- should not depend on universal deployment to be effective
This is highly desirable, but not a rigid requirement.
none of these are rigid requirements. actually this is a pet peeve of mine -
if these exercises get seen as attempts to establish lists of requirements
then everyone tries to justify his pet feature as a requirement. they are
"criteria" or "desirable goals". and it's perfectly possible and reasonable
for one desirable goal to run afoul of another one.
the point about universal deployment is this - if in order to be useful, a
spam-blocking scheme requires wide deployment, and no benefit results until
that condition is met, then if there is doubt that the scheme will be widely
deployed, it becomes difficult to get people to invest in it. all things
being equal, a scheme which can be incrementally deployed and which produces
some immediate benefit for those who invest in it is more likely to succeed.
If I choose to adopt an email solution that requires senders to
me to use a particular new technology in order to reach me,
is that a dependency on universal deployment?
it depends on how you define success. if by imposing the particular new
technology you get rid of spam without discouraging people from sending you
mail that you want to receive, then it was successful for you. if it does
discourage people from sending you mail that you want to receive, then it
probably was not successful for you. of course, what is successful for you
might not be successful on a large scale.
Keith
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg