Geospatial Semantic Web Blog

Geospatial Web and Semantic Web are two major discussion topics of the Social Web Technologies course. In the past few classes, we talked about GIS, Google Maps API, geotagging and Geonames.

When introducing Geonames to the students, I decided to do a little experiment. I used Geonames as a tool to teach students the basics of GIS and provide them an opportunity to experience a “social-able” Geospatial Web.

An Annotation Competition

The experiment was relative simple. I spent few minutes introducing Geonames to the students. And then, I asked them to play a game. The class was divided into two teams: Team1 and Team2. Using Geonames, the teams competed with each other in identifying landmarks, buildings and roads that are located within the close vicinity of the UMBC campus. Each student signed up for a free Geonames user account. Using the wiki-style annotation tool provided by Geonames, students tried to annotate as many spatial features as they can in 10 minutes. The team produced the most annotated features would win.

To keep track of the features that each team had annotated, students were asked to tag their features using their team ID: “team1″ and “team2″. Using the Geonames search tool, I displayed the real-time progress in front of the class.

• Features identified by Team1
• Features identified by Team2

Lesson Learned

• It’s fun to use Geonames in a collaborative environment. Students enjoyed the process of creating annotations while chatting with each other and arguing about the location of a specific landmark. It was a social-able experience.
• Geonames has a relative open policy for users to make contributions — whatever the user enters, Geonames stores it. In general, this is a good thing. However, this policy can also lead to unintended creations of duplicated data. For example, because students were entering data simultaneously, we frequently saw multiple annotations of the same location were entered and they had different coordinates values assigned.
• It seems that using the Web as a platform can encourage non-GIS experts (e.g., students) to do GIS tasks (e.g., annotation). Not sure if this is an inherent feature of the Web or just because of the UI of Geonames is well designed.

LinuxWorld recently features an interview with Marc Wick, the founder of Geonames. In the interview, Marc revealed many interesting facts about Geonames.

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.

Interesting facts about Geonames:

• Geonames data initially came from NGA, USGS and the worldgazetter. Today the number of sources has grown to around 100.
• Geonames API initially supports a single search service. Today its API has grown to around 30 different services.
• Prominent web sites use Geonames, including LinkedIn, BBC and Nike. Popular mashup tools like Microsoft Popfly and Yahoo! Pipes feature predefined Geonames modules for building customized mashups.
• The largest Geonames user base is in Spain.
• Geonames runs on PostgreSQL and Tomcat. Its search implementation is built on Apache Lucene.
• Geonames has a “marketing team” of 30 people — Geonames ambassadors, who help with questions regarding their countries and serve as local contact person for national data providers.
• Geonames’ core development team consists of 4-5 hard working engineers.
• The biggest technical challenge is getting data — the team has more liberal access to US Army data than to Europe or Australian “public” data.

Geonames is a great service in a niche market. It’s open source model will help the project to continue to grow and expand its user base. The structured data support in Geonames , publishing data in RDF/OWL, JSON and XML, can foster the next generation of web services for knowledge sharing. Without Geonames, the development of many geospatial web applications (e.g., gnizr, HeyWhatsThat and ongmap) would be extremely difficult if not impossible.

Spotted on: http://tinyurl.com/ysgbj2

Geonames recently added 70,000 geocoded hotel data from three major hotel booking sites: hotels.com, diytravel and laterooms. This is part of geonames’ latest initiative to include more Point of Interest data into its open source geonames database.

Integrating data from different data providers is not a task without challenges.

The challenge in this task was to integrate and match data from various data providers. Names and addresses of hotels as well as data quality may vary dramatically among providers and it is often difficult to figure out whether two hotels are actually the same hotel or not.

Thinking about this data integration problem, at first I thought the problem could be easily solved if all data providers share a common hotel ontology, but later I realized it’s not that simple — at least building a such ontology is not straightforward.

Here are some data modeling issues must be considered:
Read the rest of this entry »

Reading Martin Soutschek’s question to my previous post triggered me to think of a new way to do photo geotagging. The idea is to annotate photos with machine tags that point to geonames features.

For example, you can tag Golden Gate Bridge photos with

geonames:feature=5352844

How does it work?

Read the rest of this entry »

Geonames announced a new experimental service that includes thumbnail images of geographical locations in its query results. For example, a search query to lookup London UK will return a thumbnail image of London. All thumbnail images produced by Geonames are acquired from various geographical location pages on Wikipedia.

Marc at Geonames Blog introduces this new feature:

Thumbnail images for wikipedia articles are a new experimental addition to the geonames webservices, the full text search and the maps mashup. Around a third of all articles on geonames have thumbnail images. A simple algorithm determines which image to use as thumbnail if more than one image could be parsed from the original article.

I think this new feature could be useful to many Web mashup applications, especially those that need to display photos of some arbitrary geographical locations — maybe in a semantic-web travel tool.

Geonames announced the release of its Geonames ontology v1.2. The new ontology has few enhancements. It introduced the notion of linked data and made clear distinction between URI that intended for linking documents and for linking ontology concepts.

Read the rest of this entry »