On Fri, Apr 21, 2006 at 08:51:32 +0000, Chris Cannam wrote:
Paul Winkler:
OTOH, it's pretty obvious why this is the
case.
Imagine if it *did* have to resolve to something.
What would that mean?
That's a slightly perplexing detail even with the
situation as it is. What if the URI *does* resolve as
a URL? Would a host that looked up the URL and did
something with the results (doesn't matter what, but
let's imagine that the host defines the thing it will do
and does it consistently) be committing a fundamental error?
There is no particular reason for a machine to ever resolve that URI, it's
just an identifier, but a human might well want to.
- Steve