Ed W wrote:
Tom Ritchford wrote:
There seem to be a lot of these... is there not some Apache or otherwise rule that can eliminate people who simply take Wikipedia images with no accreditation?
Usually the cheapest trick is to look through the logs for image downloads where the "referrer" is not another page on your own site.
as it happens, i wrote a script to do exactly this a few months ago. the intent was to find how much bandwidth this wastes (answer: not actually very much), but it could probably be extended to create a "worst referer" list of some kind...
http://www.knams.wikimedia.org/~kate/logmon.pl if anyone wants to have a go.
Ed W
kate.