Message-Id: <9311122205.AA20562@expresso.bunyip.com>
From: Peter Deutsch <peterd@bunyip.com>
Date: Fri, 12 Nov 1993 17:05:35 -0500
In-Reply-To: Mitra's message as of Nov 12, 12:07
To: mitra@path.net (Mitra), masinter@parc.xerox.com
Subject: Re: URN to URC resolution scenario
[ You wrote: ]
> Peter
>
> I contend that DNS is an existance proof that resolution can be accomplished,
> although I agree taht we have a more complex case and more work to
> do to make this thing usefull.
I'd agree with that. I was only trying to respond to
Larry's question with the observation that we should have
some interaction model in mind as it will possibly impact
our design decisions. Eg. gopher as an interactive protocol.
Thus, batch-oriented operations (such as coming in and
fetching an archie-like set of all menu items) is not all
that easy. This is fine, but if they'd had such a
requirement in place from the beginning, we might have
additional options in place now.
> DNS also gives us another usefull example. I contend that if DNS
> didnt give us *A* reliable way to resolve host names, then we'd still
> be using IP numbers for most things. I believe that without such
> a method for URN's we'll stay using URL's .
I agree, and as I indicated in my last posting think it
should be considered while designing the URNs.
- peterd
-- -----------------------------------------------------------------------------"The Internet destroys the Greek tragedy of time and space..."
- Daniel Pimienta <pimienta!daniel@redid.org.do> -----------------------------------------------------------------------------