On November 28, 2008 at 10:51 mouse(_at_)Rodents-Montreal(_dot_)ORG (der Mouse)
wrote:
It seems to me that fairer "pay fairly" would have the recipients
assuming those costs to the extent that they value the mail. When I go
visit a friend, I don't ask the friend to pay for half the transport
cost (gas, transit ride, whatever); the visit is valuable enough to me
for me to assume those costs myself. (With some friends, visits go
both ways and mostly balance out; with others, they don't, but, in the
cases where I do the visiting more, the visits are valuable enough for
me to be willing to assume an "unfair" share of the transport cost.)
Fair is therefore a very dangerous word in this context.
The conception is if you are a bulk emailer you should be paying for
your email sending, just as you would if it were paper mail tho no
doubt at a much lower per piece rate recognizing the costs.
If you're not a bulk emailer then rates can almost certainly be such
that your costs would be effectively if not actually zero as my
example in a previous message indicated but briefly:
If we assume a spammer sends O(10^9) pieces per day and scale the rate
to be about $1,000/day for that privilege (just a hypothetical) then
your sending even 10^4 msgs/day would amount to about a penny a day.
And even that assumes an overly simplistic rate structure. Just toss
in 1,000 msgs/day free for the average user, or even 10,000, and your
spammer can't live with that, they can't send 10K/msgs/day and stay in
business, so they're well into the paying area.
At 1c/10K beyond that lists like this one might cost $50/year and even
that could be waived or a much larger cap imposed (like 100K/day) for
valid not-for-profits etc, perhaps some modest yearly flat fee.
etc etc.
I doubt very much that someone like Amazon either sends 1B msgs/day or
would balk at costs like $100K/year for what they do send if that was
the way it was. There is value in such a scheme to them if it were
imposed fairly, mainly that their mail would no longer compete with
all that spam in every mailbox.
BECAUSE, the underlying assumption, spammers wouldn't be able to pay,
their business model relies on massive amounts of free resources.
Ok, from there the mind naturally turns to enforcement.
But since in my mind we haven't gotten past this first hurdle of how a
rational pricing model might even work I see no need to plow headlong
into that subsequent topic immediately.
Same goes with email, only the fundamental costs are _much_ lower, and
are almost entirely human time. (Yes, there are other costs, but they
are _tiny_ compared to the human-time costs.)
That's basically what e-postage is based upon: an attempt to impose
fair payment based on a notion of fairness somewhat like that one.
(Not that this says anything about the myriad of other reasons
e-postage won't work in anything like today's world, such as the
impossibility of a suitable micropayment scheme. Just that I don't
think "fair" is quite as simple here as you make it out to be.)
There's no need for a micropayment scheme if it's the bulk senders who
pay.
So at least what I have proposed is independent of any micropayment
scheme whatsoever.
How would we keep from killing all the mail we want? The only method
anyone has ever proposed is for the recipients to skip charging
settlements to senders they like. But once you know who you wouldn't
charge, you don't need the charges, you just whitelist the ones you
wouldn't have charged and filter the rest.
That too isn't quite as simple as you make it out to be.
First, that assumes that the "skip charging" decision is based solely
on the sender's identity. This is a reasonable first approximation,
but not more than that; it's totally plausible to me that I might want
some mail from a sender but not other - as with a company who sends me
both stuff I want (order confirmations, say) and stuff I don't (ads).
Today, we object to such things being mixed in the same email stream,
but that's because we need (FSVO "need") to make mechanized filtering
decisions based on simple prima-facie properties of the email.)
Second, that assumes that the decision to charge or not is based on
properties simple enough to be tested mechanically. If we had this
mythical payment scheme, I know of some people I would wave off the
charges for most mail from, but would definitely charge for things like
chain email or yet another copy of the latest bigoted "joke" meme.
Again, the idea of each recipient making the decision is, to my mind,
entirely unnecessary and undesireable.
The ONLY decision that has to be made is by each MTA:
Does this email have valid "postage"?
YES: pass it on.
NO: drop it.
Again, can immediately jump into what is meant by valid postage but to
my mind we haven't even settled on something as simple as the above
example. Unless we do we can't really proceed to matters dependent on
some meeting of the minds on this much.
Third, it assumes that we have sender identities reliable enough that
we can mechanically detect whether the mail is from someone you don't
want to charge (or filter). (I suppose this can be handwaved under "if
you're discussing a mythical system, it can have as many counterfactual
properties as you want".)
Another reason why, like the post office or many other such systems,
we only care if the message has valid postage, not who it is from.
PPS: To anyone who is about to say "then don't do it that way, do it
some other way", we all want a pony, too.
Speak for yourself - I live downtown in a big city and have nowhere at
all to keep a pony. I'd much rather have about another thousand square
feet of machine-room space. :-)
As a resident of the City of Boston I am told I still have grazing
rights in The Commons (park) downtown but, due to some tragedy back in
the 1600s I am discouraged from transporting livestock to/from there.
--
-Barry Shein
The World | bzs(_at_)TheWorld(_dot_)com |
http://www.TheWorld.com
Purveyors to the Trade | Voice: 800-THE-WRLD | Login: Nationwide
Software Tool & Die | Public Access Internet | SINCE 1989 *oo*
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg