The Future of Online Video

We have all grown up in a world where the unit of TV currency is 'The Programme' and where TV time inexorably moves to the right. You can see it on any EPG: blocks of half an hour or an hour where the future can be seen on the right, and the past is gone forever - unless you recorded it.

It is very easy to assume that this is the natural state of video, and that moving from broadcast TV to online video (on computers or smartphones) is just a case of adapting to a different user interface (mouse or touch screen) and a less casual 'lean forward instead of back' way of viewing.

And the first phase/generation of online players reflects exactly this: YouTube, Hulu, the MSN Video Player, and the BBC's iPlayer all do the obvious by moving TV onto computers or smartphones, adapting the familiar metaphors to the new environment and making catch-up TV something that you can do away from the main TV's large screen and surround sound setup. This works very nicely, and lots of people now watch increasing amounts of video content on computers, laptops, portable mp3/media players, smartphones, etc.

So what happens in the next phase/generation? What is the end-game? I've been thinking about this for more than 10 years, and I've been through research, invention, development, and founded a startup to exploit the ideas and commercialise them. I think that the next phase throws away those two fundamental characteristics of TV: programmes and time. Let's start with an analogy:

Suppose a door-to-door salesperson turns up at your door and instead of trying to talk to you, they hold up a portable DVD player and press the 'Play' button when you open the door. The screen flickers into life and a recorded sales pitch starts up. After about five seconds you know you aren't interested, but the DVD player plays on regardless. On and on. Killing any chance of a sale, and possibly endangering the well-being of the salesperson holding the player too...

It just isn't going to work, is it? The human being is going to be a much better salesperson than the DVD player because they can do two things which the DVD player can't: they can alter the pitch to suit the person who answers the door, and they can add or remove details and emphasis to make the best pitch to that person, based on the conversation, body language and any other feedback. There's no point pushing the virtues of the product to a child's uncomprehending gaze, nor is continuing to speak English when the recipient is talking to you in a different language...

So the next phase/generation does just two things to video: throws away programmes and time. Actually, the two are closely inter-related anyway. Suppose you wanted to catch-up on what has been happening in a soap opera? Would you watch the last few episodes, or would you ask someone who is a fan to tell you what has been happening? What you actually want is probably a few minutes of overview about the major events, mixed with something about your favourite characters or story-lines. A fan can give you exactly that, and the nearest current online, VOD or DVD substitute would be to fast-forward through those programmes, hoping that you can figure out what is happening. You don't need the complete programmes, and you definitely don't need them to play as they were originally made. You need the video to do what the fan does: tell a shortened version of just the highlights of the story that is relevant to your interests.

And that's what Real Time Content's technology does. It turns video content from a single recording of a fixed story, into multiple stories where the duration, the specific storyline, the amount of detail, the rating, the product placement, the references to dates, times and places, the politics, the language, and anything else that you can think of, can be adjusted, in real-time, on-demand. If you want 20 seconds of video content on a specific aspect of home insurance, then that's fine. If you want a news broadcast where science is the top story and there is no sports coverage at all, then that's fine too. If you want a brand reinforcement video where people can view themselves singing karaoke where there are trillions of different versions of the video, then that's okay too.

By taking away programmes and time, what you have is a video delivery technology that creates individual programmes from pre-prepared video clips, on-demand. Radio, DJs and stand-up comedians have been doing this type of thing with different media for years, and now the technology is appearing for video too. Just as stand-up comedians have a rough idea of what they will say in a performance, the actual jokes and stories that get used will depend on the reaction of the audience, and so each performance is different but shares a similar overall flow of story. By having an underlying set of stories that can be told using the video clips, with variations for length, for rating, for product placement, etc, the story that you see can be changed in response to your control, or feedback.

