I tend to agree that using cascading menus for navigation is generally
not a good idea. On a purely technical level, though, there are ways to
address some of the concerns you raise about acquisition and error
rates. Check out the the left-hand nav at --
at least in IE 6 for Windows, they've managed to address the "shortest
distance is a straight line" problem. This is easiest to see on the
"Information for... Business Professionals" menu.

BTW - I don't mean this as an endorsement of that nav scheme, but just
wanted to cite it as an example of a solution that addresses some of
the technical problems with cascading menus. Overall, the interaction
looks and feels nice (if you've got the right browser), but it seems
almost arbitrary that out of 27 entries, just 3 have sub-navigation.
The only two explanations I could think of were: (a) it makes the page
fit neatly in a browser window that's maximized to 1024 pixels high and
(b) the sub-nav under Business Professionals sort of makes sense, but
having sub-nav on just one entry would be too strange. Of course, then
there's the problem of the competing navigation areas/schemes, but
that's a whole other issue...

Adam Korman
[log in to unmask]

On Tuesday, November 26, 2002, at 06:44  PM, UED wrote:

> William H posted:
>> Has anyone any direct experience of this approach? Any good or bad
>> examples? What do people think? (I know Nielsen recommends against
>> cascading menus in any event.)
> In what little testing I have done on the site I'm working on now, I
> have found that cascading menus are, *in general* not a good idea for
> the web.


> Speaking of the
> problems with cascading menus, Snyder (2001) related that "the
> shortest distance between two points is a straight line. And that's
> exactly what users do -- they try to take the diagonal route to their
> desired menu option rather than across and down. And as soon as they
> move off the menu, they lose it (or even worse, get an
> unrelated one)."
> (ref:
> dwzone=web )
> This, in turn causes significantly higher acquisition times and error
> rates as Chaparro, Minnaert & Phipps (1998) found in their study...

    Tip of the Day: Use the archives to research common questions
     CHI-WEB: POSTINGS: mailto:[log in to unmask]
              MODERATORS: mailto:[log in to unmask]