Dealing With Real Time Media

I attended a panel yesterday (at the Personal Democracy Forum 2008) on the live web (Robert Scoble, Bhaskar Roy, Max Haot, Keith McSpurren). A couple of observations about streaming live video, tweets, etc.

1. The production of these streams leaves little room - or time - for self editing.
2. The consumption of this data leaves little room for filtering - why would we consume real time data yesterday?

One problem that I see with the tools we currently have for handling this data is that they will follow the path of email. Email clients rapidly evolved to the 3-pane approach to consumption and never went anywhere else. There are no real analytical tools (except, perhaps, Xobni) for helping us deal with email.

I see an opportunity in this space for user facing tools that leverage the advances in social media analysis (including text mining, network analysis, etc.) to help us summarize and select the data, making the data more relevant, and the consumption more efficient.

Comments

[Akshay Java]

Perhaps the difficulty in self editing is compounded by the host of other services/tools that consume these streams. Lets say, I decided to delete an old Tweet. If consumption of my updates was limited to other Twitter users, I would still have some opportunity to edit/delete post facto. But now that the tweets are consumed via APIs through a host of other services, deleting it in one place does not mean it is gone. And ofcourse, since the value of a real time post quickly diminishes so does the incentive to self edit. In some sense it is almost like a "write only" stream.

[Sam Kuper]

I often read emails from mailing list archives (usually because, or to find out whether, someone else already asked a question similar to one I was thinking of asking). That is a significant kind of email consumption, and one that is open to analysis. It falls outside of the three pane model, too.

[Stan]

Live streams can have high production values, it is just very expensive to do. We will see perfect examples of highly edited live streams this summer coming from Beijing and Denver.

Even these high production live streams will have significantly smaller audiences than the resulting recorded versions that are viewed on demand some time after the event.

So why do people create unedited live streams? Generally because they are lazy, lack production skills, or are to cheap to create quality content.