Received: from localhost by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id SAA26194; Thu, 30 Nov 1995 18:50:17 -0500 Received: by cs.cs.utk.edu (bulk_mailer v1.3); Thu, 30 Nov 1995 18:50:04 -0500 Received: from mercury.Sun.COM by CS.UTK.EDU with ESMTP (cf v2.9s-UTK) id SAA26141; Thu, 30 Nov 1995 18:50:01 -0500 Received: from Eng.Sun.COM by mercury.Sun.COM (Sun.COM) id PAA14051; Thu, 30 Nov 1995 15:49:28 -0800 Received: from sage.Eng.Sun.COM by Eng.Sun.COM (5.x/SMI-5.3) id AA29135; Thu, 30 Nov 1995 15:49:25 -0800 Received: from sage (localhost) by sage.Eng.Sun.COM (5.x/SMI-SVR4) id AA24678; Thu, 30 Nov 1995 15:49:23 -0800 Message-Id: <9511302349.AA24678@sage.Eng.Sun.COM> X-Mailer: exmh version 1.6.5 11/22/95 To: ietf-drums Subject: comments on smtpupd-01.txt X-Url: http://www.smli.com/~bwelch/bwelch.html X-Face: "HxE|?EnC9fVMV8f70H83&{fgLE.|FZ^$>@Q(yb#N,Eh~N]e&]=>r5~UnRml1:4EglY{9B+ :'wJq$@c_C!l8@<$t,{YUr4K,QJGHSvS~U]H`<+L*x?eGzSk>XH\W:AK\j?@?c1o First - I just took a plane ride and read draft 00. I'm taking a quick look at draft 01, but some of my comments may have already been addressed in 01. I agree with the general outline proposed by Paul Hoffman, but I haven't looked at it in detail. As someone that never read 821 before, I agree with Pauls notion of why/how to order things. For example, I found the leap into a discussion about extentions and EHLO to be pre-mature. In fact, I'd like to see a list of known extentions and references to the RFCs that describe them. Three paragraphs after the DISCUSSION in Section 2.1 the following statement is made. It is backwards: "More generally, the MAIL address is a forward-path and the RCPT address a reverse-path." The address in the MAIL command is the reverse-path, and the address in the RCPT command is the forward-path. The description is made correctly in section 4.1.1 Section 3.1 begins with somewhat vague statement: "An SMTP session is initiated by the client opening a connection to the server and the server responding with an opening message." More specifically, the server listens on a specific TCP/IP port, a client opens a TCP connection to that port, and the server writes an opening message back to the client. I dunno - does the use of TCP/IP go without saying? Is the port number specified anywhere? The R: and S: notation in the examples is not explained. There is a cryptic note a bit eariler about SMTP-Sender == client SMTP-Receiver == server but this should be clearer Section 3.8 - a general comment. There is a brief mention earlier that relaying is a bad idea, right? However, there is a lot of detail in the spec about how to do it and how it affects the forward and reverse path. A more detailed comment: One more sentence to explain the syntax of the forwarding path would be nice. All it says is Of the form "@ONE,@TWO:JOE@THREE". You could add something like Colon (:) separates the final address from a forwarding path that is a comma (,) separated list of mail server hosts. Section 4.1.1.3, end of second paragraph, the word "user" is used instead of "use" The Time Stamps in the Example 10 do not follow the timezone format urged in the text. It is just "PDT", not "-0700 (PDT)" Section 4.6.4.1, last paragraph of the DISCUSSION section. "Note that an SMTP can generally determine" seems like it should read "an SMTP client can" Appendix B, the section about Bcc: and when to insert a Bcc: header was confusing. Does the recipient of a Bcc always see the Bcc: header in the mail they get? Or do they usually not see themselves in the headers at all? envelope is not in the glossary. -- Brent Welch Sun Labs brent.welch@sun.com