To me the most interesting thread in bringing Architectures of Participation to the geospatial world is the creation and maintenance of geographic information itself. I believe it has the greatest potential to have a true open source type movement around it, and indeed the first signs of it have already emerged: the mash-ups we’ve seen around Google Maps, Open Street Map, and others are pointing the way forward. These thoughts aren’t new to anyone who has seriously thought about mapping and open source/’web 2.0′, it’s the logical next step. But my posts on this are going to attempt to present the ideas to those who may not have been embedded in these thought streams, and I will ground the thoughts in Weber and Benkler, the two leading thinkers in my mind on bringing the ‘open source process’ to domains other than software. I will point to examples of how this is already happening in the geospatial realm, and I’ll also articulate my technical vision for the next wave, building on standards and existing GIS technologies. And I’ll touch on where I hope to see some of this stuff end up, and any related things I want to bring up along the way, as that’s the luxury I get with a blog ;).
Weber argues in ‘The Success of Open Source‘, that the most interesting thing about open source is the process, and that it theoretically could be applied to any digital information, as it is all infinitely copyable at no cost to the owner. Benkler similarly sees a broader social-economic model in open source in his ‘Coase’s Penguin‘. He calls it a third mode of production, the “commons-based peer-production”, characterized by groups of individuals collaborating on large scale projects with motivations that are not drawn from either the pricing of markets or the directions of managers (market and firm modes of productions, respectively).
Digitized geographic data certainly is infinitely copyable, but there are few examples of an people using open source process with geographic data. One can start to understand how open source geographic data might work by re-examining my metaphor of legos for open source software in the context of geospatial data instead of source code. Just as source code is a number of small files that fit together to make a program, so too does geographic data (points, lines, polygons) fit together to make a map. The ‘instructions’ in the case of geodata are not the human readable source code, but instead the raw data that can be used to make maps. Just as a binary program is a pre-assembled lego car, so too is a printed or online map un-modifiable. If someone wanted to change the map, to remix it for their purposes, or even just fix a street they know to be wrong, they would need the raw data (raw data = source code = instruction bookle of the lego metaphor). Most users are fine with the pre-assembled version, the actual map, but motivated users could likely do much more with the raw data – such as generate new maps, change the ‘style’ (the colors and data displayed) to emphasize different aspects, and make corrections to errors – that they could share with others. A license that stipulates that users of the data must also make their modifications open to others would certainly be possible, just like the GPL does for software.
In a future post I’ll explore the criteria Weber speculates as needed to build an open source process around domains other than software, and compare it against geospatial data. But for now we’ll hold off. I just want to start with raising the point that when information is digital, and is a ‘non-rival’ good, that is it doesn’t cost me anything if you have a copy of it, then ‘scarcity’ becomes much more of an artificial construct. The only thing enforcing that scarcity is intellectual property laws, and the open source software movement has shown that an initially small group of motivated people can turn that scarcity on its head. I’d like us to take a similar approach, to cooperate to build maps that are even more accurate and up to date than commercial providers and spy agencies can provide, taking that traditional source of power and putting it in the hands of all. It sounded silly with open source software – to build a better operating system than one of the most dominant companies in the world can – but just as that is coming to pass as many huge players rush in to help out, so too I think we could see the biggest buyers of commercial data flock to a solution that has them cooperate in a more economically efficient mode.