E-postage would not only help fund the system, but it could stop spammers

Rohit Khare (khare@www10.w3.org)
Tue, 21 Jan 1997 12:52:34 -0500

[Dr. Metcalfe: seriously, I'm not trying to stalk your column or anything,
it's just that this week's was an excellent hook to present my theory about
community currency to my friends. Just thought I'd cc: you directly]

Bob Metcalfe's latest "From the Ether" column hits a few good points about
the virtue of charging for e-mail, but I think his scenario is missing a
really smart way to migrate to it.

First, though, a letter to the editor at InfoWorld about Bob's closing

> Is anybody at the Internet Engineering Task Force (IETF) working
> on e-postage? No? Hello, earth to IETF, come in, please.

The IETF is an excellent consensus-drivent standardization body. Not a
think tank, industrial lab, or startup. The IETF should not be in the
business of inventing technology, it's there to look at what works and
tweak it for interoperability. Once in a great while, within
well-understood and hallowed areas, it will take on a
design-competition-like role in soliciting and evaluating technology (IPv6
security, for example). Otherwise, please call on the venture capitalists,
advertisers, and academic researchers of the world to investigate this
problem -- they're the ones with the answers!

Rohit Khare

Now, on to my point about the one-sided-ness of Bob's proposal [rehash of
Kudos ahead]...

The answer is: people should pay for my time to read their (email | catalog
| advert-site-ment). The question is: with what? How about the good will
people generate for *creating* good content? Why not charge people for
their attention in the same unit -- attention span itself?

I should be able to "make" money whenever I get someone's attention for my
insightful and witty commentary -- everytime I merit an 'attaboy' from the
Net. And I can spend this unit to buy other people's attention. Time *is*

So how about it? It's a fabulous new economic experiment, a currency backed
by a very tightly controlled infinite reserve: time itself?

Well, not measured in seconds. 10 seconds of my poorly-compensated time is
just not worth as much as 10 seconds of Aaron Spelling's, say. Furthermore,
10 seconds of my high-bandwidth attention is worth more than a slower
reader's. A Kudo is pegged at a floating exchange rate: it's how much
attention *everyone else* thinks it's worth for a certain context. 3 kudos
for a personal message to me may merit more attention than 30 for a piece
of 'junk mail'. Conversely, if I like an article on stir-fry cookery, I
might send a kudo, but I'd send 100 for a good MacareNeXT parody.

Kudos are kool because they monetize gradually. They are not backed as a
reserve currency in any traditional sense, requiring someone to back it
with dollars or frequent flier miles. They can come into existence just
because they're 'cool' -- after all, everyone has time (universality), but
only a little (store of value, noninflationary). Over time, a site that
accumulates lots of kudos (say, The Squat), can put up ads and start
converting kudos to dollars. Or an advertising agency can put up an
infomercial site and change dollars to kudos by handing out kudos to
qualified leads.

Kudos are a scenario that motivates a new style of cryptographic
micropayment protocol. Rather than "mints" per-publisher or per-broker with
aggregation in the market to 'exchange' these proprietary currencies, we
need micropayment schemes designed without a central bank but still proofed
against conspiracy -- much like the Surety timestamping model manages to

And the social impact could be enormous. So I think there's good work to be
done in this arena.

Rohit Khare

Rohit Khare -- World Wide Web Consortium -- Technical Staff
w: 617/253-5884  --   f: 617/258-5999   --  h: 617/491-5030
NE43-344,  MIT LCS,  545 Tech Square,  Cambridge,  MA 02139


January 20, 1997

E-postage would not only help fund the system, but it could stop spammers

Thanks to the many of you who invested the cost of an e-mail to write Editor in Chief Sandy Reed in support of my year-end bonus. And thanks even if you demanded that this column be discontinued because I failed to kill myself after not being entirely right about the Internet's collapses in 1996.

Now notice that all those e-mails you sent, and Sandy's confirmations, were carried by various Internet service providers for no extra charge on top of a flat monthly fee; they were free. Starting work on my 1997 bonus, this week I would like to sketch some arguments for why you should really want to pay postage on each of your e-mails.

Proponents of flat-rate pricing and free e-mail argue that the Internet would not have taken off if people had been asked to pay for what they used. Maybe, but the Internet has now taken off. It must soon lift its landing gear and do whatever else planes do to sustain flight.

Proponents of free e-mail note that Prodigy tried charging for e-mail, and all that got them was sued. Of course, not charging for e-mail didn't help Prodigy much either.

Proponents of free e-mail say that Internet usage will be retarded if people are asked to pay for what they use. OK, this is likely true, and, therefore, "e-postage" may even be beneficial.

Usage of our postal services, long-distance telephones, automobiles, food, and water, although perhaps retarded, have all been managing to grow substantially despite their consumers having mostly to pay as they go. And many of the exceptions, such as free water in California, are problems.

You really deserve improved e-mail service. This can be had over time if carriers cover their costs. Paying e-postage for what you use of e-mail is the best way to cover costs. Competition among e-mail carriers will over time drive prices down, getting us to universal service at just about the perfect time.

We should probably pay for each message we send based on how big it is, on how many copies we're sending, and on how far it's being sent. We should probably pay extra for special services such as urgency, acknowledgment, storage, privacy, and authentication. We should be able to send e-mails collect and bill them to willing third parties.

Some say it would cost much more to charge for e-mail than it costs to provide it. Even if this were true, so what, and as we grow out of ASCII e-mail, this won't be true for long.

Too bad the Internet's designers had no apparent interest in economics and left money out of e-mail standards. Internet servers exchange mail around the world without asking the obvious questions such as, "How much should who pay for this message?"

Unfortunately, even if this question were being asked today, there are no micromoney systems in place to handle the required small exchanges of value: no coins for stamps. So paying e-postage will have to wait for micromoney and the upgrading of e-mail to work with it.

Too bad we're going to have to wait a long time to pay Internet e-postage. Free e-mail will tend to be worth every penny paid for it. And too bad, because e-postage is the ultimate solution to unwanted e-mail, most of it poorly targeted advertising or "spamming."

Think of all the mail you'd receive through the United States Postal Service were it not for the cost of paper, printing, and postage. The root cause of spam on the Internet is that there is no paper, printing, or postage. With e-postage, anybody sending you e-mail has to pay for it.

Now a mail server that is seeking to forward e-mail to your mailbox server could be asked to describe each message intended for you.

Your e-mail "acceptance" agent could compute incremental e-postage based on that description. This way, you could set the price that advertisers could choose to pay for your time to read (or subsequently discard) their messages.

With e-postage, there would be pressure on spam and just enough over time so as not to require any major overhauls of the United States Constitution and its First Amendment.

Is anybody at the Internet Engineering Task Force (IETF) working on e-postage? No? Hello, earth to IETF, come in, please.