Looking at the U in UBE

[p***@sweavo.34sp.com]

This is apropos of nothing much but maybe it'll find a sympathetic ear on this
list. I signed up years ago and activity has been so low that I have other
projects and little time to spend on thinking about email protocols.

That said, It seems to me that decent sender authentication would open up many
candidate solutions. Like this one:

An email costs the sender $0.50 to be placed in the receiver's inbox (or, more
strictly, for the sender to be notified of its presence). When the recipient
reads it, he/she decides whether it was wanted or not. If so, $0.499 is
refunded. If not, the sender loses his/her 50 cents. MUAs can allow
recipients to configure a list of senders who are always refunded.

If the church wants to send out a mailshot, it can send 1000 emails with a
single dollar. But it'll need say a $5 buffer to allow 50 mails to be sent.
the rest would sit in the outgoing queue until refunds started arriving.

The 0.1 cent per email levy is to contribute toward infrastructure costs and
give ISPs a reason to bother.

This is predicated on
1) Senders' bank details being made known
2) Certification of senders
3) more formal relationships between serious service providers

It gives you
1) Wanted mail is almost free to send
2) Unwanted mail is much less cheap to send
3) Spamming as we currently know it will either need a lot of financial backing
or will involve serious felonies like defrauding a bank.
4) Mailing list owners will have more of an interest in pruning their databases

[The Famous Brett Watson]

Post by p***@sweavo.34sp.com
An email costs the sender $0.50 to be placed in the receiver's inbox (or,
more strictly, for the sender to be notified of its presence). When the
recipient reads it, he/she decides whether it was wanted or not. If so,
$0.499 is refunded. If not, the sender loses his/her 50 cents. MUAs can
allow recipients to configure a list of senders who are always refunded.

This is more or less what Vanquish proposes with their "Personal Value
Control" engine (marketroid blah blah). It's explained, albeit not very well,
at the following location and surrounding pages.

http://vanquish.com/lab/lab_taking_control.shtml

I know about this because I met Phil Raymond, CEO of Vanquish, at last year's
Conference on Email and Anti-Spam. It's like a challenge/response system,
except that the challenge that the sender must meet is to place a bond with
an escrow agent (Vanquish, in this case). Once the bond is in escrow, the
delivery is completed. The recipient is given the option of taking the bond
money, or releasing it. Typically the bond will be returned in full if the
message is acceptable: the escrow agent makes money by charging the recipient
for the service, not by taking a fraction of a cent from each sender.

So long as the problem of falsified sender identities can be solved, there's
nothing drastically wrong with this approach, so long as you don't mind the
fact that 99% of people who might have emailed you spontaneously will no
longer do so, having been put off by the idea of getting their credit card
involved in the process. On the other hand, if you're particularly famous,
people might not be put off by the money side of things, but that's a special
case.

Regards,
TFBW

[Casey Allen Shobe]

Any of these money-based approaches I've heard scare me, and not just a
little.

I have 45,000 emails in this account - you propose that people would have had
to invest a total of $450.00 in me to let me read them, at risk of paying a
lot more?

You're posting to a mailing list, but you seem to forget mailing lists in your
approach? If the mailing list has 2000 subscribers, there's no way anyone
will pay $20 to send a single email, plus the risk of some people saying it's
not wanted. If you say "well it's a single recipient" then who's going to be
in charge of deciding for the mailing list in an unmoderated case?

$0.50 may be a lot to you for an unwanted mail, but it's nothing to the people
who make money off of sending them out.

E-Mail is free and should always be free. Any system that is not will never
succeed.

Cheers,

[James Craig Burley]

Post by Casey Allen Shobe
E-Mail is free and should always be free. Any system that is not will never
succeed.

More to the point, any such system ceases to be "email" in the proper
sense of the word.

After all, any pay-based system for email is going to run into the
rather serious problem that the Internet that underlies any such
technology is, itself, prone to the exact same problems that email
faces. email is simply one of the more tempting targets.

Unless the Internet is to be changed into a pay-as-you-go system, in
which merely sending a *packet* to an external host is expected to
carry with it the "promise" of payment if the receiving host decides
it didn't care to receive it, email won't go in this direction with
any useful degree of success.

