Internet Info 1997 December/Internet_Info_CD-ROM_Walnut_Creek_December_1997.iso/ietf/urn/urn-archives/urn-ietf.archive.9607/000009_owner-urn-ietf _Mon Jul 15 13:03:25 1996.msg

Internet Message Format | 1997-02-19 | 5KB

Received: (from daemon@localhost) by services.bunyip.com (8.6.10/8.6.9) id NAA21516 for urn-ietf-out; Mon, 15 Jul 1996 13:03:25 -0400
Received: from mocha.bunyip.com (mocha.Bunyip.Com [192.197.208.1]) by services.bunyip.com (8.6.10/8.6.9) with SMTP id NAA21511 for <[e-mail]>; Mon, 15 Jul 1996 13:02:50 -0400
Received: from acl.lanl.gov by mocha.bunyip.com with SMTP (5.65a/IDA-1.4.2b/CC-Guru-2b)
        id AA22726  (mail destined for [e-mail]); Mon, 15 Jul 96 13:02:46 -0400
Received: from legiron.acl.lanl.gov (legiron.acl.lanl.gov [128.165.147.188]) by acl.lanl.gov (8.7.3/8.7.3) with SMTP id LAA03268; Mon, 15 Jul 1996 11:01:04 -0600 (MDT)
Message-Id: <[e-mail]>
X-Sender: [e-mail]
X-Mailer: Windows Eudora Pro Version 2.2 (32)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 15 Jul 1996 11:05:25 -0600
To: [e-mail] (YON - Jan C. Hardenbergh), [e-mail]
From: Ron Daniel <[e-mail]>
Subject: Re: [URN] What about getting stuff off of CDs?
Cc: [e-mail], [e-mail]
Sender: [e-mail]
Precedence: bulk
Reply-To: Ron Daniel <[e-mail]>
Errors-To: [e-mail]

At 07:33 PM 7/12/96 -0500, YON - Jan C. Hardenbergh wrote:

>I want to use URNs to get stuff off of CDs, with the browser knowing
>how to map the URN to the right mount point (some level of indirection).
>
>The idea is to explicitly give the browser permission to substitute
>a matching <NSS> without checking with a registration authority.

There are a couple of topics to address. The first is - "given an ID for
some content, can we look it up on CD instead of always going to the net".
The second is "what should that ID look like to conform to the notion of
URNs"?

First topic:
We certainly want clients to be able to optimize the resolution process
in such fashions. We want browsers to look in local caches or do any of a
number of things rather than go through the basic resolution procedure.

There are a number of technical details that the browser vendors will
have to confront to look at your CD for info, but that is between you
and them. We certainly encourage you to do it.

Second topic:

>Would "URN:x-vrml:<library name>:<NSS>" be obscene? Truly chastised forever?

Some comments and questions:

1) "URN:" is religious, you need to be able to cope with its presence or
   absence. I suggest you use it, but not everyone will agree with that.
2) If you want lots and lots of people to use this, "x-" is not appropriate.
3) You should (IMHO) still have information on the VRML naming scheme
   registered at urn.net. Assume that the browser looked on the CD and did
   not find the referenced object. Now they contact vrml.urn.net. Where
   should we send them? Is there info in the library name that can
   be used to build a domain name, so that we can just refer people to it?
4) Syntax issues - does the charset follow that of URLs? Are ':'s
   illegal in library names?
5) Namespace - who is going to adminster the vrml namespace to make
   sure that library names are unique? This also involves the same BS
   that DNS has had to deal with in terms of trademarks, name buyouts,
   ...

>The CD is a mirror of some sort, and we want a Namespace ID + Authority
>to grants the browser the right to use local data. This implies trust;
>speed is more important than document integrity.
>
>Is this just wrong?

No, although there will be some situations where document integrity is
more important than speed, those situations will be rare. In a later message
you said:
>I guess I want the handshake to have the browser tell the authority, Oh
>by the way, I have this library in cache, on CD, or locally, and the
>authority to delegate to the broswer that namespace.

There is no way to enforce this, short of having the authority pass back
a key that the browser uses to access the local material. Currently, if
the browser finds this material in its cache, or a proxy finds it, the
authority is not notified. Is notification *really* that important? If it
is, then you have a lot more work to do.


Regards,
Ron Daniel Jr.                       email: [e-mail]
Advanced Computing Lab               voice: [phone]
MS B287                                fax: [phone]
Los Alamos National Laboratory        http://www.acl.lanl.gov/~rdaniel/
Los Alamos, NM, USA  87545       tautology:"Conformity is very popular"