[prev in list] [next in list] [prev in thread] [next in thread] 

List:       imap
Subject:    re: base specification on sequences
From:       Mark Crispin <MRC () cac ! washington ! edu>
Date:       2002-09-23 16:46:20
[Download RAW message or body]

Here's my own commentary on the current wording:

                       [...]  For message
                       ; sequence numbers, these are consecutive
                       ; numbers from 1 to the number of messages in
                       ; the mailbox

This does not belong in the definition of "set"; it belongs in "sequence-num"
instead.

                       ; Comma delimits individual numbers,

This echoes the ABNF, but does so incorrectly (as was pointed out in the
message that I quoted).  I don't think that it adds any information, and it
adds confusion, so it should be removed.

                       [...] colon
                       ; delimits between two numbers inclusive.  Note
                       ; that n:m and m:n are equivalent.

"delimits between two numbers inclusive" is the cause of a lot of problems.
People confuse this use of "inclusive" with the mathematical concept of
"inclusivive subset of a natural sequence".  Reasonable people accept being
told that IMAP doesn't use the mathematical definition of "inclusive".
Unfortunately, there are unreasonable people who insist that the mathematical
definition trumps IMAP, even if IMAP's definition is more useful.

I didn't even write the wording in the current document; I eschew "inclusive"
in my speech or writing.  My wording was "all numbers between those two
numbers", which was deemed unsuitable since some people claimed that it did
not include the two endpoints.

The addition of "Note that n:m and m:n are equivalent" is an attempt to fix
the current problem.  Perhaps there is a better solution?

[prev in list] [next in list] [prev in thread] [next in thread]