OTOH, I think it's possible that email (or, perhaps more likely and
more elegantly, something like my GUIXP idea) will evolve to the point
where software agents negotiating a transaction (such as the exchange
of an email containing, or allegedly containing, a particular kind of
*content*) might have the *option*, based on profiles of the persons
or organizations on whose behalf their acting, of executing a contract
that includes financial remuneration under different scenarios.

We're not too far away from that idea already. A consultant offering
arbitrary clients answers to technical questions, whose web site
allows them to pay via PayPal, might, in another few years, be using
his own software agent to perform substantial aspects of the
negotiations over payments.

In a scenario like that, it's the receiving party that would be
expected to pay the *sender* of an email, since the email contains
information more useful to the receiver than to the sender.

Email *itself*, therefore, needn't, and probably shouldn't, impose a
substantially higher penalty on its use over whatever people pay to
access the Internet in the first place, because, first and foremost,
it's a *communications* medium.

And, as a communications medium, the costs to use it are best spread
out among *all* of those who have access to it, since its
availability, not just it's actual use, represents substantial value
to anyone who is online, as well as substantial expense to maintain as
an infrastructure. It's a "commons", akin to roadways.

[p***@sweavo.34sp.com]

Post by James Craig Burley

Post by Casey Allen Shobe
E-Mail is free and should always be free. Any system that is not will never
succeed.

More to the point, any such system ceases to be "email" in the proper
sense of the word.

...

Post by James Craig Burley
Email *itself*, therefore, needn't, and probably shouldn't, impose a
substantially higher penalty on its use over whatever people pay to
access the Internet in the first place, because, first and foremost,
it's a *communications* medium.

Right, because you never pay to send surface mail.

Or make cell calls

Or long distance

Or to place an advertisement in a shop window

Or buy paper to write on

[James Craig Burley]

Post by p***@sweavo.34sp.com

Post by James Craig Burley

Post by Casey Allen Shobe
E-Mail is free and should always be free. Any system that is not will never
succeed.

More to the point, any such system ceases to be "email" in the proper
sense of the word.

...

Post by James Craig Burley
Email *itself*, therefore, needn't, and probably shouldn't, impose a
substantially higher penalty on its use over whatever people pay to
access the Internet in the first place, because, first and foremost,
it's a *communications* medium.

Right, because you never pay to send surface mail.
Or make cell calls
Or long distance
Or to place an advertisement in a shop window

Up to this point, your examples all fall into the category of paying a
centralized resource a *specific* amount to utilize it, in accord with
a usage pattern and economic arrangement in which it makes sense to
"pay as you use".

Counter-examples that pertain specifically to yours:

You don't pay to make local calls. You're already paying for the
right to have access to the infrastructure; economics and usage
patterns dictate that charging for each local call doesn't make
sense from the infrastructure owner's point of view; and... most
importantly...you don't *pay* people to *accept* your calls!

You don't pay to place an advertisement on public bulletin boards in
all sorts of locations. These are freely available resources that,
nevertheless, cost time and $$ to maintain, and, like email, can be
exploited.

Post by p***@sweavo.34sp.com
Or buy paper to write on

In an email context, this suggests that, since you buy computers to
compose email on, and buy Internet access, you've *already* paid for
email you send!

And that's the point. We're *already* paying for email. It's
therefore rather silly, wasteful, and unworkable to try to foist a
whole new email system based on any built-in idea of pay-per-message.

Consider how people feel when ATMs, which they generally use for
*free* to perform transactions with their money, start charging them
(up front or behind the scenes) per transaction, for whatever reason.

[t***@bluex2.com]

Can you guys remove me from your address book.

I have asked a few times but still getting emails. Please remove
***@bluex2.com I'm really not interested in your cause and registered
in error.

Cheers

Tony

Post by James Craig Burley

Post by Casey Allen Shobe
E-Mail is free and should always be free. Any system that is not will never
succeed.

