Re: [Michael Kay] RE: <enumeration> of duration types from Dave Peterson on 2005-04-18 (www-xml-schema-comments@w3.org from April to June 2005)

From: Dave Peterson <davep@iit.edu>
Date: Mon, 18 Apr 2005 10:58:45 -0400
To: ht@inf.ed.ac.uk (Henry S. Thompson)
Cc: www-xml-schema-comments@w3.org
Message-Id: <a0611040dbe8977e1f4bf@[192.168.3.6]>

At 3:39 PM +0100 050418, Henry S. Thompson wrote:

>So what about P365D?

P365D <> P1Y ("incomparable", not just "not equal") because
     1903-07-01T00:00:00Z + P365D = 1904-06-30T00:00:00Z ,
     1903-07-01T00:00:00Z + P1Y = 1904-07-01T00:00:00Z ,
     1904-06-30T00:00:00Z < 1904-07-01T00:00:00Z , and therefore
     1903-07-01T00:00:00Z + P365D < 1903-07-01T00:00:00Z + P1Y
whereas
     1903-03-01T00:00:00Z + P365D = 1904-03-01T00:00:00Z ,
     1903-03-01T00:00:00Z + P1Y = 1904-03-01T00:00:00Z , and therefore
     1903-03-01T00:00:00Z + P365D = 1903-03-01T00:00:00Z + P1Y

(1903-03-01T00:00:00Z and 1903-07-01T00:00:00Z are two of the
four designated dateTime values that are used in duration
equality/order calculations.)
-- 
Dave Peterson
SGMLWorks!

davep@iit.edu

Received on Monday, 18 April 2005 14:58:58 UTC