One of the problems with describing this project is that, given it’s predominantly a HTML5 web application, until a substantial portion of the UI is constructed one finds oneself waving arms around frantically and using hypebole to explain what it should do, or could do, or what you hope it will do! And obviously arm waving doesn’t work so well on a medium like the web.
Although there’s still a number of bugs, a large part of the facade is now in place, meaning it it possible to give a comprehensive walk-through of what the tool will do.
Also, the terms used by the tool have been standardised — after much thought it has been decided to go with the same vocabulary employed by the semantic web and the Exhibit project. The terminology might be a little foreign to non-technical users at first, but hopefully they’ll get used to jargon like facets (and besides, trying to use more friendly alternatives is sure to cause more confusion in the long run!)
In the screen-cast we start by selecting a Creative Commons (BY license) video from You Tube. We then want to add a few items relating to this video — in this example the items are key people involved in the events of World War II. For each person item we can add a description and a list of web links to other resources. We add semantic data to each person by creating facets to document their sphere of influence and their nationality. Finally we drag thetrack bars to describe which passages in the YouTube video relate to each person.
In the Exhibit-driven semantic page generated by the tool (not shown on the screen-cast) the visitor will be able to use the facets to study the relationship between people (for example “show me all the American politicians”, or “how many scientists mentioned in the video were also involved in politics?”, etc), and/or they can simple watch the YouTube video and have the relevant people details appear or vanish on the page automatically, and in context.