No Silver Bullet for Terrorist Tracking

Every time a terrorist event or attempt happens, numerous vendors flood the market with claims that they can solve or prevent them. Preventing is pretty much an impossibility for video surveillance or analytic solutions. There is simply not enough unique characteristics of terrorists for a video system to detect prior to an explosion.

For this week's NYC terrorist attempt, review a surveillance video of the suspect below:

Also read about details on the difficulty of searching for video in Times Square.

Update: NYT is now reporting that the man in the video above how who puts down a bag briefly and and pulls off a shirt 'might not be connected to the failed bombing.'

Update: Suspect arrested - definitely not the guy in the video.

A more basic but misunderstood problem is the issue of quickly reviewing video and finding suspects. Doing so is hard logistically, if not technologically. 3 reasons:

• Lack of standards amongst recorders: On any given street, there may be a dozen brands of recorders, none of which can be accessed by the same software. The police will have to connect to each one.
• Lack of network access to recorders: Recorders in private businesses can almost never be accessed remotely. This requires the police going to each location.
• Lack of controls to access recorders: Even if you had network access and the right viewing software, private businesses would be unlikely to provide total 24/7/375 access by the police. An agreement would have to be reached on when and with what permission these cameras could be accessed.

Can this be resolved in the near future? Unlikely.

'Standards' for recorders are in development but they are very early (see notes on PSIA and ONVIF recording). Even if they gain traction, it could take years before they are broadly adopted.

Network access can be achieved through opening up holes into internal networks. However, this is time consuming and is widely viewed as a poor Informational security practice. We are seeing a number of hosted/managed video providers enable 'plug n play' remote access but this is still very early (it could take a decade before such offerings are widely used).

With some cities are attempting to link private cameras, the technical barriers may it expensive and complex to do so, limiting its practical viability.

Addendum: High definition cameras are a technology that we do believe would help investigating terrorist events. It would address at all the issue of accessing video. However, it would help provide more details of suspects (the video above being a good example).