If a browser implements, or can look up, links of type 'SCOnav', then
it executes the named function. The URL form is simply there to provide
a 'hint' as to where to find it, if in fact that browser has a way to add
in dynamic behavior (i.e. as HotJava does). All browsers would be expected
to understand links that begin "linkas:html3/..." but implementing other
sets of links is optional. SCO, for instance, might expand its SCOnav
types, strictly for use with its own browser, and use them liberally in
its documentation. Foreign browsers would just interpret the links as
'gotos'. No problem.
> what kind of thing should the browser expect to find? An enhanced
> style sheet? If it doesn't specify an object, then it shouldn't be
> called a URL, and probably shouldn't look like what. In that case,
So you object to "mailto:..." as well ?
> what exactly is a browser supposed to do with it?
Run that function, if it knows how. If not, just 'goto' the named HREF
as if the link type was not there.
I really don't see the big issue. Same as text-only browsers showing
you a graphic page, you lose some of the intended presentation style, but
nothing crashes, and authors know what will happen if they use odd links.
Same as if I put MPEGs on my page, not everyone can look at them. Doesn't
affect the integrity of the web. Behavior is the same, user ends up 'at'
the HREFfed page after the link has executed.
-- Craig Hubley Business that runs on knowledge Craig Hubley & Associates needs software that runs on the net mailto:craig@hubley.com 416-778-6136 416-778-1965 FAX Seventy Eaton Avenue, Toronto, Ontario, Canada M4J 2Z5