Somebody asked me the other day how come they get hits for webpages that don't exist, or for pages that are somewhere else. They had one of those fancy statistics pages, like AWstats or something, and they were pretty confused about what they were looking at on their statistics page. Turns out, they were looking at the referrer logs.
A referrer is a website that sent the visitor to you. Sometimes odd results happen because the referrer log gets this information from the web browser of the person looking at your site, and the information is only as accurate as that web browser is honest.
The way it is supposed to work is that if www.fark.com has a link to my site, www.springdew.com, and a visitor is at fark.com and clicks that link, my referrer log is supposed to show me that this visitor came to me from fark.com.
However, if someone was looking at slashdot.org and then typed in my web address into the address bar, or clicked their bookmark for my site, my referrer log might say they came from slashdot, or from their hard drive, or somewhere else. This is because web browser programs are not as smart as they are supposed to be.
