RE: Year 0000

Thank you for your detailed answers.  This seems worthy to appear as a
footnote to the XML Schema specification to clear up misunderstandings
that might easily appear since the mapping from lexical space to value
space will sometimes not match naive expectations.  

	-----Original Message-----
	From: zongaro@ca.ibm.com [mailto:zongaro@ca.ibm.com] 
	Sent: Tuesday, September 04, 2001 7:22 AM
	To: Ashok Malhotra
	Cc: Andrew Layman; lmartin@ca.ibm.com;
www-xml-schema-comments@w3.org; w3-xml-schema-wg@w3.org
	Subject: RE: Year 0000
	
	

	
	Hi Ashok, 
	
	     Sorry if my note seemed to imply that there was a
discontinuity in the Gregorian calendar.  It certainly wasn't intended.
In particular, I was trying to indicate that there are no days ommitted
from October, 1582 in the proleptic Gregorian calendar; i.e., it's a
month of 31 days, just like any other October rather than a month of 21
days.
	
	     As for whether AD 1 is a date in the Gregorian or Julian
calendar, I was describing common usage.  When one reads, for instance,
that Miguel de Cervantes was born on October 9, 1547, that's a date in
the Julian calendar; the same date in the proleptic Gregorian calendar
is October 19, 1547.  To the best of my knowledge, historians don't
usually translate dates prior to 1582 into the Gregorian calendar,
though they might need to for dates after 1582, prior to the Gregorian
calendar's widespread adoption. 
	
	     Dates in the proleptic Gregorian calendar coincide with
dates in the Julian calendar between March 1, 200 and February 28, 300.
Outside of that range they diverge.  The year 0000 in ISO 8601 is the
year 1 BC in the proleptic Gregorian calendar, but it's not the 1 BC of
historians or of common usage, which is a year in the Julian calendar.
It seemed to me that that was the question Andrew was asking. 
	
	     Historians using dateTime and the related data types will
have to be aware of the difference. 
	
	Thanks,
	
	Henry
	
------------------------------------------------------------------------
	Henry Zongaro      XML Parsers development
	IBM SWS Toronto Lab   Tie Line 778-6044;  Phone (416) 448-6044
	mailto:zongaro@ca.ibm.com
	

	Please respond to "Ashok Malhotra" <ashokma@microsoft.com> 

	To:        Henry Zongaro/Toronto/IBM@IBMCA, "Andrew Layman"
<andrewl@microsoft.com> 
	cc:        Lisa Martin/Toronto/IBM@IBMCA,
<www-xml-schema-comments@w3.org>, <w3-xml-schema-wg@w3.org> 
	Subject:        RE: Year 0000 
	
	
	Henry:
	Your response to Andrew was not quite accurate.  ISO 8601 and
XML Schema
	use what is called the "prolaptic Gregorian calendar".  This
means they
	refer to years, months, etc as if the Gregorian calendar had
been in use
	since the beginning of time; specifically before 1582.  If you
take this
	viewpoint there is no discontinuity.
	
	ISO 8601 in its usual cryptic fashion says (section 4.3.2.1 Note
2)
	'Also note that the year numbers for years before the year
[0001] differ
	from the year numbers in the "BC/AD calendar system", where the
year"1
	BC" is followed by the year "1 AD"'   The Encarta article on
calendars
	
http://encarta.msn.com/find/Concise.asp?z=1&pg=2&ti=761560321&cid=8#p8
	says that "The Gregorian calendar is also called the Christian
calendar
	because it uses the birth of Jesus Christ as a starting date. "
So, the
	BC/AD calendar, I assume from the above, uses Gregorian years.
Thus,
	year [0001] corresponds to AD 1 and ... this is my
interpretation ...
	year [0000] corresponds to 1 BC and so on.  This is consistent
if a bit
	weird.
	
	BTW, your notes appear in tiny font.  I assume this is some
idiosyncracy
	of Lotus Notes.
	
	All the best, Ashok
	
	                -----Original Message----- 
	                From: zongaro@ca.ibm.com 
	                Sent: Mon 8/27/2001 6:25 AM 
	                To: Andrew Layman 
	                Cc: lmartin@ca.ibm.com; Ashok Malhotra;
	www-xml-schema-comments@w3.org; w3-xml-schema-wg@w3.org 
	                Subject: RE: Year 0000
	                
	                
	
	                
	                Hi Andrew, 
	                
	                     Perhaps I was a bit imprecise in saying
that the year 0000
	in ISO 8601 is what is commonly called 1 BCE.  The year 1 BCE is
a year
	in the Julian calendar, whereas ISO 8601 treats the Gregorian
calendar
	as if it existed prior to 1582-10-15, and extends it
indefinitely back
	in time.  So 0000 in ISO 8601 and 1 BCE overlap to a great
extent, but
	not necessarily completely. 
	                
	                     1582-10-05 is the day after 1582-10-04, and
1582-10-14 is
	the day before 1582-10-15 in ISO 8601.
	                
