On Wed, 8 Apr 1998, Ron Daniel Jr. wrote:
> >> When an HTML document receives a URN (DOI-like or other) and it can be
> >> resolved into a URL (via a registered resolution service known to your
> >> browser?) it does not make very much sense to paste the URN in the meta
> >> tag of the document itself (that is like making a self-reference).
>
> Actually, I think it make a LOT of sense to put the URN into the
> document. Just as books carry around their ISBN, document could
> contain their URN. That way, if it is no longer offered from the
> original URL, you can go back to the URN and get a new URL for it.
>
I agree that a URN embedded in the resource is useful, but not for
retrieval of that same resource.
A URN, to be useful for retrieval purposes, should preferably be present
in a description external to the resource and retrievable independently
from the resource.
> >> Hyperlinking mechanism:
> >> -----------------------
> >> It seems to me the Internet community would be looking for a solution
> >> similar to the URL one, where you can either enter a URL at the browser
> >> window level or you click on a hyperlinked URL (<A href="">) in a given
> >> document and the browser fetches the requested document.
> >> Does anyone know if/how URNs are/will be hyperlinked and resolved using
> >> HTML encoding and HTTP protocol?
>
> RFC 2168 and 2169 discuss such a mechanism. There is supposed to be some
> free code that plugs into one of the two major browsers, once some
> licensing terms are worked out.
>
Does anyone know if the "magic" DOI-button, Frank referred to earlier,
works according to this mechanism?
If not, how does it work?
thanks, for your comments!
gr., Titia
------------------------------------
Titia van der Werf
Library research
Koninklijke Bibliotheek
National Library of the Netherlands
The Hague
The NETHERLANDS
e-mail: [log in to unmask]
tel : +31 70-3140467
http://www.konbib.nl/persons/titia/
------------------------------------
|