[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Asrg] Economic model is borken. (sic.) Let's fix it

Brad Templeton wrote:

I was one of the first people to propose e-stamps for internet
mail.  I have since renounced the idea.

Fully detailed reasons are at:

    http://www.templetons.com/brad/spume/estamps.html

But in short:
    a) You can't get there from here, since you can't reject mail
       that is unstamped until lots of people use stamps and nobody
       will want to use stamps unless people are demanding them.

    b) CPU coins are better than money but still face problem (a)
    c) Putting an artifical cost on something that was designed to
       be cheap is a lousy answer.
    d) It has free speech implications.  (in both senses of the word.)
    e) Mailing lists are a problem.
    f) Major virus problem if you use real money
on the other hand, I'm a strong proponent (still) of using postage with e-mail. Take a look at HTTP://harvee.billerica.ma.us/camram/ for a paper describing the current state of camram project addressing all of the issues you listed above.

In a nutshell:

a) you can get there from here. You just need to use a mixture of techniques
in order to deal with unstamped mail.

b) they certainly are. Including the reason that real money will have
everybody sticking their hand out as the message flows by.

c) yup. It's a trade-off
d) The free speech implications are less than filters. If you generate a
stamp, your message gets delivered. It's no bigger burden than physical
postage.
e) not anymore. Strangers pay, friends fly free. Mailing lists are your
friends (if you signed up for them)
f) I'm not sure I understand the virus problem. I'll take a look at your
paper and see if it gives a better explanation.

I'm currently working on a paper describing the 12+- properties I think an antispam system should have. I'll publish it in the next few days.

another thing that seems to be missing from this discussion is that e-mail systems really are two separate protocols that interact. The low-level protocol is the transport system (i.e. SMTP). As much as people complain about it, the fact that it has survived for twenty plus years speaks to its strengths. It also speaks to how difficult it will be to dislodge. Look at the success we have had with IPv6 and BEEP. Moving away from SMTP will take about 15 years.

the second level protocol is the client to client protocol defined by rfc2822. Granted, it is a one-way protocol but it's still communicating instructions and operations to the recipients e-mail client. In the camram environment, I've tried to focus on modifying just the client to client protocol by adding postage. The rest of the infrastructure focuses on appropriate interpretation of that message and how to dispose of it.

at the end of the day, I look at antispam techniques through the lens of this principle:

Any measure for stopping spam must ensure that all non-spam messages
have a means for bypassing that measure to reach their intended
recipient.

---eric

_______________________________________________
Asrg mailing list
Asrg@ietf.org
https://www1.ietf.org/mailman/listinfo/asrg