Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site hou3c.UUCP
Path: utzoo!watmath!clyde!burl!hou3c!KIRK.TYM@OFFICE-2.ARPA
From: KIRK.TYM@OFFICE-2.ARPA (Kirk Kelley)
Newsgroups: net.mail.msggroup
Subject: Re: user interfaces (identifier date time)
Message-ID: <439@hou3c.UUCP>
Date: Tue, 3-Apr-84 18:23:00 EST
Article-I.D.: hou3c.439
Posted: Tue Apr  3 18:23:00 1984
Date-Received: Sat, 7-Apr-84 02:42:13 EST
Sender: ka@hou3c.UUCP (Kenneth Almquist)
Lines: 21
To: Robert Elton Maas <REM@Mit-Mc.ARPA>
To: John Covert <covert%castor.DEC@purdue-merlin.ARPA>
Cc: MsgGroup@Brl-Aos.ARPA

In

   DATETIME := DD M YY H MM

Transposing YY and DD would be ok.  The rationalization for persisting with the 
"standard" ordering is to minimize the amount needed to uniquely identify a 
message from a certain person in your personal monthly or yearly subset in the 
case where you happened to get only one message from that person that day.  I 
SAID it was a rationalization.

The original spec reveals why it is H instead of HH.  The hour is represented as
a single letter from A to X.  For example, a message sent April 2, 1984 at 5:57 
am GMT would have an identifier with a datetime component thus:

   02A84E57

In the design of this format, conciseness was as important a requirement as 
human decipherability.  Other proposals that better meet both requirements would
be more interesting than complaints about how cryptic (or vebose) this one is.

 -- kirk