vog (a BETA videoblog)

This is a video that I have shown literally dozens of times in lectures, conference presentations, and seminars. It is probably one of the better examples I've made that demonstrates some of the key ideas of softvideo. In this work there are nine video panels. It is a triptych made up of three different views of Collins Street. One view looks along the street. One looks directly at the street, square on from one corner, and another is made up of snatches of views of buildings and things along the street. I filmed Collins Street because at the time my tram travelled down it on the way to work, so it formed a significant part of my daily commute.

Each video pane has an image track (a jpeg) that overlays. This is toggled by mousing into the video pane, and it has a timer attached so that after a predetermined interval the image disappears and the video below appears again. In addition the work has three simultaneous soundtracks. The default is all three playing at a low level, but mousing into the video panels varies these, so that the first triptych can have one commentary track at full volume, another triptych causes a second commentary track to play at full volume, and the middle panel allows the street atmos track to play at full volume.

I think of it as observational documentary with a multinear or interactive aspect. I have used it to demonstrate softvideo so much because it is a movie made up of nine video tracks, each being their own object in the finished work. There are nine sprite tracks, which control the sound levels and the graphics overlay, and of course three simultaneous soundtracks which can have their respective volumes varied during playback. You cannot do this using traditional video editing tools (which people still struggle to understand). Yes, I can have three soundtracks and layer them, and set their levels. I can change these. But once I export that movie for playback they are now fixed. Soundtrack one will be at level x, soundtrack two at level y, and so on. But in this project these vary in real time, during playback, based on where the mouse is. To help make this more concrete I suggest an experiment where instead of the street it might be a conversation between two people. Mouse onto person one we hear their internal commentary, mouse into the middle we hear the actual conversation, mouse over the other person we hear their internal commentary. This usually helps people to understand what is going on here. YMMV.