Take a basic understanding of KML, notions of Python to go through the Internet, add a bit of regular expressions and there is basically no limit to what you can do in terms of geo-mash-up.
Several space agencies (including CNES) are involved in the International Charter Space and Major Disasters. It aims at providing a unified system of space data acquisition and delivery to those affected by natural or man-made disasters.
On the site, there is a page describing all past activation. It is more than tempting to represent all this on a world map. Here it is ! You can click on one particular event to have more information about it.
There is basically no magic behind that. All information is available on the charter website. For recent activations, the website is generated automatically, so the structure is regular and easy to parse. The difficulty comes with older activation.
The first problem is that some pages are in a mess as they have been written by hand (human is no match match for a fully automated task). Playing around with regular expression solve that (but make the code ugly).
The second problem concerns the location of older events. No information apart from the event name is provided. The trick consist in looking for location terms in the description or the event name and find their latitude and longitude with geoname webservice. Of course, it lacks accuracy as it will sometime base its guess on the country name only… but it’s better than nothing!
Update 13/09/08: with some improvements, my proposal made it to the official charter activation map as you can see.