Because of its brand, TiVo always gets noticed when it does something that should be obvious. Approaching Google and Yahoo his week to form partnerships relating to video indexing, is obvious, but it still rates as news.
Faultline has been saying for two years that the only way that video content owners can truly bypass its existing route to market which takes much of its margin, is through the creation of a group of searching technologies that will reliably find video content files at any point on the internet, and make them available to play in living rooms across the world.
If that point is ever reached, why would anyone do anything except watch advertising free programming that is either free, pay per view, one off purchase or rental, apart from free to air TV which would be relegated to promoting new TV programs.
We are conditioned by the ever presence of TV into accepting that TV will always be the way it is, but it has always subtly changed. Free to air TV was once pretty good, but then the best programs were shifted into cable and satellite networks, which charged a subscription fee. More recently people have been buying DVDs and DVRs in order to isolate the programming they want.
If every program maker also sold its services directly as a protected file download over the internet, then revenue would rapidly switch to direct sale or rental over the internet, as it has for films, which have rapidly switched to DVDs over the past two years.
But none of that can happen if digital rights management systems are not standardized, or at least methods of interoperability found between different, existing DRMs. But even then, some pretty big brands will have to solve the technology issues surrounding indexing videos and use their brands to make a substantial proportion of files available over the internet.
We would really expect this to begin when Sony takes the top 500 films from its Columbia, Tri-Star and MGM properties onto the internet later this year. If it’s delivery system allows a company like Google to index the trailers for each film, and to integrate into its billing system directly, then we will have the perfect marriage of search and content. What TiVo might bring to this equation is a standard way of storing the content on a remote DVR, for viewing at any time.
This whole business is a chicken and egg problem, which will come first the internet content or the search engine to find it?
It is not guaranteed that TiVo is the right partner here. TiVo has about 3 million installed DVRs and because it is a brand in its own right, would want to limit the technology to its own devices or at least to devices built using its technology. Given that most DVRs are built owing something to the TiVo licenses, that wouldn’t be too limiting a factor, but it depends on your point of view.
What is needed is a fully standard way of conducting the transaction, finding the right piece of video, working out how to get rights to it (either buying it, using some existing credits to view it, or joining some kind of overarching subscription agency that has deals with every content owner).
The biggest problem is the indexing. Recently Faultline has heard everything from using the close captions, the initial video storyboarding, the script, right up to indexing each scene manually, as the solution to this problem. In the end a standard set of metadata needs to be built (in a standard format) into every piece of video. It needs to be easy enough to add retrospectively to existing video libraries, but it also needs to be functional enough to make all future searching feasible.
But video search engine technology changes with how you are planning to use the content. If it is to be delivered in file downloads, it might be a classic 110 minute movie to a TV or DVR, but it might also be selecting elements of this week’s sporting action to view on a mobile phone.
We need to search for a ‘home run’ or ‘goal’ and be able to identify the scorer at the same time as a cross indexing against the team name, for the sporting fans, while still identifying directors and major actors and even the screenplay writers and music directors, for any motion picture film. We need ‘genre’ for TV series and we need language versioning for all of them. It’s not a trivial task and the best we can hope for is if one service, for one sector, in one country, gets it right and then licenses and expands its technology as the idea takes off.
TiVo, not being, international, probably isn’t the company to do much of this, but Google and Yahoo, both being extremely international and full of ideas, might be.
Again the reason the TiVo discussions hit the news last week is because US newshounds are fascinated and obsessed by what is a highly popular, but mostly local, phenomena of TiVo. It’s not that other companies are not working frantically on this technology and there are thousands of patents on it, and an entire MPEG community (MPEG7) devoted to it. But it will need a breakthrough led by at least one, preferably two, strong brands, a little like Apple brought about with iTunes.
Another way that it will take off, is in Europe where IPTV is taking off. If all the content on all of the IPTV systems across Europe shared an index and if there was a clearing house (automated market) for cross system and cross border sales, then a customer in Sweden might purchase a film that is Video on Demand on a Homechoice system in London. There are issues of licensing to overcome, and there are issues of technology to overcome, but the search and index technology might look identical to a full internet video search engine