Sometimes spam(-trapping) can be funny....

Thu May 9 05:33:24 PDT 2002

Normally, I wouldn't bother y'all with Yet Another Spam... but the
way this worked out managed to strike me as funny.

The main BayLISA mailing lists are technically not "moderated";
rather, they are set up so that certain sets of folks are able to
post directly, subject to certain content-filtering (which, if
triggered, causes the message to be intercepted and directed to the
list owner for disposition).  The "set of folks" typically includes
subscribers to the list itself, but there are additions in some
cases -- for example, any subscriber to the baylisa list is permitted
to post to the blw list (because that made sense to me, so I did it).

And the list that catalyzed much of this was, of course, baylisa-jobs --
and for this list, the content filtering is fairly extensive (and is
itself somewhat amusing, if that's the sort of thing that you are
inclined to find amusing).  In particular, I got rather fed up with
headhunters trying to spam baylisa-jobs with irrelevant posts, so I put
some traps in there to catch the more common ones (and I add to it when
I've received suitable inspiration...).

Now, way back in the 1980 - 1992 period I used to wrangle IBM mainframes
for a living; the first couple of years as an applications programmer,
and the rest as a systems programmer.  Now, in the IBM mainframe world,
the job of "systems programmer" is pretty much what is known as "systems
administrator" in the UNIX world (though networking back then was, shall
we say, ... different).  So I'm somewhat sympathetic to the notion that
someone with only IBM mainframe "systems programming" skills ought to be
welcome at BayLISA -- certainly at least as much as any other non-UNIX
skillset.

And care & feeding of CICS -- a package that allowed remote terminal
users to access applications -- certainly is a technical skill; I would
not dream of disputing that:  I've done my share of that sort of thing.

But it is sufficiently removed from the norm of what BayLISA is about
that a post to baylisa-jobs that mentions it is somewhat suspect, and
likely deserving of human inspection, so I included a "trap" for the
regex /CICS/i in the baylisa-jobs list definition.

Thus, I found the following trapped spam rather amusing, given the
difference between the apparent intent of the spammer vs. the
reason(s) the spam was trapped.  Note especially the last clause in the
subject of the "bounce" message vs. the subject of the spam itself:

>From owner-baylisa-jobs at baylisa.org Thu May  9 04:47:15 2002
>Date: Thu, 9 May 2002 04:47:09 -0700 (PDT)
>From: owner-baylisa-jobs at baylisa.org
>To: baylisa-jobs-approval at baylisa.org
>Subject: BOUNCE baylisa-jobs at baylisa.org:    Non-member submission from [bramesh35 <bramesh35 at hotmail.com>]   taboo body match "/<html>/i" at line 5  Message too long (>40000 chars)  taboo body match "/cics/i" at line 2843  
>
>>From david Thu May  9 04:45:29 2002
>Received: from out008.verizon.net (out008pub.verizon.net [206.46.170.108])
>	by www.baylisa.org (8.11.2/8.11.2) with ESMTP id g49BjS319515
>	for <baylisa-jobs at baylisa.org>; Thu, 9 May 2002 04:45:28 -0700 (PDT)
>Received: from Jgkxcn ([63.111.170.138]) by out008.verizon.net
>          (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with SMTP
>          id <20020509114525.WHHO25518.out008.verizon.net at Jgkxcn>
>          for <baylisa-jobs at baylisa.org>; Thu, 9 May 2002 06:45:25 -0500
>From: bramesh35 <bramesh35 at hotmail.com>
>To: baylisa-jobs at baylisa.org
>Subject: Re:japanese girl VS playboy
>MIME-Version: 1.0
>Content-Type: multipart/alternative;
>	boundary=KO6F4Gpvkx9MO724Yf768P5
>Message-Id: <20020509114525.WHHO25518.out008.verizon.net at Jgkxcn>
>Date: Thu, 9 May 2002 06:45:27 -0500
>
>--KO6F4Gpvkx9MO724Yf768P5
>Content-Type: text/html;
>Content-Transfer-Encoding: quoted-printable
>
><HTML><HEAD></HEAD><BODY>
>....

Of course, maybe I'm just desperate for humor....  :-}

Cheers,
david
-- 
David H. Wolfskill				david at catwhisker.org
Microsoft products -- for those times when reliability just doesn't matter.