VAS is basically a way to make comments on a video. If you’ve used the commenting feature on Viddler or on YouTube, you get the general idea, and VAS can be used similarly. It can also allow you to link to images or video from a particular point on the timeline.

You might ask why, with Viddler and YouTube providing these services, another is needed. I’ll give a clear example: what if you want to annotate–as I do for an upcoming course–The Matrix. You can try uploading the film to some of the various video hosting sites, but you won’t get far.

One solution would be to get your own Flash or HTML5-based video player. You could use Flowplayer or Kaltura to either pull comments from a database as the video plays or lay them over the video itself. This doesn’t get you out of the issue of The Matrix, though. If you want thousands of people to comment on it, you will be hosting content that is copyrighted and that will last for all of five minutes.

VAS works by modularizing the process. It consists of three parts:

The first part is an XML file containing associations between particular times in the video–down to fractions of a second–an HTML segment that may contain plain text or HTML markup that links to other resources, and an indication of what File and Time this might be a reply to. The author of the association, and the time the association was made are also included in each annotation. Finally, for the XML document as a whole, there is metadata for the film itself, optionally including the name, year, and similar information, along with a link to a page with more information: IMDB or Wikipedia for example.

Because this is a framework, the VAS structure can be implemented with any player, but it makes sense that its first implementation is built as an interface module on the internet’s friendliest player: VLC. As a practical matter, reading VAS files and displaying them is not dissimilar to any of the half-dozen formats for subtitling that already exist on the web. But because there are more than can possibly be displayed within the film itself, the VAS interface provides a view to the side that is threaded. For textual comments, it fetches the comment and displays it. It is able to thread comments when they are indicated as replies. The interface also allows users to add comments to a current annotation file as they watch the film.

Because the interface can load and view multiple VAS annotation files, it can act as an aggregator. And just as RSS has OPML to link to multiple RSS files, you can aggregate multiple VAS files to that you can view the annotations of everyone, or just your friends and family.

Unlike web-based viewers, however, because VAS uses VLS, you are able to view the film itself in a copyrighted DVD format, layering the annotations on top of the film. For cult classics, you can find and exchange MST3000 style comments, or if you are part of a group of film aficionados you can add your own “if I were the director” tracks. The uses for education are limitless.

* You probably haven’t heard of it because it doesn’t exist. Won’t someone create it, or suggest ways that it can be done with existing software?