Designing Mobile Applications with Locative Media for Navigation and Location Annotation

People create an abundance of digital content, like photos and social media messages. A lot of it is geo-tagged, i.e., assigned real-world location coordinates to indicate it relates to that location. This locative media forms a digital layer of information about cities and other areas. It constitutes a basis for providing services for the locals and visitors, for example help for choosing a place to visit and finding one's way there. This thesis studies how to support navigating and annotating locations with locative media. It belongs to the field of human-computer interaction and the approach is constructive, i.e., we have designed prototypes and conducted field studies where people take the systems into use. By analyzing the design process and field studies, we describe user needs and present design implications to help designers of future systems. Our empirical research probes the variety of location-based annotation types people would be likely to create and the value they perceive gaining by creating them. Furthermore, we describe the user needs related to sharing the annotations with specific people, as well as annotating ephemeral targets and larger areas than just singular GPS positions. We apply annotations and other locative media to tourists' navigation and create design implications for systems aiming to support it fluently and seamlessly. While observing city navigation behavior we identify clearly defined phases, but our studies also highlight that tourists move through the phases fluidly, reacting to the opportunities they see around them. We show that foraging theory can be used to guide the design of city exploration support tools but applying it requires some considerations. In fact, a lot of the exploration is quite exhaustive and the value maximization behavior foraging theory predicts becomes visible only when tourists do not have enough time or energy to explore comprehensively the area. The prototypes we have created vary in how they present the locative content since different phases of navigation have different requirements. Our evaluations confirm the advantages of egocentric visualizations of the content when aiming to support in-situ city exploration, as displaying the content from the point-of-view of the user makes it straightforward to know the direction it lies and minimizes the time needed to focus on the device. Conversely, for pre-planning done at home or at a hotel, third-person visualizations allow free inspection of any part of the city, regardless of where one happens to be, which suits planning well. Conducting field trials that let participants use our prototypes freely for the tasks that are relevant for them, allowed us to observe more realistic behavior than more controlled studies would have. This increases the confidence that the results apply outside our study settings, too.