Update: This case has purportedly had a tragic, very sad outcome. From the little the police have released — they have been extremely tight-lipped — it sounds like the grainy couple of frames from the high school camera up the street were the roots of this entire case (which led them to a suspect, who then led them to a Home Depot video from a city 70km away), without which...who knows where they would be today, or how many more victims the culprits would have claimed. I remain more convinced than ever of the incredible value of pervasive, decentralized monitoring.

Eight year old Victoria Stafford went missing from the nearby town of Woodstock on April 8th. She had left her school at the end of the day, starting the short walk home alone, but purportedly never made it.

Video from a high school up the street (see the enhanced version as well), discovered the next day, shows Victoria walking with a thus far unidentified female (a sketch, purportedly based upon a witness account, has been released. Given the sudden appearance of this supposed witness account two weeks later, it seems to be an attempt by the police to put a little pressure on a suspect to see if they break or make a panicky misstep).

The town in question, Woodstock, Ontario, sits on a major highway that goes from Windsor to Toronto and Montreal and beyond, with a nearby branch going to the Buffalo area. It seems noteworthy that Victoria’s public school is a very short distance from an onramp.

I find this case distressing. These things generally don’t turn out well.

Whether the abduction is real or not (there’s a general cynicism about cases like this because of vile, murderous sociopaths creating a cry wolf situation, leading many to automatically disbelieve), it usually forebodes very bad things.

Of all of the video cameras that blanket our society, the best they’ve got — at least that they’ve publicly announced — is the single grainy video from the high school (see the graphic I made of approximately the zone covered by this video, as determined by correlating landmarks with satellite imagery.) There were lots of people around, but history has shown time and time again that people are really, really terrible witnesses of anything, and that seemed to have held true in this case.

Who is the woman in the video? Apparently no one knows.

This has me thinking about pervasive, distributed monitoring. Where in time of need – like an incident like this (yes, “think of the children”) – swarm media and electronic capture can be combined to zero in on the truth. Preferrably in a way that utilizes the enormous talent and load distribution of the public.

High definition video capture is becoming dirt cheap. Solid state storage is rapidly evolving.

I’m not talking about the Big Brother 1984-style of central government monitoring, with the endless pitfalls and abuses that entails, but rather a situation where almost incidentally most everything is recorded by the public using a distributed array of devices.

It seems inevitable, for instance, that in just a few years every car on the road will have forward and rear facing cameras. The former is already in place on most police vehicles (helping to keep police in line as much as to capture public malfeasance), and the latter is making inroads on large vehicles that are prone to backing over people and things.

Video retention will inevitably come next, under the auspices of road safety (similar to how your car is ready and willing to rat on you for speeding if you get in an accident). Soon it won’t be an option, but will be a legal requirement for using the roads.

“Sorry, bub, but your car’s video system clearly shows you blowing through that red light. The other guy’s video system shows that he was in the right.”

99.999% of the time the video loops over and is erased and inconsequential and irrelevant, but every now and then it serves an important purpose in getting to the truth.

Imagine, for instance, that they could put out a call for anyone who drove down that street or that neighbourhood in the time period in question (presuming they don't already know from telemetrics via systems like OnStar, again soon to be the legal norm as every road becomes a toll road with vehicle self-reporting), from which they got a number of different time and position videos.

Videos of vehicles parked in the vicinity. Of the woman in the white coat waiting.

Add the video capture on private buildings of all sorts (homes, businesses, parking lots, etc). Merge it all together into a exhaustively documented, fact-based accounting of what happened.

Cellphones of course play a part as well.

Someone premeditating a crime would likely leave their cell phone at home (though I would wager that the woman in the white coat probably had one on her), or disable it in some way, but it would nonetheless allow the net of discovery to close in on the truth. The lack of certain data is often indicative.

Say to contact a guy in the area (via text message of course) and find out that he was in the neighbourhood taking pictures of doors, and wouldn’t you know it but he happened to have one that has a suspect matching the description walking towards the crime scene 9 minutes before the crime.

That’s if you even need to contact him. Maybe he uploaded his GPS-and-time-tagged photos to Flickr, and a police investigation entails a photo search, again drawing from the enormous distributed capture that happens every day.

It is neither a utopia nor a dystopia, but I think pervasive, distributed, decentralized recording and archiving will be a good thing for society.