By reading this, you have been infected with an idea. From now on, whenever you watch online video you will want to know why you have to sit through the third recap and the fourth repeat of those flashy motion graphics. You will ask why you can't have just 5 minutes on what that character did in the first half of the season, or why the TV news programme always has to cover all the types of story when you always flick through some sections of a newspaper and read every word in others. You will realise that online video is not all about how many thousands of hours of video are available, but what you can do with that video to make it relevant to your needs, and to your available free time, and not the other way round.

Unfortunately, the cure for this infection is definitely not online video as we know it (or as it is often portrayed: Video On Demand, VOD), but a very different type of online video that is just starting to appear now. In ten years time, be prepared to answer questions like:

'So what was it like when all video was half an hour or an hour long, and there was only one generic un-customised non-personal version that everyone saw?'

Related articles by Zemanta

• Americans consume 33 billion online videos in December 2009 (downloadsquad.com)

Live Generative Art From Music MRQC 1bb (Mid-Quality)

Sometimes, just sometimes, the world seems to get locked into a wierd over-reaction. Here's just one example. I used software that I wrote to generate moving art from a piece of music. The piece of music was 'Darkside (Instrumental)' by Trystero, which is licensed to, and available for adding to videos on, YouTube as part of the 'AudioSwap' feature. Unfortunately, having uploaded the video, the music was detected by YouTube's copyright infringement check, and I got a rather forcefully written email telling me that copyrighted music from AudioSwap had been detected, and that details of my use had been passed to the copyright owners, who were a company called Rumblefish.

The ONLY option that YouTube provide in this type of case is a 'dispute' mechanism. So I disputed it. I said, excuse me, but this music is available via AudioSwap, and so I should be able to use it on a video on YouTube, and I lost. I seem to have lost because the dispute mechanism only seemed to allow me to say 'no, this isn't owned by Rumblefish' when what I actually said was 'yes, this is owned by Rumblefish'. Rumblefish, the licensors, listened to the music, verified it was what I said it was, and allowed me to keep using it, although the YouTube 'punishment' for me seems to be that they can put adverts on the page when the video plays, and the video is now marked with an annotation that says I lost a copyright dispute. (Neither the ads nor the annotation cause me any problem.)

In the meantime, I have now exchanged emails with Rumblefish, who turn out to be very nice, reasonable people, which is exactly what one would expect from a company that makes its living licensing music.

I really and truly can't understand why copyright has become such a strong area of contention that my usage of a licensed track was treated as if I had done something wrong, when I had only used it in the way it was licensed to be used. Rumblefish confirmed that I was okay to use the licensed music in the way it was licensed, by the way. But the detection mechanism does seem to trigger a whole series of actions in the YouTube machine... I'm not blaming YouTube either here, since they are obligated to protect copyright, and I'm assuming that their actions are completely appropriate in a normal case of copyright infringement.

In one of those 'wish I had been a fly on the wall' moments, I would love to know what Rumblefish thought, when they were asked, by YouTube's automated copyright infringement detection system, to listen to a piece of music which I said was licensed by them via AudioSwap, and which they then confirmed was licensed by them! Either they furrowed their collective brows quite a bit, or else they've seen it all before...

So, I'm now the proud posessor of a few emails which seem to be designed to assert something very strongly, when I understood and had already acknowledged that assertion, and I didn't need to be reminded in such a powerful fashion. And here's me trying to do things completely correctly, legally, etc. What sort of emails do you get if you don't care and ignore legality etc? They must be even more strongly written than the ones I got, which I'm now assuming were mild and routine. I just hope I never see such an email, because it must be truly formidable.

As I say, and as I will keep saying, it is a strange world.

My visualisations now on YouTube!

Sometimes it takes me ages to 'get around to it', and this is certainly true for putting more examples of my music visualisation research onto the web.

One of the key stumbling blocks has been grabbing video in sufficient quality, but buying Snapz Pro X from Ambrosia software has solved this (very nice software - I've been a happy Ambrosia customer since my G3 days).

So you can now view some video examples here