Why open source needs an open geographic dataset.

Open source has won. The fact that free software now dominates practically every sector of computing (with the main exception of the desktop) is proof of that. But there is something even more important than the victory of open source itself, and that is the wider success of the underlying approach it embodies. People often forget just how radical the idea of open, collaborative development seemed when it appeared in the 1990s. Although it is true that this philosophy was the norm in the very earliest days of the field, that culture was soon forgotten with the rapid rise of commercial computing, which swept everything before it in the pursuit of handsome profits. There, a premium was placed on maintaining trade secrets and of excluding competitors. But the appearance of GNU and Linux, along with the other open software projects that followed, provided repeated proof that the older approach was better for reasons that are obvious upon reflection.

Open, collaborative development allows people to build on the work of others, instead of wastefully re-inventing the wheel, and it enables the best solutions to be chosen on technical, rather than commercial, grounds. The ability to work on areas of personal interest, rather than on those assigned by managers, encourages new talent to join projects in order to pursue their passions, while the non-discriminatory global reach of the open method means that the pool of contributors is much larger than for conventional approaches. However, none of those advantages is tied to software: they can be applied to many fields. And that is precisely what has happened in the last two decades, with the ideas underlying free software producing astonishing results elsewhere.

Arguably one of the most important knock-on effects of free software is the open nature of the World Wide Web. In his book, Weaving the Web, Tim Berners-Lee writes that he originally wanted to release his creation under the GPL and decided in 1993 to place it in the public domain only because key computing companies indicated that they wouldn't support the original idea. Also influenced by Richard Stallman's work is Paul Ginsburg, who set up arXiv.org in 1991 as an open repository of scientific preprints. This idea of unrestricted sharing of academic knowledge later gave rise to the open access movement. Similarly, the GNU project's concepts and even name fed into Nupedia, the little-known precursor to Wikipedia. The latter's open collaborative approach has been so successful, it's hard to imagine modern life without this extraordinary resource.

Those are all major projects that are familiar to many people. But there's another application of the core ideas of openness and collaboration that is not as well known as it should be:

OpenStreetMap is a world-wide collaborative project aiming at providing free map data, under an open license, to anyone who wants it. Volunteers all over the planet contribute their local knowledge and their time to build the best map ever.

OpenStreetMap began in 2004. It was inspired by Wikipedia, and it took off thanks to Linux, as its creator, Steve Coast, explained in 2014:

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.

Despite its low profile, OpenStreetMap is arguably one of the most important projects for the future of free software. The rise of mobile phones as the primary computing device for billions of people, especially in developing economies, lends a new importance to location and movement. Many internet services now offer additional features based on where users are, where they are going and their relative position to other members of social networks. Self-driving cars and drones are two rapidly evolving hardware areas where accurate geographical information is crucial. All of those things depend upon a map in critical ways, and they require large, detailed datasets. OpenStreetMap is the only truly global open alternative to better-known, and much better-funded geodata holdings, such as Google Maps.

The current dominance of the latter is a serious problem for free software—and freedom itself. The data that lies behind Google Maps is proprietary. Thus, any open-source program that uses Google Maps or other commercial mapping services is effectively including proprietary elements in its code. For purists, that is unacceptable in itself. But even for those with a more pragmatic viewpoint, it means that open source is dependent on a company for data that can be restricted or withdrawn at any moment.

There is a more subtle problem with using a proprietary dataset. Free software is inherently about freedom because it allows users to do what they wish with a program. Proprietary code, by contrast, is under the control of the company that produced it, which means its users are subject to any constraints that are built in to the software: "Code is law", as Larry Lessig famously wrote. It is the same with mapping data and services. If the dataset and mapping software are proprietary, they come with the constraints and biases of the company that created them built-in, and they are impossible to circumvent. This might mean certain businesses are highlighted in an area because they have paid to be given preference. Services that help users find optimum routes also may be biased in socially harmful ways—for example, avoiding districts that the map provider deems "unsuitable".

The lack of a wider appreciation of the importance of OpenStreetMap to the future of open source is bad enough. But the situation seems to be even worse if a recent post by a long-time contributor to the project is correct. Serge Wroclawski worked on the OpenStreetMap project for around eight years, and he helped set up OpenStreetMap US, a nonprofit organization dedicated to OpenStreetMap in the United States. In 2014, he wrote a widely discussed article "Why the world needs OpenStreetMap", which explores many of the points mentioned above in greater detail, and with greater authority. It's well worth reading, as is his latest exploration of OpenStreetMap, ominously called "Why OpenStreetMap is in Serious Trouble". As he writes:

While I still believe in the goals of OpenStreetMap, I feel the OpenStreetMap project is currently unable to fulfill that mission due to poor technical decisions, poor political decisions, and a general malaise in the project. I'm going to outline in this article what I think OpenStreetMap has gotten wrong. It's entirely possible that OSM will reform and address the impediments to its success—and I hope it does. We need a Free as in Freedom geographic dataset.

Leaving aside the political issues Wroclawski raises, one key point to emerge from his critical examination of OpenStreetMap's current status is that there are significant opportunities for the Open Source community to help take OpenStreetMap to the next level. Assuming Wroclawski's analysis is correct, there are plenty of ways programmers could help rectify some of the deficiencies of the project he identifies. In the process, they could tackle exciting and worthwhile new coding challenges. It often seems that many of the major free software projects have reached a slightly boring maturity, with all that this implies for offering sufficient incentives for people to join. The world of mapping, by contrast, has the potential to provide ambitious coders with the scope to make their mark in a new field of interest to billions of users.

So far, OpenStreetMap has spawned only a limited range of free software tools, and it badly needs many new ones if it is to match proprietary offerings. That's a relatively straightforward task. Much harder will be creating open versions of online services like Google's Waze, which describes itself as "the world's largest community-based traffic and navigation app". Doing so will require working with well-funded organizations—perhaps the Mozilla Foundation—or with open-source companies like Canonical, since the investment required will be considerable.

Although undoubtedly difficult, creating high-quality map-based services is a challenge that must be tackled by the Open Source community if it wants to remain relevant in a world dominated by mobile computing. The bad news is that at the moment, millions of people are happily sending crucial geodata to proprietary services like Waze, as well as providing free bug-fixes for Google Maps. Far better if they could be working with equal enthusiasm and enjoyment on open projects, since the resulting datasets would be freely available to all, not turned into corporate property. The good news is that OpenStreetMap provides exactly the right foundation for creating those open map-based services, which is why supporting it must become a priority for the Open Source world.