The search for intelligent television

The electronic program guide (EPG) on my digital cable box is next to useless, it’s a vast scrollable entity with no search function, same goes for the PVR. There are categories and various ways to jump between days and pages, but it’s not like searching on the web, which is what you really want, that and a type recommendation engine, oh and social media connectivity so that I can see what my friends and contacts are watching, recording and recommending, or not as the case may be. It’s rumored that “Google TV” might revolutionize the televisual experience somehow and wrapped up in those rumors is the idea of it bringing search to the experience (and more ads, no doubt).

It’s all tied up in the notion of information overload and our ever increasing need to see, read, listen and experience as much as we can pack into our waking hours. Indeed, back in the 1980s when I was still a student, I borrowed a “subliminal” learning cassette tape from a friend that was supposed to help you cram for exams even while you slept (bullshit, of course, there are no such thing as subliminal messages and brainwashing, thankfully).

There is a simple answer to information overload, tune out, switch off and log out, but that’s not the answer for most of us who not only “want” to assimilate as much nuanced data as possible and to be entertained while doing so, but actually “need” to accrue that information for professional, domestic, and personal reasons.

As Google’s creators and many others before them recognized many years ago, as information grows exponentially so finding the information you want/need becomes increasingly difficult without some way to order, index, and abstract it. It’s not about location, location, location, we all know roughly where the information is – it’s on millions of computers scattered across the globe and connected with this structure we know as the Web, or strictly speaking the Internet for non-web data. It’s not even about gathering it up; anyone with large enough resources can aggregate all the data they care to. No, it’s about filtration, evaluation and integration.

On the day I learn that Google News is going to “employ” editors at major news outlets to select the best content for its aggregator and running in parallel with Rupert Murdoch’s imminent manual blocking of Google with a few quick tweaks of his robots.txt files, I read a research paper that says the goal is about developing intelligent agents to do the leg work for us.

Mila Nadrljanski and V. Batinica of the University of Split, in Croatia, point out that database users – whether trawling TV guides, web indexes, or research journals – don’t need a vast amount of information, they simply need just that data that correspond best to their needs and interests. While the concept of intelligent agents is not new, Nadrljanski and Batinica suggest that it is time to evolve this concept to fulfill their promise and to help us home in on the data we want, when we need it.

Research Blogging Icon Mila Nadrljanski, & V. Batinica (2010). Intelligent media agents Int. J. Intelligent Defence Support Systems, 3 (1/2), 128-138 : 10.1504/IJIDSS.2010.03368

Author: David Bradley

Freelance science journalist, author of Deceived Wisdom. Photographer and musician.