I know - but let's assume that since I and google could reach the document it was inadvertently put on a public server. Since it was a mistake it won't be filtered in robots.txt either.

For example, in the weeks before an Apple event, people tend to visit presumed unannounced product pages (by guessing) etc.