Re: Why the Web needs to change

• Mail folder: WWW Talk Jan 94-present
• Next message: Piglet: "html+ - remove <p> and </p>"
• Previous message: Daniel W. Connolly: "Re: URL decisions in Seattle, & changes "
• Maybe in reply to: Bert Bos: "Re: Why the Web needs to change"

Errors-To: listmaster@www0.cern.ch
Date: Thu, 31 Mar 1994 19:39:09 --100
Message-id: <9403311836.aa10042@ainur.ee.surrey.ac.uk>
Errors-To: listmaster@www0.cern.ch
Reply-To: ee01th@surrey.ac.uk
Originator: www-talk@info.cern.ch
Sender: www-talk@www0.cern.ch
Precedence: bulk
From: Piglet <ee01th@surrey.ac.uk>
To: Multiple recipients of list <www-talk@www0.cern.ch>
Subject: Re: Why the Web needs to change
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
Content-Length: 2379

This is an outline of and comments on replies I have
recieved...

****

> From: Marc Andreessen <marca@eit.com>
> Message-Id: <199403301018.KAA11229@threejane>
> Date: Wed, 30 Mar 1994 10:18:02 GMT
>
> The IETF URI group has been working for decades on defining such a
> beast -- the URN (Uniform Resource Name).  It's currently expected by
> the year 2009.

****

> From: "Peter Lister, Cranfield Computer Centre" <P.Lister@cran.ac.uk>
> Message-Id: <9403301036.AA02573@xdm039.ccc.cranfield.ac.uk>
> Date: Wed, 30 Mar 94 11:35:59 BST
>
> You just described URNs and local caching. URNs are not by any means unique to 
> WWW in any case and are discussed separately by uri@bunyip.com; don't 
> underestimate the scale of the issues.

How do I subscribe to uri@bunyipcom ??
>
> Local caching is happening now - there is a UK cache at unix.hensa.ac.uk.

It may well be, but what's the point unless my client know about it?
My argument is that my client ALWAYS BUT ALWAYS asks its `local'
server for the document, and that server responds with an uptodate
copy, be it locally cached, `area'ly cached, or from the source.

****

> From: "Roy T. Fielding" <fielding@simplon.ics.uci.edu>
> Message-Id:  <9403300815.aa17472@paris.ics.uci.edu>
> Date: Wed, 30 Mar 1994 08:15:49 -0800
>
> So, to make a long story short, you want hierarchical, caching proxy
> servers that use location-independent URNs.  Join the crowd ;-)
> Actually, I think you will find that URNs are only useful when the
> document is a fixed standard, such as an Internet RFC, or externally-
> published book, but that is an even longer story.

I disagree.  That's the whole point of checking to see if the
document is uptodate.  In fact, I am quite willing for the URL to be
the URN, just that the way the client gets the document be different.

>
> I suggest looking at the hypertext archive of this mailing list
> at <http://gummo.stanford.edu/html/hypermail/archives.html>
> and reading all the threads on caching, forwarding cache requests,
> and URNs.  This is going to take a while, but it's the only way to
> find out what's already been discussed.

I've looked here, and couldn't find much... maybe I'm looking for the
wrong keywords... any suggestions?

Timothy Hunt

<a href="http://www.ee.surrey.ac.uk/People/T.Hunt.html">About
Timothy</a>
<a href="http://www.ee.surrey.ac.uk/">About his department</a>