	                Thanks,
	                
	                Henry
	                
	
------------------------------------------------------------------------
	                Henry Zongaro      XML Parsers development
	                IBM SWS Toronto Lab   Tie Line 778-6044;  Phone
(416) 448-6044
	                mailto:zongaro@ca.ibm.com
	                
	
	                Please respond to "Andrew Layman"
<andrewl@microsoft.com> 
	
	                To:        Lisa Martin/Toronto/IBM@IBMCA, "Ashok
Malhotra"
	<ashokma@microsoft.com> 
	                cc:        Henry Zongaro/Toronto/IBM@IBMCA,
	<www-xml-schema-comments@w3.org>, <w3-xml-schema-wg@w3.org> 
	                Subject:        RE: Year 0000 
	                
	                
	                In that case, is 1582-10-05 a valid ISO 8601
date?  What day
	follows
	                1582-10-04?  What day precedes 1582-10-15?  Are
the answers to
	these
	                questions consistent with the assumption that
year 0000 is 1
	BCE?
	                
	                -----Original Message-----
	                From: lmartin@ca.ibm.com
[mailto:lmartin@ca.ibm.com] 
	                Sent: Friday, August 10, 2001 2:08 PM
	                To: Ashok Malhotra
	                Cc: zongaro@ca.ibm.com;
www-xml-schema-comments@w3.org;
	                w3-xml-schema-wg@w3.org
	                Subject: RE: Year 0000
	                
	                
	                Yes, I'll add it to our Issues list.
	                Lisa.
	                
	                
	                "Ashok Malhotra" <ashokma@microsoft.com> on
08/10/2001 04:11:14
	PM
	                
	                Please respond to "Ashok Malhotra"
<ashokma@microsoft.com>
	                
	                To:   Henry Zongaro/Toronto/IBM@IBMCA
	                cc:   <www-xml-schema-comments@w3.org>,
	<w3-xml-schema-wg@w3.org>, Lisa
	                     Martin/Toronto/IBM@IBMCA
	                Subject:  RE: Year 0000
	                
	                
	                Yes, I came to the same conclusion after
rereading the 2000
	version of 
	                 ISO 8601. I think we need to consider this as a
possible errata
	item.
	                Lisa, can you please add to the errata list.
	                
	                
	                
	                All the best, Ashok
	
===========================================================
	                
	                    -----Original Message-----
	                    From: zongaro@ca.ibm.com
[mailto:zongaro@ca.ibm.com]
	                    Sent: Friday, August 10, 2001 1:01 PM
	                    To: Ashok Malhotra
	                    Cc: www-xml-schema-comments@w3.org;
w3-xml-schema-wg@w3.org;
	                    lmartin@ca.ibm.com
	                    Subject: RE: Year 0000
	                
	                
	                
	                    Hi Ashok,
	                
	                         Although ISO 8601:2000 isn't entirely
clear on the
	point, it
	                does
	                    say that the year numbers are contiguous.  I
take that to
	mean that
	                    0000 is the year before 0001, and -0001 is
the year before
	that.
	                That
	                    would mean 0000 is what is usually referred
to as 1 BC (or 1
	BCE)
	                and
	                    -0001 is 2 BC (or 2 BCE).
	                
	                         If my understanding is correct, that
means the years
	that are
	                    commonly called 1BC (0000) and 5BC (-0004)
are leap years in
	the
	                    proleptic Gregorian calendar, but 4BC
(-0003) is not.
	                
	                    Thanks,
	                
	                    Henry
	                
	                
	
------------------------------------------------------------------------
	                
	                    Henry Zongaro      XML Parsers development
	                    IBM SWS Toronto Lab   Tie Line 778-6044;
Phone (416)
	448-6044
	                    mailto:zongaro@ca.ibm.com
	                
	                
	                    Please respond to "Ashok Malhotra"
<ashokma@microsoft.com>
	                
	                
	                    To:        Henry Zongaro/Toronto/IBM@IBMCA,
	                    <www-xml-schema-comments@w3.org>
	                    cc:        <w3-xml-schema-wg@w3.org>, Lisa
	Martin/Toronto/IBM@IBMCA
	                    Subject:        RE: Year 0000
	                
	                
	                    I found additional confirmation that the
2000 version of ISO
	8601
	                    does,
	                    indeed, allow the year 0000.  I don't know
what it maps to
	in terms
	                on
	                    AD and BC because AD 1 == 0001 and BC 1 ==
-0001.  Should we
	                discuss
	                    as
	                    a possible errata item.
	                    Ashok
	                
	                                    -----Original Message-----
	                                    From: zongaro@ca.ibm.com
	                                    Sent: Thu 8/9/2001 2:02 PM
	                                    To:
www-xml-schema-comments@w3.org
	                                    Cc:
	                                    Subject: Year 0000
	                
	                
	                
	                
	                                    Hello,
	                
	                                         I just saw a copy of
ISO 8601:2000.  I
	was
	                    surprised to
	                    discover that it defines 0000 to be a valid
year, unlike the
	                    specification of dateTime in the "XML
Schema:  Datatypes"
	                    recommendation
	                    [1].  I gather that in ISO 8601:2000, the
year 0000 is
	roughly 
	                     equivalent to what people usually refer to
as 1BC, and is a
	leap
	                year.
	                
	                                         Should dateTime follow
ISO 8601:2000 in
	this
	                    respect?
	                
	                                    Thanks,
	                
	                                    Henry
	                                    [1]
	
http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#dateTime
	                
	                
	                
	
------------------------------------------------------------------------
	                
	                                    Henry Zongaro      XML
Parsers development
	                                    IBM SWS Toronto Lab   Tie
Line 778-6044;
	Phone
	                (416)
	                    448-6044
	                                    mailto:zongaro@ca.ibm.com
	                
	                
	                
	                
	                
	                
	                
	
	
	
	

Received on Wednesday, 5 September 2001 15:00:50 UTC