Message-Id: <9310251624.AA10840@merit.edu>
To: "Chris Weider" <clw@merit.edu>
From: "Fred Swartz" <fred.swartz@umich.edu>
Subject: Re: Permanence of URNs
In-Reply-To: Your message of "Mon, 18 Oct 1993 11:59:27 EDT."
<9310181559.AA22644@merit.edu>
Date: Mon, 25 Oct 1993 12:24:14 -0400
[Looks like I forgot to reply to this last week]
> From: "Chris Weider" <clw@merit.edu>
...
>In my opinion, the URN should be permanent. We discussed this in both
>Columbus and Amsterdam. Electronic citations require permanent URNs and the
>committment to maintain these mappings for decades, perhaps even longer.
I'm in complete agreement that certain applications require permanent URNs.
The question is who will provide it. My particular use for URNs
doesn't happen to have permanence as a requirement, although some
might have a very long life -- it will be a function of the how long
the underlying documents are useful.
So what do we do with URNs for which there are no longer URLs? I'm
making the assumption that the yet-to-be-defined protocol will allow
returning error information that could inform the client requesting the
URL field that there is no possibility of getting it. Alternatively it
should be able to suggest a URN for a proposed replacement, eg perhaps
a newer version of something.
>I agree that this would be a perfect function for libraries, who in the past
>have done essentially this function with paper books, where they take a
>persistent (and fairly permanent) reference, such as the LOC specification,
>and map it into their current holdings.
Exactly, although you noted that they only provide info on their
holdings. Perhaps they (or someone) could take on the new job of
keeping permanent citation information for the URNs.
-- Fred (fred.swart@merit.edu)