[Top] [All Lists]

Re: [ietf-smtp] Anti-spam - paid network

2015-12-03 15:38:43

--On Thursday, December 03, 2015 13:20 -0800 Brandon Long
<blong(_at_)google(_dot_)com> wrote:

/me looks at the number of messages we send a day, multiples
by 10 cents, laughs
/me looks at the amount of money an ad supported service earns
per user, looks at per message costs of 10 cents, laughs

Right.   I agree that the economics are unlikely to work in
practice, even if one titrated the amounts.  I tried to say that
in a more general way.

Regardless of any other complication, charging enough to
matter to bad senders but not enough to deter real senders...
this whole thing is a non-starter.

See above.

As for equating DMARC (which is an anti-spoofing /
anti-phishing proposal, not an anti-spam proposal) with this,
I don't even...

I didn't intend to "equate"  them and certainly understand the
distinction.  DMARC was mentioned only because it provides a
fairly recent example of a method based on assumptions of proof
of authorization to send mail from a particular address or
domain whose deployment caused disruptions.

Personally, I don't believe this is worth pursuing even to the
extent of more traffic to explain why it is a bad idea.  There
are a lot of reasons why a general fee-for-mail (especially a
fee that, as you point out, is small enough to not provide any
deterrent) is not going to work and a lot of independent reasons
(a few of them actually hinted at in Jacob's note) why it would
be likely to be impossible to deploy in practice.  

I was objecting almost exclusively to the style in which the
proposal was dismissed, independent of whether it is a good idea
or even one that should be taken seriously.


ietf-smtp mailing list