The Long and Winding Road
When Jim Sparks started out on his journey as the State of Indiana’s new Geographic Information Officer (GIO) in 2007, the first thing he wanted to do was to get the lay of the land – visiting as many groups doing GIS in the state as he could. In many of his meetings throughout the first 90 days in his new position, he kept seeing overlapping goals and initiatives. One of these – the idea of aggregating and homogenizing several critical datasets from local data stewards in Indiana – was in the early planning stages at the Indiana Department of Homeland Security.
This was a more ambitious goal than it may seem. Many of the obstacles had less to do with technology (which FME Server deals with handily, as we will see), than with inter-governmental collaboration and funding. The first challenge? There are 92 counties in Indiana, each one with their own data management practices and datasets, and limited resources. When Jim arrived on the scene, only a handful of counties had decided to participate. A long road lay ahead – but the good news was that other interest groups, including the Indiana Geographic Information Council (IGIC), working from an NSDI CAP grant, and other Indiana state agencies, were looking for a similar solution, and by tackling it collaboratively, everyone could win.
The Architecture
The data was to ultimately be compiled into and made available through the already existing IndianaMap online GIS data portal
(www.indianamap.org). The datasets requested were a subset of those already maintained by the counties – point addresses, parcel shape and ID numbers, local government geoadministrative boundaries, and street centerlines with address ranges. No personal information about individuals was to be included.
Some very intelligent early decisions were made regarding the architecture – most importantly, the data sharing process needed to be non-invasive, and not require that counties change their current methodologies, formats, data models, and platforms. And so, a central model was designed based on FGDC standards, and the counties were asked to simply add an OGC WFS stream to their current processes. Transformation and homogenization to the destination model, residing in a central Esri ArcSDETM database, would be handled by FME Server, and be the responsibility of the GIO.
The Real Challenge
The architecture was in place, but the real challenge lay ahead. As Jim tells us, “Early on, there was a lot of doubt that we could gain the support of 92 counties. In fact, I was told that if we got 20 on board, we’d be doing good.” The counties’ resources and personnel were stretched to the limit already, and even though adding a WFS was minimally disruptive, it wasn’t always an easy task to convince people of the value of contributing.
Jim and his partners embarked on what was to become a multi-year campaign to gain buy-in from the counties. There were objections after objections – data security, privacy concerns, lost revenue, and more. But gradually, county by county, meeting by meeting, they gained acceptance. And it worked.
Making It Work
According to Phil Worrall, Executive Director of IGIC, there were three fundamental keys to the success of the Indiana Data Sharing Initiative. The first was external funding from project sponsors. Indiana DHS provided $14,497 to each county who agreed to participate, to defray the cost of adding the WFS service. The second key factor – enabled by FME Server – was that they weren’t asking the counties to change their data. The changes would be handled by a series of FME workspaces consuming the feature streams, harmonizing, and loading to the central database. Third, Phil noted that it was important to make this data easily accessible to a wide audience. “The more the data gets used, the greater the return on the investment. So adding these high value statewide data layers to the IndianaMap was an obvious requirement.”
Somewhere along the way, over half of the counties in the state had signed on. The project caught the attention of VerySpatial.com, who profiled it on “A Very Spatial Podcast” as a noteworthy model for the creation of a national map. And they kept gaining momentum – 55 counties signed on, then 64, and as of this writing all but five counties are sharing their data, and one of those five has verbally committed to participate.
Rave Reviews
So, five years down the road, was it worth it? There’s no doubt in Jim’s mind. “Having this regional data readily available is far and away more valuable than the sum of its parts,” he says. “Things like natural disasters, air quality, water quality, economic development – they don’t respect administrative boundaries. Earlier this year, when Indiana was hit with tornados, every minute counted. And the data was ready to support response efforts.”
When a major auto manufacturer was recently evaluating business locations, they weren’t looking at any one county – they wanted information about the region, and it was ready for them. “You never know who’s considering a site in the state,” says Jim. “If they have ready access to data, they may stop and take a look. If it’s not easily available, you may well get passed over and never even know it.” The data is also being used in Indiana’s new statewide first responders dispatch system, and citizens’ services applications like the Indiana Secretary of State’s “Who Are Your Elected Officials?” website.
We asked Jim if they could have done it without FME Server. His answer – “Theoretically? Maybe. Practically? No. It was the best solution we knew of, and had we tried to do it another way, we probably would have failed before we even got started.” Thanks for the kind words, Jim – and congratulations on an amazingly successful initiative. May the road forward be a smooth one!
Learn more about FME Server at safe.local/FMEServer.