Re: ietf-nntp New wording on article numbers

Brian Kantor <brian@nothing.ucsd.edu> Sat, 28 December 1996 18:21 UTC

Received: from cnri by ietf.org id aa28639; 28 Dec 96 13:21 EST
Received: from ACADEM2.ACADEM.COM by CNRI.Reston.VA.US id aa11538; 28 Dec 96 13:21 EST
Received: (from majordomo@localhost) by academ2.academ.com (8.8.4/8.7.3) id MAA03138 for ietf-nntp-outgoing; Sat, 28 Dec 1996 12:19:24 -0600 (CST)
X-Authentication-Warning: academ2.academ.com: majordomo set sender to owner-ietf-nntp using -f
Received: from academ.com (root@ACADEM.COM [198.137.249.2]) by academ2.academ.com (8.8.4/8.7.3) with ESMTP id MAA03133 for <ietf-nntp@ACADEM2.ACADEM.COM>; Sat, 28 Dec 1996 12:19:22 -0600 (CST)
Received: from nothing.ucsd.edu (nothing.ucsd.edu [132.239.1.4]) by academ.com (8.8.4/8.7.1) with ESMTP id MAA29264 for <ietf-nntp@academ.com>; Sat, 28 Dec 1996 12:19:19 -0600 (CST)
Received: by nothing.ucsd.edu (8.8.3/UCSDGENERIC.5) id SAA20971 to ietf-nntp@academ.com; Sat, 28 Dec 1996 18:18:10 GMT
Date: Sat, 28 Dec 1996 18:18:10 GMT
From: Brian Kantor <brian@nothing.ucsd.edu>
Message-Id: <199612281818.SAA20971@nothing.ucsd.edu>
To: ietf-nntp@academ.com
Subject: Re: ietf-nntp New wording on article numbers
Sender: owner-ietf-nntp@academ.com
Precedence: bulk

>The server should never reuse an article number. The GROUP command, however,
>is providing an indication of the range of articles which is currently 
>available, and that may quite legitimately be smaller due to cancellation.

Whilst the list of currently-AVAILABLE article numbers may be reduced
because the highest-numbered article has been cancelled or expired,
that does NOT mean that that article number is available for filing a
newly-arriving article.

What the above DOES imply is that a news server which provides reading
services MUST maintain an article counter for each newsgroup.  That
counter may never be decremented for any reason, and may only be set back
at an epochal reset, an event which should be (we hope) very rare.

The group command may return either the current value of the counter OR
the last available article number as an implementation issue.

In the current and forseeable future, there is a strong likelyhood
that a new article will have arrived - thus extending the article
number range - since the GROUP command was executed, so all reader
software must be prepared to cope with that eventuality.

Since the article numbers are not transmitted in a binary form, I see
no reason to artificially restrict the range of article numbers by
specifying the size or length of the representation.  A specification
that the numbers are DECIMAL and represent actual article counts should
be sufficient.  (I.e., we should not force epochal resets of a newsgroup
article count just because it won't fit in N digits.)
	- Brian