W3C home > Mailing lists > Public > xmlschema-dev@w3.org > July 2001

RE: Date

From: Ashok Malhotra <ashokma@microsoft.com>
Date: Tue, 10 Jul 2001 17:46:45 -0700
Message-ID: <E5B814702B65CB4DA51644580E4853FB8D4D90@red-msg-12.redmond.corp.microsoft.com>
To: <Noah_Mendelsohn@lotus.com>, "Henry S. Thompson" <ht@cogsci.ed.ac.uk>
Cc: "Jeff Rafter" <jeffrafter@definedweb.com>, <vdv@dyomedea.com>, <xmlschema-dev@w3.org>, "Martin Duerst" <duerst@w3.org>, "Martin Gudgin" <marting@develop.com>
Can we just let it go?

I have suggested that the type library add two types -- dateWithTimezone
and dateWithoutTimezone.  These two types are totally ordered and this
is what I think users should use.

All the best, Ashok 
===========================================================
Ashok Malhotra              <mailto: ashokma@microsoft.com> 
Microsoft Corporation
212 Hessian Hills Road
Croton-On-Hudson, NY 10520 USA 
Redmond: 425-703-9462                New York: 914-271-6477 



-----Original Message-----
From: Noah_Mendelsohn@lotus.com [mailto:Noah_Mendelsohn@lotus.com] 
Sent: Tuesday, July 10, 2001 5:28 PM
To: Henry S. Thompson
Cc: Jeff Rafter; vdv@dyomedea.com; xmlschema-dev@w3.org
Subject: Re: Date

Henry Thompson writes:

>> I actually find it pretty surprising that 
>> your users think 12:00:00Z is
>> different from 13:00:00+1

Well, only on the very practical sense that they may wish to ask:  "In 
what time zone was the time recorded?    Roughtly, did you write that at

noon in Boston or 5PM in the UK?  If I put a time on a draft of a 
technical report, for example, certain users will feel that something is

lost if I record it as 3PM in whatever timezone, and it comes back as
1AM 
in some other (presuming someone has read in the value and rewritten as 
canonical, for example.)  It's not broken, just doesn't always do what
you 
might want or expect.  By the way, this should not be taken as a comment

reflecting the needs of Lotus users in particular (that was true of the 
other comment on legal changes affecting timezones.)  I have just seen 
various correspondents on our mailing lists presume that the time zone
was 
reflected in the value space, and wanted to clarify.

>> ...when there were 25 disjoint value spaces.

I don't remember that in detail, but you are probably right.  I thought
I 
remembered a proposal to explictly add the timezone as a field in a
tuple 
representing the value in what I took to be a single value space. 
Presumably, that would have led to the equality relation (enumeration 
facet, etc.) requiring a timezone match.  I also think but am not sure 
that we would have had some freedom in defining ">" and "<">

In any case, all of that is history, and I certainly have no need to 
pursue this issue.  Whatever the merits, it was debated quite clearly at

the appropriate time, and the schema WG made a decision that is 
reasonable, though probably not the one I would have made.  I don't
think 
there is any new information since then.  I was merely trying to clarify

for readers of schema dev the way that timezones do and don't affect the

semantics, and unfortunately I somewhat confused things by giving a 
partially bogus clarification.

So, unless others want to continue, I think we can let this go.  Thanks 
very much.

------------------------------------------------------------------------
Noah Mendelsohn                                    Voice: 1-617-693-4036
Lotus Development Corp.                            Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------------
Received on Tuesday, 10 July 2001 20:57:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 11 January 2011 00:14:22 GMT