16 February 2012

Discussion of Readings 3

Goldberg, Wilson and Knoblock, 2007, "The current state of geocoding" URISA Journal



žogs said...

Awesome article!

Kevin said...

Does a street address correspond to a polygon or to a center of mass (a point)? As someone who first heard the term geocoding about two weeks ago, this strikes me as a kind of fundamental question behind this article from Goldberg, Wilson, and Knoblock.

The question has ramifications for all phases of research, from defining problems and questions, to collecting and analyzing data, and displaying results as information or knowledge. It will never converge in a single correct answer; the answer to this question has many facets. The good news about the multi-faceted question is that the technology of geocoding is increasing the attainability and usefulness of spatial data. And the tech seems to be in the phase of the curve where it is increasing faster than anyone can really track it. So with that happening, maybe we'll be able to have geocoded data as points, lines, and polygons and be able to use the most appropriate source to get the best answers to our geospatial questions of the future.

David Hanrahan said...

I found Goldberg's article interesting for several reasons. Mainly, the article takes what I originally thought to be a simple concept of referencing a physical location in digital space and illustrates the complexity, uncertainty and variety of academic research surrounding the topic. I also thought that the discussion of uncertainty with locating corner addresses increasing as one gets more rural or deals with environmental hazards fascinating. The example of the frustrating use of PO box #'s provoked me to remember when my parents address, surrounded by corn fields, was changed from Rural Route 3 to something that I can never remember like 27365 280th St. I wonder if this had something to do with geocoding? I also found it fascinating to think about some of the ways in which it might be illegal to use or distribute geocoded information where "requirements necessitate the geocode for an individual's location to be nonidentifying." And, lastly, the complexities involved in coding information to an address with multiple units or where the address is a complex of buildings.