[
Lists Home |
Date Index |
Thread Index
]
Dare Obasanjo wrote:
>
> You miss the point. An identifier is not a location.
>
> I could quite easily come up with two schemas, one with a target
> namespace of http://www.25hoursaday.com that describes myself as a
> GA Tech alumni and http://WWW.25hoursaday.COM which describes my
> CD collection. Now there is no question that both of these URLs
> refer to the same location on the web yet neither is there any
> question that they identify different things.
Actually, those can be treated as different HTTP (1.1) locations:
* http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.23
Web metadata will work much better if we consistently work towards a
situation where every "logical" resource has one and only one un-encoded
URI. In other words, we should work towards a world where our software
reflects the Web's model better than it does today. The "Host" header is
a good step in that direction. Namespaces taking a hard-line about
syntactic equivalence is also.
But this strikes me as more of a theoretical issue than one we run into
in real systems. People naturally reduce the number of names a resource
has because they don't want to mess up even the small amount of metadata
we have available today. For instance if you type in the wrong name for
a resource Google won't give you back a cache for it. And your page rank
could drop if Google doesn't know to merge multiple references to the
same resource.
--
"When I walk on the floor for the final execution, I'll wear a denim
suit. I'll walk in there like Willie Nelson, John Wayne, Will Smith
-- Men in Black -- James Brown. Maybe do a Michael Jackson moonwalk."
Congressman James Traficant.
|
