http://www.wired.com/dangerroom/2012/04/air-force-drone-data/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+wired%2Findex+%28Wired%3A+Index+3+%28Top+Stories+2%29%29 (http://www.wired.com/dangerroom/2012/04/air-force-drone-data/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+wired%2Findex+%28Wired%3A+Index+3+%28Top+Stories+2%29%29)
And this is where the USAF logic comes into play.
The problem of being able to collect more imagery than anyone can analyze(interpret) is almost as old as the technology. I began my imagery analysis career in 1962 and can vouch for the problem having existed then. We have always had the capability of collecting more pictures than we could look at and it isn't likely to change anytime soon. There have been numerous attempts at creating technology which would automatically analyze photos(images, etc) and they have always been lacking. It simply takes a human eyeball, with a human brain behind it, to do an adequate job of analyzing the image and reporting what you see.
BTW: I continued in the imagery world until I retired in 2003, so I am not mired in the lost years.
This has and always will be a problem.
Our technology has always been ahead of our ability to actualy process and analyse the data we collect.
Even back in the days of still photography....it would take days for photo analysists to go through it all to get any usable intel.
Even in science we see this. Astonomers are always going back to old photos and data collected and seeing things that missed before.
Throw in the OPS tempo of the RPA and the urgent need for real time intel....there is simply not enough people or assets to process all the data the collect.
Joe Public has a similar problem because digital photos are essentially "free", so people have hundreds they never look at.
Some of this is also self-correcting - current intel "expires" on a FIFO basis, and the history is only interesting for an investigation, etc.
so there's probably a core set of data that needs to be reviewed, and the rest goes into data farms "in case".
Quotecurrent intel "expires" on a FIFO basis
not necessarily the case.
Quoteand the history is only interesting for an investigation, etc.
Semi-categorically false.