[TriLUG] postfix spam blocking

[Kevin Hunter]

At 2:54pm -0500 Fri, 16 Dec 2011, Jym Williams zavada wrote:
> Times have changed, expectations and needs have changed, but the
> design of e-mail hasn't. I still maintain that people are relying on
> e-mail in ways for which it was never designed.

Could you elaborate on what some of these ways are?  What function(s) do 
people in your view rely on from email that it doesn't support?

> Unfortunately e-mail has no "SMTPv6" (and corresponding
> client-side) standards even ready for implementation, forcing
> everyone to make do with what we currently use, along with various
> "ad hoc" solutions like domain keys, SPF, RBLs, greylisting, etc.

You'll get no argument from me on the implementation of it, but I'm 
questioning your original assertion that it's not dependable and should 
not be used for instantaneous communication, as seen from the 
user/business standpoint.  My evidence is of statistics and profit motive.

My personal (and perhaps anecdotal) statistics suggest that email "just 
works, and instantly."  In my little world, email gets there, for all 
intents and purposes, as soon as I send it, and I receive email in 
similar time.  It doesn't get lost, and, shy of certain lists ending up 
in my spam filter (Barracuda, btw.), pretty much "just works, and 
instantly."

The profit motive evidence is the fact that companies base their entire 
existence on the expectation that email can -- and does -- "just work, 
and instantly."  I again refer to exhibit A of Gmail.  Or the ensuing 
work that other companies capitalize on, like the Barracuda Networks 
that was at the head of this discussion.

Do you have different evidence than myself?  I freely admit that I'm 
*not* an email infrastructure professional.  (In fact, where *are* the 
email professionals?!  We have more than a few on this list, and I'd 
love to read their thoughts.)

I'll follow up with my "expert" thoughts on a better infrastructure.

Cheers,

Kevin