Date: Thu, 10 Feb 1994 09:17:33 -0800 (PST)
From: "Rob Raisch, The Internet Company" <raisch@internet.com>
Subject: Re: URN functional spec
To: "P. Desjardins" <Desjardins@citix400.doc.ca>
In-Reply-To: <1310*Desjardins@citix400.doc.ca>
Message-Id: <Pine.3.85.9402100933.A20287-0100000@hmmm>
EXCELLENT!!
A clear concise description of the work. Thank you VERY much!
A few comments from my perspective regarding P. Desjardins comments:
On 9 Feb 1994, P. Desjardins wrote:
> Here are my comments on the document "Specification of Uniform Resource
> Names", DRAFT, Feb. 8, 1994
>
> On sameness...
>
> - sameness of resource names (URNs) will imply "sameness" of the
> resources (is that concept defined anywhere?)
Sameness can only be defined by the publisher. Anything else
implies control or definition of intellectual content, which is an
abstraction and cannot be defined.
> On resolution...
>
> In the statement for this functional requirement, you say: "...To be more
> specific, for URNs that have corresponding URLs,...". To me this leaves
> room for URNs that have no corresponding URLs; is that the case?
Absolutely. There are many cases where a URN must exist and URLs
might not -- content which has expired or has been replaced, for
example. We really need to divorce the thing (or it's accessibility)
from the name of the thing.
>
> Further, bullet #4 in the "design decisions" section it is said that "Naming
> authorities should guarantee that somewhere, somehow, there is a mapping (or
> the potential for a mapping) to one or more URLs". I do not quite understand
> the implications of having to guarantee the "potential for a mapping". All
> this feels like an upgraded requirement...
A potential for a mapping implies that there is an authority
somewhere which can be contacted for information about the content.
Btw, this must be revised to say "...to zero or more URLs."
-- Rob Raisch, The Internet Company