The dimensions of context and its role in mobile information retrieval

Mobile information retrieval is a sub discipline that has been the focus of much research over the last 10-15 years. Drivers of this interest include the emergence of enabling technology, particularly position determining hardware and mobile communications. A second driver has been the increased availability of georeferenced information objects, including points of interest databases and conventions for adding geographic information to documents in a structured way using mark-up languages such as GML, KML, GeoRSS and GeoJSON. In many cases this associated geographic information is a simplistic spatial representation (such as a point), but increasingly commercial implementations are giving consideration to both an object's physical location and its sphere of influence: Google Places, for example. Such georeferenced information objects have the capacity to satisfy the diverse "where", "what" and "when" aspects of queries. As well as the development of enabling technology, there has been widespread acceptance that whilst information needs are ubiquitous, so long as information searches can only be conducted in a small number of static settings, then needs will frequently go unsatisfied [1].