more on Stopping spam isn't as easy as you might hope

[Dave Farber <dave () farber net>]

> Date: Sat, 31 May 2003 15:20:54 -0400
> From: Bob Frankston <Bob18-2 () bobf frankston com>
>
>
>
> Since this is a recurring issue I should update
> http://www.frankston.com/?name=spamfixation. For now I'll simply say that
> the concept of spam itself is the result of having a word that can be used
> in lieu of understanding.
>
> There are really two issues. One is the concern about the volume of mail and
> I'm not convinced that it is that bad compared with what the capacity we'll
> need when video traffic becomes the norm. But it's really a denial of
> service issue not an email attention issue.
>
> The spam problem is really an attention problem (separating out the issue of
> slow pipes and fraud which is ever present) and is the to-be-expected result
> of having one well-known magic name that all can use to reach us be it by
> phone/fax, direct mail, email or other ways. There are also indirect ways
> such as advertisements (which are invisible only because we've learned to
> manage our attention). Many approaches to spam seem to assume that there is
> an omniscient and prescient version of Maxwell's Demon that can make a
> static distinction between those with good intent and bad intent or which
> simply knows what we're interested in at the moment. This is also the
> presumption that makes people ask for firewalls and leads to increasingly
> complex and problematic solutions when failures make people try harder
> rather than recognizing the fundamentally flaws in the concept. Aside from
> being ineffective, the appeal to a central authority is basically
> antithetical to the underlying end to end concept. Since SMTP is an edge
> protocol, there isn't a place for the demon to sit in the middle of the
> network anyway.
>
> What we really need are edge tools that help us manage the demands on our
> attention and capabilities rather than names to manage our personas. Of
> course these are all going to be imperfect since there cannot be a static
> metric.
>
> As an aside, the payment system ideas are tempting and can be implemented at
> the edges. While I know reporters who would like to put their fax on a 900
> number I don't know of any who have actually taken that step. There actually
> is a version that is implemented -- advertisers pay third parties to
> piggy-back on their reputation and reach their viewers or readers. There are
> even publications that are 100% advertisements that you choose to view.
>
> I'm especially concerned with proposals that make the .COM problem far worse
> by presuming that the DNS is the source of all authority (and wisdom) rather
> than just a binding mechanism. Just because I have an MX record somewhere
> should not subject me to commercial conditions of all the zone owners in DNS
> path. As John Levine pointed out, some proposals are like making it illegal
> to drop a (paper) letter in a mail slot other than the one near my house.
> This is the kind of rigid centralism that threatens the Internet.
>
> The anger at spam is real but so is the danger of misguided solutions that
> only create worse problems while only addressing the symptoms.


-------------------------------------
You are subscribed as interesting-people () lists elistx com
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/