The great thing about recreation is that depending on how you define it, recreation can mean a lot of different things and a lot of different sports. That wonderful variety can present a challenge when you try to inventory all of the physical resources that are commonly considered recreational resources. The data behind the www.goPLAYkansas.com website is generally split into three formats of GIS data: polygons, points, and lines. The type of format used depends on the resource.

Parks are represented in the GIS data as polygons. Polygons are the best way to show parks because they represent the boundaries of the site and can include resources located within them. In can get a little tricky if parks cross political boundaries, or if two parks are located adjacent to each other (think of Pawnee Prairie Park and Tex Consolver Golf Course in Wichita).

Points GIS data represents the recreation resources (i.e. playgrounds, fields, picnic tables, etc.). the points data is utilized for this category of features because the features vary in size from very small to very large and it would be time consuming to draw the boundaries of each one out. The points work well to illustrate single items out in the field in a quick and easy format.

Trails and paths are represented in the GIS data as lines. The line data can be broken up into segments that have individual attributes (i.e. type of material, width, etc.). This is how the website is able to display multiple types of trails.

One of the difficulties that came to light as the team was working with the data was the ability to represent the nearest access to trails on a map. The problem was that sometimes the nearest way to a path is down a cliff or across a busy roadway. The team came up with the solution of plotting access points for trails using the point data. This way, the distances to trails are accurately represented on the map. This feature is not available yet, but the team is working on it – so keep an eye open for those updates.

In the next post we can take a look at the challenges behind the use descriptions.

Pulling together disparate data from multiple sources and putting a new spin on it is a challenging project. The way such a project starts is the realization that there is a vast sea of information that you know is ‘out there,’ but your vision to it is limited, or you just don’t understand it. You know that you could use this information to inform and notify others, if it was presented in a manner that is easily digested.

When I first looked at the data we had downloaded from Wichita and Sedgwick county, it was obvious that the fundamentals of what we wanted to do with goPLAYkansas were there, but it was in a format that most people would not understand.

Parks were made up of points (latitude and longitude) and there were points within those parks representing amenities. In some cases these amenities were obviously tied to activities like softball and baseball.

Where are the trails?

There is another set of data points that represents all the sidewalks, and trails in Sedgwick county.

This data, of the parks and the trails, is held in a special geo-spatial format. A format that is not compatible with the types of database servers available to web designers on hosting providers. It is a special format that the guys into such things have developed to make things like your car GPS possible. Most of us are all familiar with the most popular type of search perfromed on this data: “What is the shortest distance between two points.”

Since this type of search has become common place, the leading web database, MySQL, now includes geo-spatial functions. Like any good file format, however, there are different versions from different systems. The geo-spatial format for our current version of MySQL may, or may not, be the same version or format that the geo-spatial guys are using today. Just like MS Office documents, one version is not always compatible with another.

Our first hurdle was to convert the data into a format the database system on our hosting provider would understand.

Now that we had the data, what can MySQL do with it? After many hours of frustration and Internet searches with Dogpile we realized that our version of MySQL did not support the most common search, a distance between two points. Another Dogpile search reveals that we are not the only ones beating our heads against this limitation, and bird watchers have the solution. A fancy trigo-geo-something-or-other-function that you can apply to latitude and longitude points, (in plain text, not geo-spatial format) to get a distance between points. Best of all it was designed for MySQL.

With the distance calculation in hand we had a pretty good web site. You could find the closest parks from your starting point. You could find the type of amenities in the park, and with some data massaging, the type of activities the park supported.

But where are the trails?

We had to find a way to associate trail data (that included city sidewalks) to the parks. Thankfully, our version of MySQL supported the Intersect function on geo-spatial data. Intersect told us where all the trail points intersected with all the park points. With that query we were able to generate a new lookup that puts trails in our near the parks. Scripts were written to create those tables, so we know what steps to take as we collect data from the rest of the state.

goPLAYkansas is one of those new breed of websites that uses data and information from around the web to create a new look at the vast sea of information that is ‘out there.’ It makes full use of modern web browsers, especially the Javascript programming language. goPLAYkansas works best with an updated Internet browser that has a fast Javascript engine. If you are still slugging along with Internet Explorer 6 or 7 you may find the website sluggish, especially with parks that have odd shapes and lots of amenities or trails. Upgrading to Internet Explorer 8, or the latest version of Firefox or Safari will ensure that you have a fast Javascript engine and give you the best experience.