[Asrg] Re: Consent protocols - was E-postage

On Fri, 30 Apr 2004, John Levine wrote:
> >> f. Sender pretends to be 300 other people and sends you buckets of spam.
> Sorry, that was supposed to be:
> f. Sender pretends to be 30,000 other people and sends you buckets of spam.

Sure. You seem to be saying "Does this scale?". I suggest that it scales
better than having to do content inspection, or manual filtering of
spam from X hosts.

> >Now, if you like, add greylisting for unknown senders, so that those
> >300 msgs won't be accepted on the first try. Then, add a distributed
> >blacklist which you check your mailqueue against before delivery to
> >users' inboxes. Application of the blacklist can be part of whatever
> >other content spam filters the user has.
> 
> DCC already does that.  It works reasonably well, with no need to
> invent new consent protocols.

DCC today, is used as a first line of defence. Your MTA needs to accept
all X messages, and check a DCC for them.

With the scheme I described, you make DCC the second line of defense,
and save a lot of work in two out of three cases.

Case one: For Sender IPs that you develop Trust for, you may choose not
to bother checking the DCC. In my case, this would account for roughly
half my mail (the non-spam half).

Case two: For Sender IPs that you develop Distrust for, you may choose
not to accept connections at all. 

Case three: For new Sender IPs, you do all the filtering you do now,
_and_ you offer a chance for them to self-identify as spammers, by
exceeding your specified rate limit.

I realize that the argument about whether to give negative feedback to
spammers is not a closed issue by any means, but I like that in Case
two, you reduce bandwidth demands across the backbone by not letting the
spam flow to you in the first place.

-- 
David Maxwell, david@vex.net|david@maxwell.net --> Although some of you out
there might find a microwave oven controlled by a Unix system an attractive
idea, controlling a microwave oven is easily accomplished with the smallest
of microcontrollers. - Russ Hersch - (Microcontroller primer and FAQ)

_______________________________________________
Asrg mailing list
Asrg@ietf.org
https://www1.ietf.org/mailman/listinfo/asrg