Laura Bliss is a staff writer at CityLab, covering transportation, infrastructure, and the environment. She also authors MapLab, a biweekly newsletter about maps that reveal and shape urban spaces (subscribe here). Her work has appeared in the New York Times, The Atlantic, Los Angeles, GOOD, L.A. Review of Books, and beyond.
As OpenStreetMap grows, companies like Mapbox are a threat to the long reign of Google Maps. And that’s a good thing.
Maps may be a necessity of an empire, as Paul Ford writes on the rise of Google Maps in The New Republic this week, but whether that empire belongs wholly to Google is up for debate. Right now, the entire digital mapping industry is being re-mapped.
Last week, Mapbox, a map development company based in Washington, D.C., announced that it has raised some $52.55 million in Series B funding, a sum CEO Eric Gunderson called the biggest ever for a mapping company.
Mapbox doesn’t exactly make maps, though. It builds towers of software that organize sets of geo-spatial data for other kinds of businesses—real estate, transportation, agriculture, government, smartphone apps. Take, for example, the Foursquare app, which Mapbox powers:
It layers information about a variety of destinations, and what Foursquare users think about them, on top of a pre-existing street map. Mapbox builds the software that enables this “stack” of geo-spatial data sets.
Ditto for the interactive maps on the United Nations Development Program’s website, which charts thousands of U.N. projects and their funding in 177 countries. Mapbox’s tools helped build this:
What’s key here is the basemap underpinning the above examples. You can see its tiny attributions in the corners: OpenStreetMap (OSM). Though Mapbox uses a mix of open-source, public-domain, and proprietary sources to generate maps, it relies heavily on OSM data as the bedrock of many of its products.
Founded in 2004 by Steve Coast, a British entrepreneur and cartographile, OSM is a totally free and open-source map of the world. Like Wikipedia, it is constantly being updated by a community of now more than two million members, who use GPS tools, satellite photographs, and their own local knowledge. Anyone can edit it. Many argue—and studies have shown—that the breadth of its team of contributors has given rise to data that is as good or even better than Google Maps. Community members verify new entries and correct mistakes.
And anyone—including businesses—can use the OSM API to build their own maps. (Think of an API, as the New York Times once put it, as a “programmers-only side entrance” into the mapmaking machine.) OSM is not a household name like Google, but it may be soon.
Google Maps defines the way we navigate from A to B, for free, and it does so extremely well. It also sells its API to its a number of businesses. As of 2012, Apple, Foursquare, Craigslist, and Wikipedia (to name just a few) all built their maps using the Google Maps API.
But today, none of those companies are using Google—partly because of how much Google started to charge for its services and data, and because of the limitations it draws around what companies can do with them.
All four of the aforementioned companies moved to using OSM (partially, in Apple’s case) because it’s free, and often as good as Google. And because the value of proprietary map data is rapidly plummeting as OSM gets better and better.
The world-mapping empire
Besides OSM, there are three other global data sets: Google, TomTom, and Nokia’s HERE. You know Google’s map-making empire very well. The company employs a small army of professional mappers, editors, developers, designers, and 3-D-camera-mounted-van drivers to build and maintain its digital map. TomTom and HERE have similar, if smaller, operations, and they sell their data to other companies like Google does.
But it’s extremely costly to sustain these massive mapmaking operations through licensing schemes alone. Nokia is learning that lesson right now. The Finnish IT corporation purchased HERE, formerly known as Navteq, for more than $8 billion in 2007. Since then, HERE has aggressively competed to create digital maps as advanced as Google’s.
But in April, Nokia announced that HERE was for sale. Bids from companies ranging from Uber to Microsoft to a suite of carmakers have rolled in around the $2 and $3 billion mark. It’s a lot of money, but compared to what Nokia originally paid for, it’s bargain-bin. That confirms what Mapbox seems to have known for a long time: that the value of proprietary map data is approaching zero. In large part, that’s because of free sources like OSM.
It’s not the only way the geo-spatial landscape has rapidly evolved. “Ten years ago, maps were all about a navigation-use case,” writes Marc Prioleau, a location-based services expert who sits on Mapbox’s board of directors. “We looked at streets, addresses, POIs, maybe a little traffic. That's not true today. Today, it's about imagery, social media, real-time content, and much more. And three years from now, there will be only more data”—especially as location-based sensors increasingly populate the built environment.
Mapbox’s suite of data-organizing services is flexible, customizable, and cheap—because they’re getting some of their most fundamental data for free. Google is not. According to some market research, the location-based-services industry will be worth some $40 billion by 2019. That’s what Mapbox is tapping into, and what its investors are betting huge chips on.
Slightly strange bedfellows
Mapbox may be getting OSM data for free, but since its inception in 2010 it has made a point of collaborating extremely closely with the project. “We think of Mapbox as an OSM company,” says Gunderson. “We believe OSM is going to win.”
The company tasks a number of staff members with editing and improving the map daily. It also hosts mapping parties and workshops, helps organize yearly conferences, and regularly submits donations. Shortly after Mapbox announced its fresh round of funding, it donated $20,000 to the OpenStreetMap Foundation (OSMF.), the non-profit, volunteer-staffed entity that supports the project.
“Geospatial data is a project that’s too big solve by a single commercial entity with a business model based on commercializing that data,” says Alex Barth, who heads data at Mapbox. “Open collaboration around global map data is the future of maps. That’s it.”
But there are obvious tensions, particularly as more eyes turn to OSM as a profitable data set, and to Mapbox, as the most successful company yet to have turned a profit from it. Some—though not all—OSM members are wary of commercial entities coming into “productify” what is an open-source, community-driven project. OSM’s license states that all uses (private, commercial, government, humanitarian) of its data are equally permissible, so long as the user gives proper attribution, and shares back to the OSM community any improvements to the map that she makes public.
“There are people who spend hours and hours every week on OSM, as volunteers,” says Kate Chapman, a director of the board of OSMF. “Those individual contributions are what makes the map so rich and detailed, so you know, the idea that someone would be getting rich off their efforts—which is an oversimplification—but, you can imagine not everyone’s going to be OK with that.”
These tensions bubbled up earlier this month in a meeting at State of the Map U.S., an OSM conference held in Manhattan. More than 800 gathered from the commercial, humanitarian, and government mapping worlds. Mapbox was highly engaged in the event, as a conference co-sponsor and with a number of employees giving talks—including Barth, who was particularly visible.
In one discussion facilitated by Barth, conversation quickly turned to dissatisfaction shared by business leaders about the piece of OSM’s license that requires users to “share back” their improvements to the data. In some places, the legal verbiage is vague on how much and what kinds of data must be contributed back, which makes some businesses nervous that OSM s legally incompatible with their own data (i.e., user data). OSM members see the license as a way of keeping information truly open—and of preventing users from simply profiting from the data without contributing anything back.