![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Robert Elz writes: [ about changing " " to "211-" in a proposed FTP extension ] > That's certainly just a cosmetic change, I'm continually amazed at how little attention the IETF pays to these basic protocol engineering issues. In the real world, Qualcomm has just released a version of Eudora Pro that's incompatible with thousands of non-sendmail SMTP servers around the world. It fails whenever it tries to send a styled message, because it incorrectly terminates SMTP lines with \n instead of \r\n. Oops. Exercise for the reader: How could SMTP have been engineered so that Qualcomm would not have fallen into this trap? (For some hints, see http://pobox.com/~djb/proto/design.html. For a complete answer, see http://pobox.com/~djb/proto/qmtp.txt.) Back to the proposed extension. I object to FEAT on the grounds that its response format is much more complicated than it needs to be. The format should be redesigned to make client parsing as easy as possible. In particular, the initial text should be scrapped, and the current "\r\n " separator should be replaced by a single-byte terminator such as "/". ---Dan P.S. I'm sending this to ftp-wg at hethmon.com and ietf at ietf.org. Thanks to the misguided ftp-wg MLM, ftp-wg subscribers probably won't see the complete To line.
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.