For a long time we turned to live media because it’s “immediacy” (more on the scare quotes in a second) was unsurpassed. You got literally current news from live updates. Of course “live” generally had a built-in delay. Not a huge one, but an appreciable (seconds to minutes) one. Still, this was pretty darn close to immediate, and we came to associate the concept of live media with the idea that “media doesn’t go any faster”.
Except that it does. Because, traditionally, “live” media has been filtered through the same publishing apparatus as other heavily produced media, and that has built-in delays. Modern communications technology, however, has given us access to unfiltered live media production (Twitter, FaceBook, and more media-rich applications such as live streaming from a cell phone camera). And it turns out that for raw information, text is faster, more compact, and generally more useful than the forms of media we’ve traditionally considered “live”.
With news-delivery seemingly eliminated as an interesting application for live rich media, we’re left with only one obvious significant use: simulating real-world space. This is the realm of live performance (in both the entertainment and the educational sense of the word). Lectures, plays, improv, music, etc. These are things that we understand work best in live contexts where there is potential for feedback and interaction, and feedback and interaction of the sort which impacts the performance directly requires real-time speed.
One thing worth noting here is that while a screen tends to be a great way to receive a live stream of audio and visual data, they make terribly restrictive systems for creation of that data. That is to say: most live streams are one-way. It is hard to perform while watching a screen because it restricts your movements and involves multi-modal interaction with mismatched turn-taking. (That is: most feedback systems have data incoming at the same time that a performer has data outgoing. This is in contrast to performances in live space where the tendency is to coordinate turn-taking so that data is only going in one direction at a time.)
One of the reasons this problem arises is that we still haven’t solved the turn-taking problem for online discussion. In face-to-face interaction we’ve come to intuitively handle multi-modal communication in ways that allow us to pass turn-taking information on a separate channel from the one we pass actual data-content stuff. Usually turn-taking is a body-language (visual) thing while data-content is vocal (aural). Most online interaction, however, is handled purely through visual data. Or, in the case of live voice chat, there is no good way to pass visual turn-taking data.
Which, at this stage, leaves me rather disinterested in live streaming of rich media. The point of going live should, I think, be that it enables interaction, but it’s not at all clear that interaction is enabled with current communication tools. I think it is well within our ability to create new tools which support interaction at this level, but at the moment I’m far more interested in interaction by audiences around media than in the way audiences impact performance. Consider, for instance, that watching a live stream of a concert is very different from going to that concert in person, even if you have the same audio/visual setup. The difference is that you aren’t sharing interaction space with the audience. It’s audience interaction that really drives a lot of the power of live performance, and I find that extremely compelling. But the weird thing to note is that audience interaction happens even with pre-recorded media (cinema, for instance), so there’s nothing particularly compelling about the live component there.
I realize that all of this is in extremely sloppy form, but that’s about how my thoughts run on the subject at the moment.