* David Binger wrote [2005-08-29 20:30:36 -0400]:

> I assume that the problem with taking over the window.onload method  is
> that this might conflict with other code in the same page that also wants
> to take over the window.onload method.  Is that right?

Yes.

> Is there a risk, created by making this inclusion conditional, that a
> window.onload take-over conflict would not be detected until the page is
> tested with IE?

This is true; I detected the inclusion because it broke in a way visible on
Firefox but not code I had relative to IE.

However, the CSS required to make it work on all platforms is not included by
default, so I wonder if documentation would be a good enough fix.

If worried about the conditional comments (I'd not used them before myself)
changing to the addEvent approach is IE only from what I can tell (admittedly
have not spent much time researching) and serves the same purpose.

> The startList function body starts with a test of (document.all &&
> document.getElementById).  Can we just use the same test to decide if we
> should window.onload = startList ?

Sure... that works for me but we still should update the docstring or
comments or docs...