Date: Apr 11, 2013 Source: (
click here to go to the source)
Robert Cheetham, CEO and president of Azavea, spoke about the Web-based Historical Geocoder called Temporal Geocoder, that the company is developing for address-level temporal geocoding.
GISCafe Voice: Do you think this is the first time-enabled geocoder to be developed?
Robert Cheetham: There have been previous efforts to create time-based place name gazetteers. The China Historical GIS project
is a good example of a place name geocoder that has some similar ideas. There is a similar effort underway in New York City, led by the New York Public Library that is also aimed at place names. But, to our knowledge, this is the first attempt to create an address-level temporal geocoder. We hope to merge both address and place name geocoding into the same system.
GISCafe Voice: What types of technology will be employed in Temporal Geocoder's making?
Robert Cheetham: We plan to use Leaflet, Python, Django and PostGIS. There is also some parallel work being done by a sub-project of the OpenStreetMap project and we hope to collaborate with that effort as well. We plan to release the Database Editor under an open source license in order to make it possible for other communities to build similar databases as well as to cultivate a community around this type of work.
GISCafe Voice: How will the information for the historical aspect be displayed?
Robert Cheetham: We plan to create two basic software tools, both of which will be web-based. The first will be a database editing software tool that will enable people to indicate changes in the street network as well as street name changes and aliases. This Historical Street Database Editor will be able to display, a) the current streets; b) the street grid for a specific historical reference period; and c) a historical reference map that has been scanned and georeferenced.
GISCafe Voice: Where will the data come from?
Robert Cheetham: This initial NSF grant will only support the development of a prototype, and we are planning to build that prototype around Philadelphia data. We have many years of experience working with historical material in Philadelphia; we have several local partners with historical street maps that can be used as reference material; and we have a City with an open data policy, which helps to give us a head start. We plan to start with the contemporary road centerline plus add an extensive list of street name changes that we have assembled from both local historians and our own research. To these, we will add several keystone historical maps that have been digitized and georeferenced as well as paper maps available in the
municipal archives, the Philadelphia Free Library and other sources. If we are successful in proving the feasibility of the concept, we will be eligible to apply for a Phase II grant from the National Science Foundation, but we would like to have a working prototype at the end of this Phase I project.
GISCafe Voice: Will there be a way to compare historical documents from one time period with those of another?
Robert Cheetham: We may need to be able to incorporate scanned, georeferenced historical maps as reference material for developing the time-linked street map, but we do not expect to make historical documents available as part of the project. Rather, our goal is to be able to support more accurate geocoding of historical documents by the many archives, libraries and museums around the world.
GISCafe Voice: Will this be a relatively easy to use tool?
Robert Cheetham: Azavea is a skilled UI/UX design team that works on every one of our projects. Our objective is to make this a high quality user experience that may require some experience working with historical documents, but will not require a lot of training to understand. We have also recruited an Advisory Board that will be made up of experienced historians, librarians and archivists, and one of their roles will be to provide feedback on usability.
