Re: [Asrg] 2a. Analysis - Developer Perspective

[Kee Hinckley <nazgul@somewhere.com>]

At 8:33 PM -0400 9/12/03, Hector Santos wrote:
> Listen, I'm well experience enough to look at all these proposals, using
> your +/- check off list and tell you that the ultimate solution is a
> restriction/authorization concept.    Do you design, develop and market a
So, you've made the technical decision.  Now the next question of 
course is a market question.  Will it be adopted?  (And this 
discussion has occurred in the archives many times.)

Prior experience indicates that the adoption of similar systems takes 
years.  In the meantime such systems need to run in parallel.  Since 
unconstrained gateways would simply import the spam problem, the 
system needs to remain unconnected to SMTP.  So anyone who is an 
early adopter still has to deal with all the spam they get via SMTP. 
They then find themselves with two systems.  One of which has a cost 
of of entry that is non-trivial in terms of support and registration 
costs (and you need to figure out how to scale that cost across 
national borders, or else cut out large portions of the world), has 
no spam, and which communicates with virtually nobody.  The other 
system has spam, but communicates with everyone they want to talk to. 
Or in other words--early adopters pay more, but get no benefit.

Or, since we are putting this in terms of dead email systems.  What 
you are proposing is very similar to back in the late 80's when I was 
developing mail clients for X.400.  X.400 has a lot of the attributes 
you are looking for.  Strong registration and directory services, 
very centralized.  It wasn't adopted. There was this simpler system 
out there that allowed everyone to talk to everyone. I can see no 
reason why it would be adopted this time around (even if 
significantly simplified).  The barriers to entry for a new system 
are large, and early adopters receive no benefit until far more than 
50% of the population has switched.

What FidoNet didn't see coming, and what also killed Prodigy, is the 
simple fact that people like to talk to people.  The web has brought 
people to the internet.  But the ability to talk to anyone keeps them 
here.  Any move to a centralized and restricted system will be 
ignored.

Also, you neglected the other post-implementation detail. 
Enforcement.  Let's assume that your system is 100% implemented.  Now 
someone from my system sends mail to someone on your system.  Your 
someone claims it's spam.  Mine claims it isn't.  Who presides over 
the dispute?  Who gets penalized?  Who pays for the cost of the 
dispute?  What happens when government borders are involved?  What 
happens when *governments* are involved?  What you are describing is 
incredibly expensive to run.  At very best you might end up with a 
highly specialized and constrained system used by corporations and 
wealthy individuals for mission critical communications.  But the 
rest of the world would use regular email for their communications. 
This group is chartered with looking at the problem with email.  The 
new technology would not be email as we know it, and since email as 
we know it wouldn't go away, our focus needs to remain on the 
end-to-end, available-to-all system that will remain.
--
Kee Hinckley
http://www.messagefire.com/  Anti-Spam Service for Companies and Individuals
http://commons.somewhere.com/buzz/  Writings on Technology and Society

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.

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