Part 2: Requirements for a future epidemic early warning system
by David Luckham
This is the second of three articles on future event-driven early warning systems for epidemic outbreaks
I will ask you to suspend disbelief for a moment. The overall task from a top level point of view is monumental. Your primary reaction will be “how could this ever happen”. Well, let’s deal with the task first, and then the technology, and finally, in the third article, a discussion of how it could happen in small incremental steps. And always remember, the key is to start small and build outwards.
The Space of Events In and Events Out
First of all, the environment in which the system will operate is going to be the mobile Internet – the Internet as we will soon know it. It will contain very large numbers of hand-held devices – billions. These devices are going to be cheap and powerful. And you’ll see them in the third world – often as refurbished
giveaways. They will contain software to perform complex functions such as streaming video, audio to text and language translation. Also the OLPC2 project may add another source of inputs – one never knows! All of this means a vast increase in real-time mobile input from sources – in third world countries for
example – that are not available to systems like MedISys, Promed-Mail and GPHIN today. Download pdf