Dave Crocker <dcrocker(_at_)brandenburg(_dot_)com> wrote:
An exercise like the one you are doing can be very useful.
I also have planned another document, which discusses which
protocols are used in what situations to send email. It will point
out which use models, and which fields of which protocols, are being
abused to send spam.
Let me strongly encourage you to re-cast the bulk of your document
into non-technical issues, first.
I'm not sure why. The intent of the document is clear: To compare
and contrast SMTP to other protocols, and to explain why and how
technical features in SMTP are being abused to send spam.
The only non-technical issue I can see being relevant is that any
proposed changes to the use or deployment of SMTP should minimally
affect the way users currently send messages. This issue can be dealt
with, without getting into any details of human communication modes.
That is, discuss messaging and a
human process, without concern for whether it is paper mail, email, or
whatever. As you offer criticisms of the human communication modes
available for Internet mail, compare whether they differ from the
modes available for other messaging.
I honestly don't see why. The intent of the document is to describe
what's wrong with SMTP today, not to describe why or how people use
SMTP to send messages. End-user human communication modes are not
relevant when discussing protocol flaws in SMTP, between MTA's.
After we have some consensus about these issues, in terms of human
communication, it will then be reasonable to see what changes are
needed to the technology.
The discussion of human intentions and beliefs about which processes
*should* be used to send email is a topic which has spawned countless
flamewars on this list. After being burned a number of times, I would
prefer to avoid those issues. If you (or anyone else) writes such a
document, I'm willing to read it. But I can't contribute to one.
By focussing on technical issues, I can avoid these human issues, so
long as I ensure that any technical recommendations do not affect
current use patterns or beliefs about email properties (whatever those
are.)
The only other reason I can see to discuss human communication modes
and intentions is to see if the design of SMTP fulfills our goals for
human communication. Since SMTP has been around for 2 decades, it's a
little late to have that discussion.
A discussion of human communication modes is best suited for a
document which describes the requirements for a protocol to replace
SMTP. In that context, it's benefical to define our human
requirements and goals; along with the technical limitations of SMTP.
This will let us be sure that we don't make the same mistakes again.
But I do not propose replacing SMTP.
Alan DeKok.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg