Location-based services have gained large impact in the last years. Also, meanwhile small and powerful end-user devices are present and available. At the same time, so-called mashup pages are spreading on the web. Mashups integrate content from existing sources into a new presentation. With the appearance of web-based map APIs, such as Google Maps, it has become easy to create mapping mashups, which present geographically annotated data on a map. Giving such a mapping mashup the possibility to utilize the user's position would make the mashup location-aware and provide additional user value.

This thesis explores how the user's position can be integrated into a mashup. Different approaches to achieve this are examined. An architecture for a system enabling location-based mashups is developed. The architecture integrates the user's position into mashups by extending the web browser. The Delivery Context Interfaces (DCI) are used as standardized interface for providing mashups with the user's position. Mashups are created on the user's device by a JavaScript client. Adaptation to the various data formats of the data sources is done by wrappers on the server side which convert the data into a uniform format. The architecture is implemented on a Nokia N800 Internet Tablet. An external Bluetooth GPS device is used for locating the user. Finally, an approach to integrate privacy aspects to the platform is presented.