More to the point, any such system ceases to be "email" in the proper
sense of the word.
After all, any pay-based system for email is going to run into the
rather serious problem that the Internet that underlies any such
technology is, itself, prone to the exact same problems that email
faces. email is simply one of the more tempting targets.
Unless the Internet is to be changed into a pay-as-you-go system, in
which merely sending a *packet* to an external host is expected to
carry with it the "promise" of payment if the receiving host decides
it didn't care to receive it, email won't go in this direction with
any useful degree of success.
OTOH, I think it's possible that email (or, perhaps more likely and
more elegantly, something like my GUIXP idea) will evolve to the point
where software agents negotiating a transaction (such as the exchange
of an email containing, or allegedly containing, a particular kind of
*content*) might have the *option*, based on profiles of the persons
or organizations on whose behalf their acting, of executing a contract
that includes financial remuneration under different scenarios.
We're not too far away from that idea already. A consultant offering
arbitrary clients answers to technical questions, whose web site
allows them to pay via PayPal, might, in another few years, be using
his own software agent to perform substantial aspects of the
negotiations over payments.
In a scenario like that, it's the receiving party that would be
expected to pay the *sender* of an email, since the email contains
information more useful to the receiver than to the sender.
Email *itself*, therefore, needn't, and probably shouldn't, impose a
substantially higher penalty on its use over whatever people pay to
access the Internet in the first place, because, first and foremost,
it's a *communications* medium.
And, as a communications medium, the costs to use it are best spread
out among *all* of those who have access to it, since its
availability, not just it's actual use, represents substantial value
to anyone who is online, as well as substantial expense to maintain as
an infrastructure. It's a "commons", akin to roadways.
--
James Craig Burley
Software Craftsperson
<http://www.jcb-sc.com>

[p***@sweavo.34sp.com]

[ Incidentally, The Famous Brett Watson send me this link which was interesting:
http://vanquish.com/ I don't recall whether he copied the list ]

Post by Casey Allen Shobe
Any of these money-based approaches I've heard scare me, and not just a
little.

Don't be scared! It'll be alright *pats on back*

Post by Casey Allen Shobe
I have 45,000 emails in this account - you propose that people would have had
to invest a total of $450.00 in me to let me read them, at risk of paying a
lot more?

How long did it take you to accumulate these mails? How many senders? In
short, yes. The longer answer is I feel there's a way in to this situation by
starting it up between corporations who have partnerships. Warranted
communication between employees in key roles. It would be good to come up with
a scalable distributed infrastructure that could work its way in that thru that
niche and slowly grow as more people want to communicate with those
organisations. Later, only AOLers and spammers would use the old style email.

Come to think of it, it could start out as a simple addition to SMTP - return an
error that a guarantor is required. Hmmm...

Post by Casey Allen Shobe
You're posting to a mailing list, but you seem to forget mailing lists in
your approach?

Hmm, yes. The "fine" part holds good, but the "creaming off" part for the ISP
falls down a bit. Looking at only the fine part for now, yes, it's a single
recipient from my point of view. The mailing list owner/sponsor has to carry
the risk of recipients not liking the mail. This leads to responsible,
hands-on administration of mailing lists.

Post by Casey Allen Shobe
If the mailing list has 2000 subscribers, there's no way anyone
will pay $20 to send a single email, plus the risk of some people saying it's
not wanted.

Point taken. I hadn't thought through the charging part, the main idea was the
fine part but then I realised someone whould have to administer the
infrastructure. I'm starting to lean towards a flexible charging scheme that
accumulates with each hop. The sender can set thresholds at which the delivery
will instantly be cancelled, and MTAs can set their own wanted and unwanted
charges.

Post by Casey Allen Shobe
$0.50 may be a lot to you for an unwanted mail, but it's nothing to the
people who make money off of sending them out.

Here I disagree. As I understand it, spam is cost effective only because it is
*SO* cheap to send junk mail. At $0.50 you might still get junk mail, but it
would be more carefully targeted, resulting in less traffic and a better
operational internet.

Post by Casey Allen Shobe
E-Mail is free and should always be free. Any system that is not will never
succeed.

