Openstreetmap - Putting Camden on the Map
Openstreetmap is an open source mapping project with the goal of creating a free to use and editable map of the world. It relies on non-licensed data and contributions from the public to grow. Besides producing a free visual map of the world, the data produced by mapping everything from roads, apartments, hotels, restaurants, and theaters, to post offices, parks, bike paths, rail lines and much more, has fostered the creation of applications that utilize the data in some form. Foursquare, Craigslist, and MapQuest all use data provided by Openstreetmap to power their applications.
While most of the roads, highways, rail lines, and etcetera of most major cities and urban areas have already been mapped in great detail by registered users, there remains many places which have yet to be digitized - the untamed frontier, or the wild west of Openstreetmap manifests itself in the swaths of drab grey nothingness emerging between the roads. Many places are lacking in data for buildings - schools, hospitals, homes, apartments, hotels, retail centers, malls, or rather any place with an address. One such place in New Jersey that could use some Openstreetmapping is Camden. I spent a week mapping about a square mile of an area in east Camden. I used license free data in the form of NJ GIN's 2012 aerial imagery to map the man made structures present in the area, followed by a visit in person in order to tag commercial and government places with names and addresses. The effort to put Camden on the map was part of a course-wide project undertaken by GIS students, including myself, at Rowan University.
First order of business was to load up JOSM, Openstreetmap's desktop editor, and examine the location. Next I added NJ GIN's 2012 imagery layer. Here is what was already mapped placed on top of the imagery:
As you can see, all of the roads and highways were already mapped. They also appear to be, more or less, mostly in line with the imagery layer. At least one land use polygon designated as "recreational" appears to have been placed as well. However nearly everything else is uncharted territory.
For the next step, I drew in polygons for every building. Where necessary, I did this making use of JOSM's tools to orthogonalize all lines to 90° angles. Here is what the map looked like after this step:
What buildings actually are is difficult to determine from above, so I left their values as 'building=yes' for now. The next step in this project involved visiting the area to take notes. Luckily, walking-papers.org enables easy printing of maps in Openstreetmap. I took advantage of this to print out a map of my own to mark down where all the businesses and government offices were during my visit.
Visiting the area on foot also gives information from a perspective not obtainable from aerial imagery. For example this building appears to be boarded up and closed.
Upon returning home I tagged all the commercial, government, and named apartment complex locations that I surveyed. Taking a few pictures while there also enabled me to add in addresses, phone numbers, and other ancillary information for some places. It was especially helpful coming across businesses that advertised information such as:
With my map update now nearly finished, I took a second look at the preexisting road and land coverage data to make some final touches. Comparing the aerial imagery to the roads showed a few inconsistencies. Things were a little, off to say the least. For example, in the interest of cartographic accuracy, I redrew some roads to not show them intersecting at angles that were not true in reality. I also gave the park polygon a face lift to show that its boundaries fell between certain roads, and beyond them. The next few picture slides show the before and afters.
And now this Camden neighborhood is fully mapped and uploaded on Openstreemap! Along with many other Rowan students during the fall of 2013, most of the city of Camden was mapped for Openstreetmap in a class wide project that sought to broaden the skills of students in open source non-ESRI software.