Video In PubPub : moving images in context

Abstract

Video has become so easy to make and distribute that it is becoming the most common way to describe many processes and ideas. However, it is often not effective at doing so. Within a context of a larger document, video is rarely a good citizen compared to text or images. It takes over the users attention and rarely points to other parts in the document.The video itself may be poorly made or it may demand too much time and investment on the part of the viewer relative to the content it provides. This thesis attempts to tackle this problem for a specific but useful circumstance: academic publishing. In this thesis, I develop and study two novel methods of using interactive video within an academic document for instruction and peer review. StepByStep explores how we can create a better tool for instruction and documentation of methods. A key property of good documentation is that it is easily skimmable for relevant information. StepByStep breaks down a video into steps and arranges them spatially so that users can skim and scroll through the steps to find what they're looking for. VideoComments is an attempt to create an alternative method of leaving feedback to a document. By pairing a video recording of a user along with their actions on the document, we allow users to reference images and embeds easily, quickly switch between different parts of the document and record their thoughts in real-time. Finally, this thesis creates an infrastructure in which interactive video applications such as StepByStep and VideoComments can be integrated into academic publishing standards and platforms.