Found this interesting today when going through a site’s Web Crawl error report in Google Webmaster Tools. I was looking at the 404s (not found) report and found a number of what looked to be partial URLs pointing at the site from hundreds of affiliate pages.

Turns out that Google has tried to follow a URL it found in a JavaScript function. The problem is that this URL is only the beginning of an address that was to be constructed using other variables from the JavaScript function. Google obviously can’t run the function and put the whole thing together. It just found an “http://www.blahblahblah.com/and-so-on” and tried to go there.

What I find most interesting about this are the ramifications. Can you exploit this as an inbound link to another page? Does this pass page rank goodness? Would Google try to follow a URL in a JavaScript comment?

Interesting indeed.