Email Web of Trust - Problem Statement

I think it would be useful if we can clearly state the terms of the 
problem which the "web of trust" is trying to solve. My try on it is as 
follows:

"One of the problems with spam is that when email is accepted there is 
no way to figure out whether a given MTA is trusted or not. An "web of 
trust" for email is proposed which will make a guess about 
trustworthiness of a given MTA based on trust granted to it by other 
entities."

(this can also be found at 
http://www.shaftek.org/wiki/wiki.pl?Web_Of_Trust_For_Email/Problem_Statement)

Yakov