Last week, at Demo07, an annual conference that showcases new technologies and startups, Suranga Chandratillake, a cofounder and co-CTO of Blinkx (pronounced “blinks”), was voted “Demo God” by the show’s attendees. The crowd was impressed not only by Chandratillake’s intelligence, but also by Blinkx’s technology, which allows users to search more than seven million hours of Internet video to find exactly the clip they want. Unlike most video-search outfits, Blinkx does not merely search the text or tags associated with a particular video. Instead, Chandratillake’s technique employs speech recognition, neural networks, and machine learning to create transcripts of the world’s videos; then, the words spoken in the videos can be searched. The method creates much more relevant video-search results. Chandratillake talked to Technology Review’s editor in chief about Internet video, the company’s ultimate aspiration to be a kind of remote control for all the world’s content, and some recent emerging technology projects.
Technology Review: What’s so great about Internet video? Why is it growing so fast?
Suranga Chandratillake: The infrastructure is in place now. The bandwidth and processing power are available. Video cameras are cheap and editing software is free (or almost). Ultimately, this all represents a massive lowering of the barrier to entry, which has utterly disrupted the supply side of this enormous global industry. Previously, all production was controlled by a handful of individuals, and the distribution of that content was tied to a small number of restricted mediums: the cinema, Blockbuster, cable TV. Changing the paradigm of entertainment production and distribution matters on multiple levels. In society, the democratization of media means we’re all able to participate in a global conversation or experience. Commercially, it’s suddenly viable to support a diverse “long tail” of content, which previously had a hard time getting an audience.
TR: What are the challenges to the growth of Internet video? How does Blinkx contribute to solving those problems?
SC: The two challenges are, first, copyright and licensing issues and the appropriate and fair monetization of content. People focus on the big media companies and their content, but the smaller publishers matter too. That’s the commercial challenge. The second issue is technical–that is, the filtering and search process. If there are going to be hundreds of millions of hours of video content online, we need to have an efficient, scalable way to search through it.
TR: When I first heard of Blinkx, in mid-2005, it was in the business of desktop search. What happened?
SC: The focus of our technology–contextual search–hasn’t really changed. We felt that the problem of search hadn’t been solved. At first, we applied our contextual search technology to the desktop, and with the proliferation of online video, we saw a great opportunity to improve the rich media-search experience.
TR: If video now constitutes 60 percent of Internet traffic (with some estimates saying that figure will rise to 90 percent within the decade), how much of that content is now searchable using Blinkx? Could you compare that with your competitors in video search, please?
SC: Blinkx is content and source agnostic, which means that we’re working to index all video content, wherever it exists on the Web, which makes us the biggest video-search engine. The other engines focus on a particular type of content. In the case of Clipblast and Furl, the focus is college-humor types of video. Truveo and Singingfish only index video that’s found in RSS feeds, which constitutes less than 10 percent of what’s available.