In short, GeoLing is a handy tool for performing statistical analyses on spatial data: You can use data from dialect surveys, transform them into smoothed maps (via density estimation), detect structures that run through the data and find groups of maps that share spatial features.
We developed this program with linguistic applications in mind, but that should not stop you from using it for any kind of spatially conditioned data.
GeoLing was written as a part of the project New Dialectometry Using Methods of Stochastic Image Analysis, financed by the Deutsche Forschungsgemeinschaft (DFG) between 2008 and 2014. All the work has been carried out by people associated with three institutions:
- Institute of Stochastics (Ulm University)
- Lehrstuhl für Deutsche Sprachwissenschaft (University of Augsburg)
- Fachbereich Germanistik (University of Salzburg)
This software is licensed under the GNU General Public License v3.0
(published on 29 June 2007). The full text of GPL 3 is available at
https://www.gnu.org/licenses/gpl-3.0 and in the file LICENSE.
Although we took reasonable precautions and conducted extensive testing, we would like to stress that the software comes without any warranty or guarantee.
An up-to-date Java runtime environment is recommended, at least Java 7 is required. You can obtain the latest version of Java at https://www.java.com/download. Note that at least 2 GB RAM have to be installed for execution of GeoLing. Depending on the data and analyses, more memory may be required - which then also implies the need for a 64-bit operating system and 64-bit Java runtime.
Official binary release: You can start GeoLing by double clicking "start-geoling-windows.vbs" (Windows) or executing "start-geoling-linux.sh" (Linux). Have a look at the supplied documentation file "user_guide.pdf" for more information on how to use GeoLing.
Source code release: The source code is supplied in the folder "src" and it contains Javadoc comments. As a first step, have a look at the "package-info.java" file in the "src/geoling" directory. You can build the executable JAR file using the build tool Ant.