NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/mail/headers/426

Internet Message Format | 1993-01-10 | 6.1 KB

Path: sparky!uunet!charon.amdahl.com!pacbell.com!ames!agate!stanford.edu!enterpoop.mit.edu!eru.mt.luth.se!lunic!sunic!aun.uninett.no!nuug!ifi.uio.no!enag
From: [e-mail] (Erik Naggum)
Newsgroups: comp.mail.headers
Subject: Anastasios Kotsikonas' list server
Message-ID: <[e-mail]>
Date: 10 Jan 93 19:36:43 GMT
Reply-To: Erik Naggum <[e-mail]>
Lines: 122

Folks,

There's a new list server out there, written by some Anastasios Kotsikonas,
that violates the spirit and the text of the Internet mail RFCs and ignores
the oral tradition that RFC 1123 tried to codify.  This is not surprising
for list servers in the Eric Thomas tradition, but this list server also
violates what we've learned to cope with from those list servers.  It looks
as if this Anastasios Kotsikonas descended upon earth like some Plan 9 from
Outer Space re-make, and managed to screw up even more royally than usual
for amateur mailer writers.  To illustrate, here's the headers of a message
I sent:

    From: Erik Naggum <[e-mail]>
    Reply-To: Erik Naggum <[e-mail]>
    Message-ID: <[e-mail]>
    Date: 09 Jan 1993 01:12:25 +0100
    To: Entropy Warrior <[e-mail]>
    Cc: Multiple recipients of list <[e-mail]>
    In-reply-to: <93Jan8.175917est.160633(2)@router.mail.cornell.edu>
    References: <93Jan8.175917est.160633(2)@router.mail.cornell.edu>
    Subject: Re: Just asking

and here's what I got back (minus Received headers):

    Return-Path: <@router.mail.cornell.edu:[e-mail]>
    Message-Id: <[e-mail]>
    Comment:  Copy Editors and Editing
    Originator: [e-mail]
    Errors-To: [e-mail]
    Reply-To: Erik Naggum <[e-mail]>
    Sender: [e-mail]
    Version: 5.5 -- Copyright (c) 1991/92, Anastasios Kotsikonas
    From: Erik Naggum <[e-mail]>
    To: Multiple recipients of list <[e-mail]>
    Subject: Re: Just asking
    Date: Fri, 8 Jan 1993 19:22:43 -0500

Now, let's look at the headers individually:

--  The Reply-To field is *clobbered*, and the old value is completely
    lost, redirecting replies to an address I did not want to receive
    messages at.  (I have received four messages destined for the list
    already.)

--  The To and Cc headers are *destroyed*, and a meaningless new value is
    inserted instead.  (My original Cc field comes from this new To field
    in the message I replied to, so the To field has no relation to my Cc
    field.)

--  The In-Reply-To and References headers are both *discarded*, completely
    destroying any threading and tracking information.  The two items above
    are bad enough, but this is horrendous.

--  The Date field is changed, so that those who reply with a mailer that
    doesn't use the Message-ID in the In-Reply-To field will generate a
    message that I can't find in my archives without manual intervention.
    *All* the information that can be used to mechanically retrieve the
    message to which a received reply refers is _gone_.  This bespeaks
    EVIL! in my mind.

--  Return-Path differs from Errors-To, which means that the list (!) will
    receive errors messages from conforming Internet mailers, and Carol
    Roberts will receive error message from non-Internet or non-conforming
    Internet mailers.

--  There is an "Originator" field which is clearly misused insofar as it
    has any meaning at all.  The Sender field is probably correct for this
    purpose, though.

--  There's a "Comment" field, which should've been named "Comments" to
    follow RFC 822.  The "Comment" field contains the information which
    should've been in the "phrase" part of the list name instead of the
    stupid "Multiple recipients of list" thing inherited from Eric Thomas'
    wretchedness.

--  There's a "Version" field with a copyright notice (!) inserted so we
    know who to blame for this atrocity.  Whether this is valuable or not
    is another question.  In at least one previous version, there was a
    whole Copyright field, which implied that Anastasios Kotsikonas had
    copyright in the contents of the messages distributed through his list
    server!

The result is that only 3 (three) headers survived close encounter of the
third degree with this (m)alien Anastasios Kotsikonas:

    From: Erik Naggum <[e-mail]>
    Message-ID: <[e-mail]>
    Subject: Re: Just asking

I mean, I could bow and scrape for the fact that my Message-ID line was
intact (most BITNET list servers gratuitously clobber it if you don't SET
listid FULLHDR), but that's pure accident.  I sent a message to the list
server to receive copies of my own messages (SET COPYEDITING-L MAIL ACK,
instead of the usual SET ... REPRO, when ACK usually means getting some
sort of message back that the message was redistributed), and the response
had the _same_ Message-ID that my request had!  Now, _that_ is bogus!

Of all the braindamaged mailers and mail-type service access points I've
used in my 8 years with access to Internet mail, this one _really_ takes
the cake.

OK, so who uses this stupid list server?  The first place I saw it used was
at ORA.COM, O'Reilly and Associates, publishers of the excellent Nutshell
Handbooks and the X Window System series, who host mailing lists for the
Davenport group.  I resigned in disgust from these lists when I got twenty
"vacation" messages because the Return-Path pointed back to the list and
the To field had the name of each recipient in it (*HATE*), _and_ this
Anastasios Kotsikonas thing claimed copyright on all the contributions.
Now, I found it in use at Cornell.

The burning questions are: How do we get Anastasios Kotsikonas to read
RFCs?  And, if this fails (which I suspect it will, as with all other
amateur mailer implementers), how do we get his atrocious list server
booted off the Internet?

Best regards,
</Erik>
--
Erik Naggum                 ISO  8879 SGML                    [phone]
Oslo, Norway                ISO 10744 HyTime
<[e-mail]>            ISO  9899 C                 Memento, terrigena
<[e-mail]>           ISO 10646 UCS             Memento, vita brevis