Ten years ago, when PostGIS was at 0.8 and the world was fresh and new, I was pretty convinced our industry was on the cusp of an open source revolution. When folks got a taste of the new, flexible, free tools for building systems they’d naturally discard their legacy proprietary software and swiftly move on to a more enlightened existence. I felt excitement, and wind in my hair.

Similarly, pretty much every year since 2000 has been heralded by someone, somewhere, breathlessly proclaiming that (finally) “this year will be the year of the Linux desktop”.

A funny thing happened on the way to the open source revolution. It turned out to be more of an open source evolution. In aggregate, change has been slow, incremental, though always in the direction of more open source use.

So in looking forward to what to expect in the new year of open source geospatial, my predictions will have to be a little circumspect — the big things will change slowly, but at the edges there will be a great deal of churn and change:

• Oracle will notice they are losing customers to PostgreSQL. While MySQL always got all the press as “the open source database” it’s been PostgreSQL that has had the enterprise features from the start to go toe-to-toe with the big guy. As Oracle continues to increase maintenance prices to please Wall Street, customers are beginning to think the unthinkable: maybe it’s time to reevaluate their database standard.
• The coolest stuff will continue to have open source at the foundations.Whether it be the Linux-running, GDAL-enabled satellites of PlanetLabs or the latest Android phablets, the coolest innovations will stand on the shoulders of open source and reach upwards from there.
• Most of the open source action will be in JavaScript. Juan mentioned that geospatial programming is increasingly polyglot, but the open source arena with greatest level of churn right now is the JavaScript world, both on the client and the server. There is a lot of sound and fury out there. Some of it signifies nothing, but some of it is laying the foundations for the standards we’ll be using for the next decade. Contemporary JavaScript reminds me of Java circa-2005: multiple projects with similar functional goals, competing design philosophies, and huge potential. Separating the signal from the noise in this kind of environment takes real expertise, so I’m glad we have some of the best and brightest JavaScripters in the geospatial world on our team.
• PaaS will join open source in the evolution revolution. As I am just starting to learn platform-as-a-service (PaaS), I feel it has both the same promise as open source, and the same long organizational learning curve. As a result, it’ll find its way into core IT only slowly, as experienced folks like me pick it up, and the next generation moves into operational jobs. Since PaaS is open source almost by definition, growth of cloud platforms will also further institutionalize open source components for system building.
• Iterative, open source style development will gain more ground. The public failure of the healthcare.gov site, and lashing of that failure to waterfall methodology, can only be good for agile development. There’s already lots of agile in enterprises, but it’s still something that “progressive” organizations do, it’s not the default. The more people think about technology in open source ways (it’s a process, not a product; it’s about managing change, not achieving a final state), the better for open source.
• Organizations will yearn to work with OpenStreetMap, and some will figure out how. While licensing continues to lock out many public organizations from participating, others will make peace and begin integrating OSM into their workflows. The lucky ones will get approval from their lawyers to work with OSM directly. The less lucky will settle for using OSM as a change tracking driver to keep authoritative maps up to date.
• Boundless will integrate even more open source technology into OpenGeo Suite, making it yet easier to get started with enterprise geospatial systems. OK, that was an easy one since Eddie agrees, but I have to get at least one right.

Have a great new year, from everyone at Boundless!

Paul Ramsey, has been working with geospatial software for over ten years and, in 2008, received the Sol Katz Award for achievement in open source geospatial software.