That's simply an emotional response, the first part of which I share, but I
believe while it remains free it is in the grip of entropy and is destined to
become the CB-radio of the 2010s.

You might as well say everybody on the internet should know each other
personally like back in the days of the ARPANET.

If there's a single reason email needs replacing it's because of spam.

If there's a single reason for the existence of spam it's because it's
effectively free to send.

That's all I have time for right now. But as per my original post on this
subject, it all rather depends on identity guarantees.

Steve.

[James Craig Burley]

Post by p***@sweavo.34sp.com
Here I disagree. As I understand it, spam is cost effective only because it is
*SO* cheap to send junk mail. At $0.50 you might still get junk mail, but it
would be more carefully targeted, resulting in less traffic and a better
operational internet.

So we'd still get spam, just better-targeted spam selling higher-end
products? That's worth charging *everyone* to exchange email
messages?

Consider this. Do people use email to transact business,
e.g. negotiate prices in exchange for goods and services?

Answer: yes. And, normally, any communications system used for such
negotiations is *not*, itself, designed to require up-front payment,
or promise of payment, in order to engage in such communications.

(Either or both parties might pay a third party to intermediate and/or
enable the communications. That's not like email; that's more like
the telephone system, or an airline.)

Generally, when system A depends upon system B to operate properly, it
is unwise to contemplate "improving" system B to depend on system A in
order to operate properly.

So, since markets are increasingly depending upon email (mainly,
two-party email) to transact business, it would be unwise to try to
make using email depend upon engaging in business transactions (which
is what any pay-by-email system basically implies).

After all, how would these financial promises, payments, etc. be
transacted, if not via email? By some *other* communications system?

And how do we keep *that* system from being targeted by spammers out
to wreck the system, if they can't use it directly for spam?

And why can't we apply *those* solutions more directly to the problem
of making *free* email work better?

Post by p***@sweavo.34sp.com
If there's a single reason for the existence of spam it's because it's
effectively free to send.

Yet, here we all are, still using it, and it seems to be working
pretty well.

One way to test whether charging for sending email works: set up a
service whereby subscribers pay to submit their email via the
service's SMTP server (presumably using SMTP AUTH or SMTP-after-POP3
or similar).

Then, see how popular the scheme is, how much of the Internet you can
convince to whitelist your entire server as a relay, etc. (The
financial arrangements might be a bit more complicated than that, but
you get the picture.)

It's cruder than most proposals involving charging for email, but most
such systems start out similarly crudely, letting the market decide
whether fine-grained controls are really necessary only after it
decides whether the *concept* will fly.

[Bryan Campbell]

Although I have not read the preceeding post . . .

I used to agree that e-mail should be free. But, I no longer believe
that. Because, it is not. Anyone who believes they are sending e-mail
for free obviously forgot about the fees they paid to their ISP,
telephone company, or cable company.

The reason that e-mail has been so inexpensive thus far is that it was
inexpensive to operate. Now, due to abuse, it is no longer so
inexpensive to operate. The cost for the operation of e-mail servers
and the bandwidth to support them may seem insignificant to the end
user. But, those costs are not insignificant.

I am subscribed to this list because I believe that the sending party
should be responsible for the e-mail they generate. If the e-mail is
sent for the purposes of marketing, then the marketing people (and not
the ISP) should foot the bill for the costs of the e-mail transport and
delivery. If the e-mail is sent for the purposes of discussion between
ISP customers, that falls within the realm of services for which we pay
our ISP.

Will customers understand? Is there a way to accomplish the goal of
making marketing people pay for the eye-balls? You bet. But, not until
something like im2000 is built . . . if for nothing more than a
reference design.




Here are a few interesting number for ya!

FYI -

45,000 e-mails . . . each read and disposed in 15 seconds (don't forget
download time and the time to delete the offending messages from the
trash) . . . consumes 187 hours of time . . . at $20 an hour . . .that
is $3,750.00. And, that is if you are only earning about $40,000 a
year. That is a little more than a month of your time. That is real
money. How much do you believe your employer . . . or any employer
would think that their employee is worth. Well, a $20 an hour employee
is worth roughly $75 an hour to any random company after taxes, and
benefits. That translates these numbers up a bit. 45,000 e-mails . . .
each read and disposed in 15 seconds . . . consumes 187 hours of time .
. . at $75 an hour . . .that is $14,062.50. Now, do you really think
that $450.00 is adequate compensation for lost time and wages.

