Page is a not externally linkable
rogerd - 7:09 pm on Jun 6, 2004 (gmt 0)
I doubt if running your own spider would be practical unless you expected the theft to occur on a specific set of sites. You could use the major search engines to generate a list of sites based on keywords/topics, then spider them for images. That's a lot of effort, though. Visibly watermarking your images might be one way to discourage theft if you feel they are highly likely to be stolen.
Unless the thief is dumb enough to hotlink the image, the only way to find stolen images is to spider the entire web. I recall hearing about an image spider that checked for images with a hidden watermark, but AFAIK that's a paid service. Usually, it makes most sense to let the search engines do what they do best (spider as much of the web as they can) and use them to spot ripoffs.