But in both cases it’s difficult to imagine how our technology could have saved the day without the sort of integration that’s likely to cause the privacy lobby and the political left – and I think we can all understand their fears – to dissolve into conniptions. 
The incidences I’m talking about are the recent terror attacks in Boston and Woolwich but I might just as well be talking about Baghdad or Makhachkala from the point of view of law enforcement need and operational functionality. In both Boston and Woolwich, the perpetrators were known to security services suggesting that radicalisation of terrorists is a process that takes a period of years and there are often multiple warning signs. 
It may be possible for video analytics of the future with reliable face recognition capabilities to integrate with public surveillance systems. These solutions could then keep an eye out for specific activities of suspects who have been flagged by intelligence or law enforcement agencies. Jogging alone they would ignore. Placing large bags in the middle of crowds at public events in their field of view they would not ignore. 
In the case of the Boston bombers, the primary perpetrator was known to the FBI and if a public surveillance system equipped with analytics had been able to identify this person behaving in a particular way at a public gathering, it’s broadly conceivable that reason-based analytics might have been able to proactively call a response. 
It’s obviously erroneous to suggest video surveillance cameras have some sort of proactive ability to leap off their mountings and defend the public interest. But as austerity bites into law enforcement funding worldwide it’s increasingly obvious that our reactive systems are going to fail the people they are meant to protect.
However, if there’s one thing the events of the past 10 years show us it’s that increasingly, video is an absolutely brilliant investigative tool, whether it comes from handicams, smart phones or HD surveillance cameras. Video surveillance allows crimes to be resolved in hours and for dangerous extremists to be neutralised. But what video does not yet do is provide advance warning. 
Something else no one’s talking about that I think is likely to become more common in the future is the use of data munching media processing – software engines that can take the inputs from thousands upon thousands of cameras and churn metadata into hard evidence. 
Such systems would be ideal for use with public surveillance solutions but they’d be just as useful for law enforcement bodies seeking to collate data from public sources – whether this be from the smart phones of citizens or video streams supplied by other organisations. And the ability to combine face recognition data with law enforcement files or images available in online ecosystems like Facebook could be invaluable as an investigative tool, too, legitimate privacy issues notwithstanding.  
While we’re on the topic of video surveillance, I can’t help but mention the public surveillance system in Nowra which was turned off after the Administrative Decisions Tribunal found the system was in breach of the Privacy Act. 
This decision was overturned by the State Government but there was something about the ADT’s gripes it’s worth installers and end users paying particular attention to – camera performance. According to the ADT, the footage of a plaintiff under the Privacy Act was of such poor quality he could not actually be identified. 
Now, I’ve not seen the Nowra system. It’s an older solution now and we all know things like camera quality, compression, motion blur, resolution, frame rate, WDR, depth of field, field of view, dust streaks and daddy long legs have a massive impact on image quality.  
But having seen some new cameras in the last few months, both fixed and PTZ, it’s impossible to imagine a modern system that does not offer crystalline images of targeted views in most conditions. Quality HD cameras now do so well in low light and backlight the idea any installation could not be fit for purpose seems a slight on us all. 
Regardless of budget constraints all you integrators and you end users in government too, have a responsibility to ensure that our public surveillance systems perform. It is foolish to pretend that there will not come a day when our needs demand that they do