I hereby coin the term "gurule."

The problem isn't "too many cross-references" -- it's how they are designed, in what context, for what kind of content, for what types of users.

You could just as easily confuse users with 2-3 cross-references, if they weren't done well.

Redundancy is very good, depending on the context. When I'm at the Atlanta airport, I'm always grateful that there are several ways to make sure that i haven't passed my concourse. In my addled, tired state, it's easy to forget which one I just passed. So the little blinky signs in the train car are a big help. However, signs that say "you are now at Hartsfield International Airport" would be much less useful :-)

andrew hinton ::

"gurule" is the best I've heard in a long time!

I have so many issues with Nielsen's assessment, especially when his pronouncements come without test results, screenshots, etc.

My main irritation is that he's forever equating usability with the lack of intimidation. He's also conflating multiplicity with complexity, and sacrificing both on the altar of usability. I can think of several interfaces in which the presence of a duplicate link or action button saves me a step. One example: being able to hit "next" at the bottom of a page instead of scrolling back up past text (an email, search results, etc). Nielsen might consider that a symptom-solution rather than a "true" solution, but I would rather scroll down as I read and hit "next" than go back to perform the same action.

I wonder how many credible users were tested on this one. I also suspect that the rules of footnoting, much like other constructions (grammar, etc) vary according to what style guide the writer is using. This makes Nielsen's statement "this gives you a normal footnote" [my emphasis] strikingly inept, and seems to reflect a personal problem with the tool rather than a tested user outcome.

Personal irritations may jumpstart a study, but they don't equal a study.