OK, over in [post=7023357]another thread[/post], there’s a malformed link. Yeah, like that’s rare. The link was http://"http//www.bartleby.com/19/1/ . Unsurprisingly, when I clicked on it, I didn’t get the Bartleby entry which panamajack intended. What I got instead was http://www.microsoft.com.
Except I’m using Firefox, on a Macintosh. Why the heck would a Mac running Firefox redirect a malformed link to Microsoft, of all places? Have the tendrils of evil really reached that far?
When Firefox detects a string which is not a valid URL in the address bar, it launces a Google “I’m Feeling Lucky” search, which takes you to the first Google result for the entry. The first Google hit for “http://” is Microsoft.com. I assume that’s what’s happening here, although I’m not sure why it’s truncating it this way.
Which means, “if you get an unknown keyword, do a Google ‘I’m feeling lucky’ search on it.”
As for why it’s truncated - I’m assuming Firefox parses along the URL line, looking for a valid URL. s soon as it hits something that prevents it from being a valid URL - in this case, the double quote http://", it feeds everything that preceded it to the Google search. Google (AFAIK) ignores “special” characters like : / " so you end up searching for just “http” without the quotes.
The next question is, why should “http” make Google think of Microsoft? Probably because www.http.com is a lame search site (don’t bother visiting - it may have cooties) that seems to want to redirect me to microsoft.com.
Because just about all links use it, and Microsoft scores higher than any other site in Google’s ranking system of cross-linking sites. This in turn will be due to the Microsoft website’s size, turn over and frequency of linking (although no-one can say for sure because Google keeps its exact ranking mechanism a trade secret.)