waze service:
Waze is all about contributing to the ‘common good’ out there on the road.
By connecting drivers to one another, we help people create local driving communities that work together to improve the quality of everyone’s daily driving. That might mean helping them avoid the frustration of sitting in traffic, cluing them in to a police trap or shaving five minutes off of their regular commute by showing them new routes they never even knew about.
So, how does it work?
After typing in their destination address, users just drive with the app open on their phone to passively contribute traffic and other road data, but they can also take a more active role by sharing road reports on accidents, police traps, or any other hazards along the way, helping to give other users in the area a ‘heads-up’ about what’s to come.
In addition to the local communities of drivers using the app, Waze is also home to an active community of online map editors who ensure that the data in their areas is as up-to-date as possible.
Although that emphasises car-based road data, it generalises to all forms of transport. Waze is about using an app on mobile devices to crowd-source real-time and current information about travel and places. It’s a great idea, with one fatal flaw: the huge amount of unique data contributed by Waze users belongs to Waze: it’s a classic case of users creating the service and then handing all the value over to someone else. So much for “contributing to the 'common good' out there”.
Things are made worse by the fact that Waze is now part of Google, which means that all this incredibly revealing information about you and your fellow travellers is being fed into the maw of the Google machine, to be processed, and cross-linked and analysed to reveal rather more about people than they probably suspect.
In fact, the problem with Google is even larger, since it owns the map service that dominates the location space as completely as Microsoft Windows does the desktop domain. Even before it acquired Waze, Google was encouraging users of Google maps to contribute “improvements” to the company, which got to keep most of the benefits, just as it does with pretty much all of its “free” services.
Mikel Maron warned about the dangers of this approach for OSM – and for us – as far back as 2011, when he wrote a post entitled “We Need to Stop Google’s Exploitation of Open Communities”:
Google’s strategy is to build market in Africa by appropriating the appearance of open data community methodologies, yet maintaining corporate control of what should rightfully be a common resource. They are specifically targetting govts and NGOs, offering to “map their country for free”, but keeping the results, and attracting customers.
What bothers me so much is how they have blatantly copied OpenStreetMap. First their MapMaker product is directly modelled on OSM, but with a restrictive data license, where you can not use the data as you see fit. Second, they have stolen the idea of Mapping Parties, a unique concept and name we developed. Third, they’re even copying initiatives to map impoverished informal settlements, like Map Kibera.
This is not stuff that only really matters to map geeks. Maron gives a brilliant example of just how important maps are in the real world of society, politics, war and peace. He invites us to compare Google’s map of an area of particular interest today – Gaza – with that of OSM. Where the former shows an anonymous, largely meaningless grid of streets, with no detail – and thus no humanity - whatsoever, OSM’s version not only names most of the main streets and buildings, but it shows the actual buildings present – an exact representation of the people and lives they contain. In other words, maps are highly political, and that makes it important that they are not controlled by a few commercial entities.
An article in the Guardian earlier this year, by Serge Wroclawski, explores this issue, and suggests how OSM is vitally important here. He singles out three serious concerns about maps provided by Google (or any other company):
Who decides what gets displayed on a Google Map? The answer is, of course, that Google does. I heard this concern in a meeting with a local government in 2009: they were concerned about using Google Maps on their website because Google makes choices about which businesses to display. The people in the meeting were right to be concerned about this issue, as a government needs to remain impartial; by outsourcing their maps, they would hand the control over to a third party.
It seems inevitable that Google will monetise geographic searches, with either premium results, or priority ordering, if it hasn’t done so already (is it a coincidence than when I search for “breakfast” near my home, the first result is “SUBWAY® Restaurants”?).
This is the usual issue of control. If the software on your computer is not free software, it is not really your computer, just one that the proprietary software vendor allows you to use in certain ways, determined by them.
The second concern is about location. Who defines where a neighbourhood is, or whether or not you should go? This issue was brought up by the American Civil Liberties Union (ACLU) when a map provider was providing routing (driving/biking/walking instructions) and used what it determined to be “safe” or “dangerous” neighbourhoods as part of its algorithm. This raises the question of who determines what makes a neighbourhood “safe” or not – or whether safe is merely a codeword for something more sinister.
That is an extension of the point about Gaza: maps are not just points of data, but represent points of view. Although a completely neutral map is probably impossible, we can at least make it as open as we can. Finally, there is the issue of how these maps are being created, and how location metadata might be (mis-)used – even more pertinent in the post-Snowden world:
map providers have an incentive to collect information about you in ways that you may not agree with. Both Google and Apple collect your location information when you use their services. They can use this information to improve their map accuracy, but Google has already announced that is going to use this information to track the correlation between searches and where you go. With more than 500 million Android phones in use, this is an enormous amount of information collected on the individual level about people’s habits, whether they’re taking a casual stroll, commuting to work, going to their doctor, or maybe attending a protest.
That isn’t an issue with OSM, which collects only the basic data it needs for creating maps, with no hidden agendas.
As I hope the above indicates, the success of OSM in the world of location matters to all of us as much as the success of GNU/Linux does in the field of software. We need a comprehensive but independent source of geodata and associated information. In its ten years of existence, OpenStreetMap has made huge strides to becoming precisely that; it is in all our interests to help it to continue to thrive and grow until it is just as successful and pervasive as GNU/Linux has now become.