DARPA dabbles in real-world ‘Matrix’

Defense Department ubergeeks at DARPA are soliciting proposals for a comprehensive, searchable database of individual human lives encompassing every communication, encounter, transaction and even 'feeling' generated by a lifetime of social interaction.

It's called LifeLog, and it would combine a plethora of sensors, including medical sensors, along with a vast transactional database and an exhaustive catalogue of media encountered, all compiled and tweaked to preserve every scrap of data that a subject might talk or write about, or be observed to do or say or confront, or be reasonably assumed to intend.

"LifeLog must acquire data to capture both the user's physical experiences in the world and his or her interactions with other entities in the world," DARPA says.

On the level of intimacy proposed, this would surround participants with a tremendous outfit of wearable devices, probes, cameras and bugging devices for their phones, computers, televisions, fax machines and so on.

"Physical data is captured by various physical sensors and is stored as multiple data streams in appropriate formats at appropriate resolutions. Transactional data is extracted principally from a number of computer applications. Detectors, recognizers, analysis tools, and heuristics are used to 'distill' the data, associating metadata, flagging keywords, and otherwise preparing the data for further categorization in terms of representations at various levels of abstraction."

That's the easy part, the 'VoyeurDorm.com' on steroids bit. The trick will be to tame all that data and make it easily and intelligently searchable so that sense can be made of it. DARPA's a little vague on how they see that happening:

"Through inference and reasoning, LifeLog generates multiple layers of representation at increasing levels of abstraction. The input data streams are abstracted into sequences of events and states, which are aggregated into threads and episodes to produce a timeline that constitutes an 'episodic memory' for the individual."

"Patterns of events in the timeline support the identification of routines, relationships, and habits. Preferences, plans, goals, and other markers of intentionality are at the highest level."

So what's it good for? Assuming any normal human being would volunteer for this sort of endless "Truman Show" scrutiny, the research could be used to make machines impersonate humans more faithfully, to improve educational and training simulations and enrich VR games. At least there's a hint of that:

"Application of the LifeLog abstraction structure in a synthesizing mode will eventually allow synthetic game characters and humanoid robots to lead more 'realistic' lives."

The computer games and porn industries will certainly benefit richly, and may well provide the vast commercial market needed to sustain such an expensive undertaking.

But what about military applications? Better simulations and better battlefield agents for strategists, DARPA reckons. Of course those 'realistic humanoid robots' will come in quite handy when DoD decides to take over the government by systematically replacing senior civilian officials and media personalities with duplicate military automata. It's comforting to learn from the proposal request that the state of the art is such that they probably haven't done so already, in case you've been wondering. ®