ietf
[Top] [All Lists]

Re: Why spam is a problem.

2002-08-13 19:03:08
From: Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu>

The main path will be that bulk mail will be taxed at about $0.10 per
target with stiff criminal penalties for evading the tax. ...

in other words, such laws would just make spam a mainstream activity.  

There seem to be some words mixed up there.  Spam is already a mainstream
activity, as demonstrated by megabytes of drivel from the Fortune 500.
Taxing spam will only squeeze the worst spammers as it emboldens the
DMA's customers among the Fortune 20,000.

       .........
 
] from "Perry E. Metzger" <perry(_at_)PIERMONT(_dot_)COM>

] ...
] I think the main good certification can do is to stop some of the
] idiocy mail forging viruses have started.

} From: Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu>

}  ...
} indeed, if we could just make it easy to trace the source of viruses
} (so that we can detect them earlier and make it more likely to catch
} them close to the source), that would seem like a big win.

Given the imperative for Redmond-style user friendliness, how will
you keep viruses from replacing any signatures of the previous hop
with new signatures using private keys found in the current machine?

Note that you certainly don't need any fancy authentication machinations
to know the previous hop of a virus.  You can simply look at the SMTP
client IP address, often in the Received header added by the ISP of
the newly infected machine.  Of course, that requires that the SMTP
headers of the virus be available, but that's no stronger than assuming
that a virus will save its authenticating tokens.
In other words, the virus problem is distinct.  If it has a simple
magic bullet, it involves competent operating system design instead
of more layers of wishful thinking and propaganda.


  ..........

| From: "Perry E. Metzger" <perry(_at_)PIERMONT(_dot_)COM>

|                                            What concerns me is the
| fraction that I can't, like the "Giant Discount!" spam that just made
| it past my filters.

Would it have made it past the DCC?  You could check by feeding the
spam including headers to http://www.rhyolite.com/cgi-bin/dccproc-demo


Vernon Schryver    vjs(_at_)rhyolite(_dot_)com



<Prev in Thread] Current Thread [Next in Thread>