GeoNames is a free and open source geographical database. Primarily for developers wanting to integrate the project into web services and applications, it integrates world-wide geographical data including names of places in various languages, elevation, population, and all latitude / longitude coordinates. Users are able to manually edit, correct and add new names with a user-friendly wiki interface. The data is accessible through a number of webservices and a daily database export.
Launched at the end of 2005, GeoNames is already serving up to over 3 million web service requests per day, and contains over eight million geographical names.
A growing number of organisations such as the British Broadcasting Company (BBC) and Greenpeace are ditching proprietary aggregation databases and turning to GeoNames. Other high profile organisations using the database include Microsoft Popfly, LinkedIn, Slide.com, and Nike.
Like many open source projects, it all started somewhat accidentally when founder, Marc Wick needed to develop a holiday apartments application. We speak to Wick about the project, the challenges involved in using public data, and the changing times in the GIS world.
What is your technical background?
I have a degree in Computer Science from the Federal Institute of Technology in Zurich (ETHZ). I worked as Software Engineer for Siemens Transportation Systems and for some major Banks in Switzerland mainly in an environment of Oracle, Java, Unix.
What gave you the idea for the project?
I was developing a holiday apartments application and needed gazetteer data for it. Since commercial data was prohibitively expensive I searched for free data and started aggregating it. Later I realized that other applications have exactly the same needs and are also aggregating free data. I subsequently released the data as the GeoNames project as it seems a waste of effort if a lot of people are doing the same task without sharing it. The idea was to share what I had already aggregated, if other people join in and help improve it, that is all the better.
(The theory being) that if we all team up together we will manage to build a better gazetteer than each of us would be able to do on our own. A lot of applications share this point of view and have switched from their own proprietary aggregation database to GeoNames. Among them are Greenpeace and the BBC.
Why did you make the project open source, and why did you choose the creative commons attribution license over other available licenses?
It is a common engineering principle to use the simplest possible implementation that could possibly work. I believe we should use the same principle when choosing a free license and pick the freest license possible. If there is no absolute requirement for restrictions like ShareAlike (SA) or Noncommercial (NC), don't use them. Let the project free. For many applications SA or NC licenses are as closed and restrictive as a commercial license. It is a pain so many 'free' projects are using only half-free licenses. There are more than a couple of 'free' geo data projects whose data we cannot use because they are using a less liberal license. They are as inaccessible to us as the commercial data providers. (Read more information on Creative Commons Licences here)
How much has GeoNames grown from when it was launched?
Initially there were three or five sources, namely the National Geospatial-Intelligence Agency in the US (an agency related to the US department of defence), the United States Geological Survey (USGS) and worldgazetteer. The number of sources has now grown to around 100.
The number of services in the web service API has grown from a single search service to around 30 different services from elevation to timezone. The number of supported administrative levels has grown from one to up to four for some countries.