Networked Video in 10 Years : Networked Video == Parseable Video

Recently, I had a chance to discuss what online video might look like in the next 10 years with a group of very smart people at the Video on the Net: Beyond YouTube? breakout session at the Beyond Broadcast conference.

There are those who beleive that the video internet is currently going through it’s growth spurt much like text internet did in the 1990s. In some respects, I very much agree. The phenominal growth of activities such as video blogging, aggregation, playlisting and podcasting have gone far to make video a normal part of the web.

In other respects, I see a long road still ahead. Mike Lanza of Click.TV outlined a thought that is very pertinant. He stated that in the current iteration of online video, interaction, particularly social interaction occurs around the video with tools that are firmly based in the world of the textual web: tagging, commenting, sharing and the like. This is evident all through the popular video aggregators and video blogs, a quick trip to YouTube should illustrate enough.

Of course, there is more that is happening. People are remixing, starting to make comments in-time with video, people are creating videos in response to other videos but these are certainly not the dominant forms.

It is obvious that online video must and is taking a different form from the video that we have all experienced over the past 50 years (namely TV). It is on-demand, lean forward and nessecarily of limited quality and duration.

What is slightly less obvious is that current iterations of the popular online video formats are black boxes. They depend on the the text around them to provide the context and searchability. Metadata, which could provide some of this information is non-standard if existant at all. In other words, we are moving from a pure text internet to a multimedia internet but that multimedia in order to be useful needs to be described or put back into text in some manner.

Now, I am not saying this is a bad thing or useless thing. We can scan text, pull out key points in a non-linear fashion, navigate through text. None these things are easy with video in it’s current form. Video is rich and has tremendious emotional impact but it also has a lot of baggage.

One of the the things we discussed in our group discussion was “What would a video wiki look like?” A wiki being a very successful example of many of the things that the web was originally designed for. Wikis are open platforms for anyone to write, edit, erase, converse and otherwise publish content online.

Unfortunately, no one really had an answer. There are thoughts that collaborative editing platforms are getting there but editing is only one aspect of the language of video. There is also all of the production in the first place. Perhaps wikis just don’t translate into something where there is an infinite number of variables. In text, language adds some semblance of the finite, in video there isn’t a defined language with parseable portions.

My thesis here (and this is not new nor original) is that for video on the net to reach the relevance of text on the net, to be truly searchable, scannable and sharable it must be parseable at the very least. We must be able to hyperlink to portions, drill deeper within it, copy and paste it and search it.

What would a video wiki look like?

Last a note: Researchers, Academics, Cinematographers and practicioners who use video have been talking about these issues for as long as video has been around. This is not a new conversation but certainly one that is becoming increasingly relevant. One place that you might find people discussing these very issues is netvidtheory Yahoo Group.

Some meta learning from Beyond Broadcast 07

First of all, major kudos to Steve Schultze and the rest of the folks that took on the organizing this year. Great job!

Some other things I knew but was reminded of regarding events and conferences of this sort:

1: It is much better to be a participant (audience member) than organizer just for the fact that you can actually pay attention.

2: Most events, Beyond Broadcast included, work better as a one day events than two day events. Energy and participation are sustained much more effectively. Cutting portions out can be tough but in the end probably the right thing to do.

3: Given the opportunity, people will enthusiastically participate in documentation and “continuing the conversation” using web technologies. Utilizing something as simple as a common set of tags to be used across sites and platforms is an effective means to enable and encourage participation. People want to and will gladly contribute if the structures for contribution have low barriers.

Want some evidence: Try the tag “beyondbroadcast” on flickr,, youtube, google blog search, technorati, and so on.

ps. If anyone has an IRC transcript or if anyone did any reporting from Second Life (I am using the word reporting very loosely) I would love to be pointed to them.