Tony Hansen wrote:
The consensus on this text
It is worth noting that the file transfer protocol (FTP [35]) uses a
very similar code architecture and that the SMTP codes are based on
the FTP model. However, SMTP uses a one-command, one-response model,
while FTP is asynchronous and the 1yz codes are not part of the SMTP
model.
is almost dead even. I'll leave this one open for two more days, then
give a consensus call. If it remains even, it will become "editor's choice".
This is all minor to me so I can go either way, but........
I voted for the removal on the basis that the 1yz is no longer part of
the spec and that throwing in 1yz may raise more questions than not for
new developers, especially those who only interest is SMTP and have not
other protocol perspective to lean on.
I believe, if I read John right, it was stated as such to explain the
removal of 1yx - "where did it go?" In my view, that is more logical if
the 1yz description was not removed. In this case, as it is stated, it
makes more sense to explain why it should only be used for extensions.
Alternatively, IMO, I think it is may be more acceptable if a rewording
was stated as a theory basis in section 4.2.1 as the initial paragraph.
Example:
4.2.1. Reply Code Severities and Theory
SMTP inherits the reply code model and theory from the file
transfer protocol (FTP [35]). However, as described in this
section, only specific reply-code ranges are acceptable.
or
4.2.1. Reply Code Severities and Theory
The SMTP reply code model uses a similar C/S (client/server)
command response architecture that is based on early
C/S protocol designs such as file transfer protocol (FTP [35]).
I like the latter because it makes no assertion that the SMTP reply
codes are based on FTP, but based on similar frameworks such as FTP.
I think the overall issue is that no assertion or implication should be
made to the reader that he/she should research FTP to understand SMTP
better.
My opinion, of course.
--
HLS