Hi John. It's been a few years since we met at UofA. Good to see you are
still active with email. Thanks for your thoughtful response below. I
would love to see a three-lecture series like you describe, and I would
gladly drop some other material to make room. Unfortunately, as a mere
"Industrial Associate" I was never able to squeeze in more than one lecture.
You make a good point about the complexity of real-world systems, and the
fact that our classroom presentations can't really capture all the
important details. There is always a compromise between being informative
and keeping it simple. http://en.citizendium.org/wiki/Email_system is my
best shot at a one-lecture compromise. Figure 4 in RFC 5598 is simple, but
not very informative. Fig. 5 is informative, but not simple.
Most of what I see in computer network textbooks errs on the side of messy
detail. What's important at this level is focus on fundamentals and
building a solid foundation of understanding. I don't let students think
that my presentation is the whole story. I show them Fig. 5 in RFC 5598 to
let them know there is lot more to learn if they ever get serious about
email systems. I firmly believe that students are better prepared to learn
these details, if they start with a good understanding of the fundamentals.
On Sat, Sep 28, 2019 at 12:58 AM John C Klensin <john-ietf(_at_)jck(_dot_)com>
wrote:
--On Friday, September 27, 2019 16:45 -0700 David MacQuigg
<macquigg(_at_)gmail(_dot_)com> wrote:
...
Imagine you are trying to teach the fundamentals of email
systems to university students in a class on computer
networks. Most are not going to become email system admins.
Most will be overwhelmed by the level of detail in Fig. 5 of
RFC 5598. You have one lecture to give these students a basic
understanding of how email systems work. How would you do it?
I'm busy imagining. I'm imagining that one of the most
important messages I'd want to deliver to such students is that
things are often more complex than they appear at first glance
and/or more complex than one would ideally like them to be. I'd
also want them to learn that it is important to know the
difference between models that reflect the real world and what
we politely call hand waving. I'd think about how much easier
it would be to teach TCP/IP if I ignored congestion control and
how much shorter the email talk could be if I could ignore the
implications of gateways to other types of mail systems and
relays, especially relays under different administration/
control than either the user-originator, MSA and associated
systems and the user-recipient, MDA, and associated systems. I
might then decide that email was not a very good example of an
application in the modern Internet because just about everything
else assumes a direct and real-time originator-recipient
connection. If I did, I'd wave my hands for maybe 10 or 15
minutes (and make it clear that I was doing that) and then move
on to a more productive example. Or I might conclude that a
well-developed lesson about growing complexity with changing
needs was really important. If I did, I'd start with the model
of the early 1980s, early enough to be well before the DNS and
particularly before RFC 974. I'd then walk through examples of
how and why the complexity developed until I reached and could
discuss the present reality, including, as Valdis indirectly
points out, distributed and mirrored mailstores with mechanisms
for moving stored messages around to provide better performance
is the user moves from one geographical location to another. It
that took three lectures rather than one, something else in the
course would need to go.
YMMD, but you asked.
best,
john
_______________________________________________
ietf-smtp mailing list
ietf-smtp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf-smtp