Re: [tlug] Email address munging in the TLUG archives

[Godwin Stewart <godwin.stewart@example.com>]

On Wed, 6 Dec 2006 08:59:43 +0900, "Josh Glover" <jmglov@example.com>
wrote:

> I am not so sure that this is true any more. I think your "average"
> spammer is now a professional spammer. Maybe Godwin can shed some
> light on this, as I think he is much more kuwashii than most on this
> issue.

Shannon is right on the whole. As long as there *are* low-hanging
fruits to pick, that's what spammers will do.

They also resort to so-called dictionary attacks. They can get lists of
active gTLD's from Verisign and the other registries, and they have
software that sits there blasting out mail to random users at those
domains. Mail that isn't rejected results in the randomly generated
e-mail address it was sent to being added to a list, which is
subsequently sold off to other spammers.

Why should they go to the trouble of deobfuscating e-mail addresses
when it's so easy to have software generate them randomly and then sell
them?

> I think spammers make decent money without having to work particularly
> hard. Everything is automated, so all they have to do is buy their
> lists of email addresses, feed them to the software, and go. I think
> the part of spamming that requires the most work is maintaining your
> business relationships with the companies whose products you are
> advertising.

What business relationship?

Do you think a spammer whose illegal pharmacy site is hosted in China,
as is his credit card processor, actually *has* any pills to sell? At
best you'll get counterfeit pills made of compacted sawdust and rat
poison, at worst the spammer will pocket the money and not bother
sending you anything. You actually receiving any genuine product
ordered from a spammer is *extremely* unlikely. You can't pester him
because any phone number on the site is either bogus or belongs to an
innocent bystander, and the domain is registered in the name of someone
who fell for a phishing scam (or ordered stuff from the spammer) and
had his identity stolen - which is likely to happen to you too if you
order from a spammer.

The only relationship that a spammer will have with the company whose
products he is "selling" is going to happen in a courtroom - if the
authorities can be bothered to go after the perp.

> > SMTP -> SMTP is the warlike spam-filled status quo
> > NEMS -> SMTP is treated as status quo (because SMTP can't do
> > anything else)
> > NEMS -> NEMS is spam free
> > SMTP -> NEMS is the decision locus.
>
> I really don't understand how this could be implemented. Can you
> provide more details?

You'll probably find them here: http://www.google.com/search?q=FUSSP

No offence to Shannon, but solutions like this have been discussed and
demonstrated unworkable sooooo many times, merely because they rely
either on immediate global adoption and/or on radical changes to current
SMTP software. Feeding into SMTP isn't the problem here, it's the SMTP
to NEMS part which isn't going to work unless new SMTP software is
written that "speaks" NEMS.

--
G. Stewart - godwin.stewart@example.com

The new Linux anthem will be "He's an idiot, but he's ok", as performed
by Monthy Python.  You'd better start practicing.
        -- Linus Torvalds, announcing another kernel patch

Attachment: pgpDkYx9kVdBm.pgp
Description: PGP signature