Post by Casey Allen Shobe
Any of these money-based approaches I've heard scare me, and not just a
little.
I have 45,000 emails in this account - you propose that people would have had
to invest a total of $450.00 in me to let me read them, at risk of paying a
lot more?
You're posting to a mailing list, but you seem to forget mailing lists in your
approach? If the mailing list has 2000 subscribers, there's no way anyone
will pay $20 to send a single email, plus the risk of some people saying it's
not wanted. If you say "well it's a single recipient" then who's going to be
in charge of deciding for the mailing list in an unmoderated case?
$0.50 may be a lot to you for an unwanted mail, but it's nothing to the people
who make money off of sending them out.
E-Mail is free and should always be free. Any system that is not will never
succeed.
Cheers,

[p***@sweavo.34sp.com]

Post by Bryan Campbell
I used to agree that e-mail should be free. But, I no longer believe
that. Because, it is not.

... [snip]

Great post Bryan, you put it much more lucidly than I could have!

[James Craig Burley]

Post by Bryan Campbell
The reason that e-mail has been so inexpensive thus far is that it was
inexpensive to operate. Now, due to abuse, it is no longer so
inexpensive to operate. The cost for the operation of e-mail servers
and the bandwidth to support them may seem insignificant to the end
user. But, those costs are not insignificant.

If you think email is expensive to operate *now*, wait until you see
how expensive it'll be for everyone to deal with a micropayment-based
email system like the one proposed earlier!

Post by Bryan Campbell
I am subscribed to this list because I believe that the sending party
should be responsible for the e-mail they generate.

I agree. But, at some point, a receiving party *will* want to take
responsibility for a given email.

SMTP and IM2000 each introduce excess expense into the process of a
receiving party deciding whether and just when to accept such
responsibility.

Therefore, I believe the solution is to design and deploy an email
protocol that reduces that expense while still providing the
flexibility of a range of options that includes the SMTP "push" and
IM2000 "pull" models.

Since SMTP is already rolled out, and IM2000 is "new", it might be
easier to make IM2000 more flexible prior to its design being fleshed
out and its deployment.

[Brian Candler]

Post by Bryan Campbell
I am subscribed to this list because I believe that the sending party
should be responsible for the e-mail they generate. If the e-mail is
sent for the purposes of marketing, then the marketing people (and not
the ISP) should foot the bill for the costs of the e-mail transport and
delivery.

That, unfortunately, means some sort of payment-with-message system as far
as I can see.

Whilst IM2000 means the sender has the cost of *storing* the message, the
recipient still has to bear:
- the bandwidth cost of receiving the notification (small, although could
become larger if they are resent for many days)
- the bandwidth cost of retrieving the message body, if they choose to do so
- most importantly, the filtering cost of scanning the notification
and/or message headers and/or message body to determine whether the
message is wanted in the first place

"Filtering cost" =>
- annoyance and time-is-money cost of having to do this manually; or
- complexity cost of having to install and maintain automatic filters,
and consequential losses from false positives; or
- monetary cost of paying someone else to filter your mail for you

Post by Bryan Campbell
Will customers understand? Is there a way to accomplish the goal of
making marketing people pay for the eye-balls? You bet. But, not until
something like im2000 is built . . . if for nothing more than a
reference design.

Well, a payment-carried-with-marketing-message infrastructure could be built
on top of IM2000 (or SMTP), but I don't see a concrete proposal yet for how
that would work.

Anyway, who gets the money? The ISP? (Then it's in their interest to deliver
as much marketing as possible). The end-user? Then ISPs have to get involved
in micropayment reimbursement schemes to each of their customers, which
would be a nightmare to administer. A pool of charities? The pool would have
to be acceptable to everyone.

Regards,

Brian.