Re: ietf-nntp New wording on article numbers

USENET news manager <newsmaster@ucs.cam.ac.uk> Sat, 28 December 1996 13:44 UTC

Received: from cnri by ietf.org id aa16963; 28 Dec 96 8:44 EST
Received: from ACADEM2.ACADEM.COM by CNRI.Reston.VA.US id aa07878; 28 Dec 96 8:44 EST
Received: (from majordomo@localhost) by academ2.academ.com (8.8.4/8.7.3) id HAA02265 for ietf-nntp-outgoing; Sat, 28 Dec 1996 07:40:21 -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 HAA02260 for <ietf-nntp@ACADEM2.ACADEM.COM>; Sat, 28 Dec 1996 07:40:19 -0600 (CST)
Received: from lyra.csx.cam.ac.uk (news@lyra.csx.cam.ac.uk [131.111.8.36]) by academ.com (8.8.4/8.7.1) with SMTP id HAA28172 for <ietf-nntp@academ.com>; Sat, 28 Dec 1996 07:40:13 -0600 (CST)
Received: by lyra.csx.cam.ac.uk (SMI-8.6/MDTG-V1.1.8@lyra.csx.cam.ac.uk) id NAA28314; Sat, 28 Dec 1996 13:40:04 GMT
Message-Id: <199612281340.NAA28314@lyra.csx.cam.ac.uk>
Subject: Re: ietf-nntp New wording on article numbers
To: Robert Elz <kre@munnari.oz.au>
Date: Sat, 28 Dec 1996 13:40:04 +0000 (GMT)
Cc: ietf-nntp@academ.com
In-Reply-To: <7700.851759720@munnari.OZ.AU> from "Robert Elz" at Dec 28, 96 06:55:20 pm
From: USENET news manager <newsmaster@ucs.cam.ac.uk>
X-Mailer: ELM [version 2.4 PL22]
Content-Type: text
Sender: owner-ietf-nntp@academ.com
Precedence: bulk

Robert Elz wrote:
>
>    Date:        Sat, 28 Dec 1996 02:49:13 +0000 (GMT)
>    From:        Jon Ribbens <jon@oaktree.co.uk>
>    Message-ID:  <199612280249.CAA15425@black.oaktree.co.uk>
>
>    This is wrong. The last article number may quite possibly be less than
>    that from a previous response for that group. The last article number
>    is the highest number corresponding to a currently available
>    article - it is *not* the high-water-mark. If there are articles
>    2, 3 and 4, then the highest article number is 4. Subsequently
>    article 4 may be cancelled, and then the GROUP command must return
>    3 as the highest available article.
>
>This can't possibly be right, article number 4 can't be used again,
>or readers that have read it before it was cancelled would never
>see it's replacement.   The GROUP command isn't intend to give any
>kind of authoritative statement about which articles actually exist,
>just what the relevant range of numbers is to look in - any of the
>articles in the range (including the lower and upper bounds) might
>be missing when requested specifically.
>
>I totally agree with the original wording, numbers reported must
>never go backwards.

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.
If the client wants to know the highest article number that has ever been 
assigned within a group, it should use the LIST [ACTIVE] command.

I've just confirmed that the INN 1.4unoff4 and 1.5 GROUP command shows the
highest available article number, not the highest that has existed prior to
cancellation. It is therefore the way (some widely used) current server
software behaves, and readers don't seem to have any problems with it, and the
wording should therefore not exclude that implementation.

                                John Line
-- 
Cambridge University Computing Service - USENET news manager. Usually John Line
newsmaster@ucs.cam.ac.uk    (alias {newsmaster,news,usenet}@news.cam.ac.uk)