The purpose of this thesis is to investigate and test the usage of HTTP/2 in dynamic adaptive video streaming as well as to take a look into how it can be used to benefit prefetching algorithms used with branched video. With a series of experiments the performance gains of using HTTP/2 rather than the older standard HTTP/1.1 has been investigated. The results has shown no significant change to player quality and buffer occupancy when using HTTP/2, though our tests has shown in a slight decrease in overall playback quality when using HTTP/2. When using a linear prefetch of two fragments an average quality improvement of 4.59% has been shown, however, the result is inconclusive due to variations in average quality between different values for how many fragments to prefetch. Average buffer occupancy has shown promise with a maximum increase of 12.58%, when using linear prefetch with three fragments. The values for buffer occupancy gains are conclusive. Two implementations for non-linear prefetching has been made. The first one uses HTTP/2 server push to deliver fragments for prefetching and the second one uses client-side invoked HTTP requests to pull fragments from the server. Using HTTP/2 server push has shown in a decrease of 2.5% in average total load time while using client-side pulling has shown in a decrease of 34% in average total load time.

When watching interactive branched video, the viewer is given the opportunity to tailor the storyline of the video playback. This type of video puts the users in control of their viewing experiences and provides content creators with great flexibility how to personalize the viewing experience for individual viewers. When it comes to regular linear videos, the concept of having a playback bar visually presenting the playback is a well established implementation used for most (if not all) video players, but for interactive branched videos, that is not the case. Instead, most branched video implementations are typically custom-made on a per-video basis (e.g., see custom-made Netflix and BBC movies) and do not use a playback bar. With the goal to fill this void, a branched video player with a generalized playback bar that visualizes the tree-like video structure was developed in 2018 by students of the Information Technology program at Linköping University and will be made public with a publication in the near future. Within the preparatory work prior to making this branched video player public, this project included a three-phase user study, where we evaluated the playback bar and its implemented features and compared the video player with alternative designs. With this thesis, we highlight the value of a branched video playback bar and provide interesting insights into how it, and other design customization features, may best be integrated into a branched video player. Furthermore, we present an improved version of the video player in which the new implementations are based on, and motivated by, the results from the user study. Finally, we describe how further investigations may be done to evaluate the improved version of the video player.

Interactive branched video allows users to make viewing decisions while watching, that affect the playback path of the video and potentially the outcome of the story. This type of video introduces new challenges in terms of design, for example displaying the playback progress, the structure of the branched video as well as the choices that the viewers can make. In this thesis we test three implementations of working video players with different types of playback bars: one fully viewed with no moving parts, one that zooms into the currently watched section of the video, and one that leverages a fisheye distortion. A number of usability tests are carried out using surveys complemented with observations made during the tests. Based on these user tests we concluded that the implementation with a zoomed in playback bar was the easiest to understand and that fisheye effect received mixed results, ranging from distracting and annoying to interesting and clear. With this feedback a new set of implementations was created and solutions for each component of the video player were identified. These new implementations support more general solutions for the shape of the branch segments and the position and location of the choices for upcoming branches. The new implementations have not gone through any testing, but we expect that future work can further explore this subject with the help of our code and suggestions.

This thesis analyses how users' mouse behaviour during a video stall gets influenced by the type of video, either branched or linear, and by the presence of a playback bar. An experiment was conducted with thirty-two participants divided into six groups. Each group was watching a different combination of four videos with stalls, the first two videos belonging to the same type of video and either with or without a playback bar, and the last two videos changed in one of the two aspects. With mouse-tracking, these aspects were studied through the variables of mouse activity, average speed, average distance between the cursor and the playback bar, and the total distance moved on the screen. The participants also filled in questionnaires about their mouse behaviour, after watching each video, and their answers were later analysed through a thematic analysis. The results showed no significant differences between the groups in any of the main dependent variables. In general, within all groups, the participants moved the mouse very scarcely, indicating that the results about mouse movement should be interpreted carefully. During videos with a playback bar, mouse movements appeared to be concentrated to the stalls, focusing the movements to the bottom of the screen where the playback bar is located. Mouse behaviour during videos without a playback bar was more evenly divided between the different parts of the video and of the screen, or the user were not moving the mouse at all. Watching branched or linear videos influenced the mouse behaviour in such a way that branched videos seemed to engage the user to interact with the video player more than the linear videos. However, no difference was noticed between these conditions for active users during stalls when a playback bar was present. The thematic analysis gave clear indications that the playback bar was an important component for understanding a stall. Based on these findings, conclusions are drawn that a stall is a situation of watching videos during which mouse behaviour may be less affected by the type of video, and more influenced by the access to a playback bar. The playback bar was shown to be a source of information about the system and the situation.