Most people have heard of the famous paper in Nature 2001 that said that papers that are available online are way more likely to be cited than the ones that are not. (Yes, apparently we do need papers to state obvious facts).
What's ironic is that none of the links to the paper, or to the author's homepage, from this citeseer
url work. Maybe someone could write another Nature paper which says that papers with no broken links are way more likely to be cited?!