Re: Proposal: WIT over USENET or Mail

Ehud Shapiro <udi@ubique.co.il>
Errors-To: listmaster@www0.cern.ch
Date: Wed, 15 Jun 1994 04:48:29 +0200
Errors-To: listmaster@www0.cern.ch
Message-id: <199406141847.VAA09422@shaldag.ubique.co.il>
Errors-To: listmaster@www0.cern.ch
Reply-To: udi@ubique.co.il
Originator: www-talk@info.cern.ch
Sender: www-talk@www0.cern.ch
Precedence: bulk
From: Ehud Shapiro <udi@ubique.co.il>
To: Multiple recipients of list <www-talk@www0.cern.ch>
Subject: Re: Proposal: WIT over USENET or Mail
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas

> From: Tim Berners-Lee <timbl@www3.cern.ch>

> ... I feel that the point of the NNTP
> protocol is that it is a global flooding protocol.  It has certain
> properties.  If an NNTP server says to you "IHAVE xxxx" you can check
> to see whether you have xxxx from any other site, you can pass it on,
> you can refer to it just as xxxx.  Either people have that message
> or they don't.  Looking at the Message-ID is all you need.
> 
> By contast, in a system with private news servers, you would have to
> use the full URI of style nntp://host/message-id as a reference
> instead.  NNTP doesn't use those URLs.  Not in the "References:", not
> in the protocol.
> 

NNTP doesn't use those URLs partly since it dates back to "pre-URL"
days.  Still, we have an option to use the client-server protocol of
NNTP for remote access (nothing precludes it in the NNTP definition;
most clients support arbitrary configuration of the host address; and I
am not aware of inherent inefficiencies in remote NNTP access.)

I feel that the key question is whether we want to offer a smooth
migration path for newsgroups, which are very popular and widely used,
to the current architecture and capabilities of the Internet (recall
that NNTP was invented in the UUCP days), or whether we want to abandon
them in favour of our "brave new world".  I feel that if a migration
path can be offered, then it should be offered.  In this particular
case there are both architectural reasons and social reasons for that,
not the least of which is allowing people without full Internet access
participation in newsgroups, if needed, and allowing existing
newsgroups easy migration to remote access instead of "flooding".

The proposal I have put forward provides direct access to interactive
discussions for the "have's", and provides indirect, e-mail based
access for the "have not's".

I must admit I wasn't aware that URLs of the form "news://host/article"
were "forbidden"; to me they seem natural, self explanatory,
architecturally sound, and useful for newsgroups whose ownsers don't
wish to flood the entire Internet, yet wish to allow access to anyone
interested in participating.  So "forbidding" them in the name of some
higher principles ("though shall abandon NNTP in favour of HTTP?" :-)
seems to me to be a mistake.

-- Udi

p.s. On further enquiry Tim was not able to recall a source for
bidirectional newsgroup-mailing list reflector, so a pointer will be
appreciated.