Today let us talk about an interesting application of internet technology. My inspiration for this was an incident a few years ago, I believe, in Oregon. A family was lost in the wilderness when their car broke down on a road that (they did not realize) was just closed for winter. Several rescue parties were launched by land and air to comb the surrounding areas. Time was of the essence as it had already been a few days and they were without water, food and were exposed to the elements. Similar situations on a larger scale could arise if there is a large area affected for example by an earthquake or hurricane and a search is on for survivors. (We keep hearing about the big one that will hit sometime in CA) How can information technology and crowd-sourcing save lives in these situations?
Google, and /or Microsoft have access to high definition satellite imagery. (Yes, it is scary that they can watch anybody if they choose to play big brother, but that is a topic for another day). Those images refreshed on that day can be divided into smaller rectangular grids that can be selected by volunteers (who can be anywhere on the internet). They carefully (visually) scan the imagery and can identify if somebody is within that grid or not. To allow for human error, each grid cell can be given to 2 or 3 volunteers to scan. This allows a high level of parallel processing and scanning within a short time frame, by people who are far away from the affected area and want to help.
When I once mentioned this to a top executive, there was a concern around privacy / security /misuse etc. That is a very valid concern and can be mitigated by pro-actively establishing a bank of volunteers. Interested volunteers would self register and get prescreened / background checked for specific countries/ geographies indicating that they will be available for these scans in such situations. They could also go thru a self paced training on how to better detect people in these scans. When such an incident is unraveling on TV, these volunteers can login and depending on their permission level (established during prescreen) can choose the grids that are available. This way thousands of volunteers can be leveraged in a short time to scan large areas to make a difference of life or death.
I realize there are several limitations of this idea as it does not help in densely wooded areas or when it is cloudy. It cannot also find people trapped under debris, but can point to debris which could be a possible trap. It can still be used in post earthquake, hurricane situations where usually roads are the first one to get affected, hampering access for search & help parties. An automated option is to build software that automatically detects motion in an area (based on before and after images) and alert rescuers. It is high time to unleash such technologies for peaceful purposes to save lives. It is time we proactively plan and innovate to avoid deaths and suffering in disasters and mishaps. Some other day I will talk about commercial uses for this, for example in law enforcement and to allow remote monitoring of construction sites by businesses.
Prasad
tvprasad 