RE: Time Intervals

From: Carl W. Brown (cbrown@xnetinc.com)
Date: Thu Feb 01 2001 - 13:04:55 EST


Mark,

I did not find anywhere in the discussion of time issues the notion that since the starting point is arbitrary that processing could be simplified by shifting the starting point to a March start. The only problem is leap seconds which are added at the end of the calendar year. But leap seconds are a problem anyway. Most systems usually compensate by adjusting the starting point so that the current time is synchronized. This is adequate for most uses. Thus the 1/1/70 epoch time does not actually match the actual historic time but has been adjusted to compensate for the leap seconds.

On a related subject I am writing a strftime to ICU date format conversion routine and noticed that ICU has no week based year support. Fortunately I don't think my client needs it.

Carl

-----Original Message-----
From: Mark Davis [mailto:markdavis34@home.com]
Sent: Thursday, January 25, 2001 9:18 PM
To: Carl W. Brown; Unicode List
Subject: Re: Time Intervals

Yes, some date calculations are easier that way. But in reference to what
are you saying this?

Mark
----- Original Message -----
From: "Carl W. Brown" <cbrown@xnetinc.com>
To: "Unicode List" <unicode@unicode.org>
Sent: Thursday, January 25, 2001 07:35
Subject: RE: Time Intervals

Mark,

Date calculations are much easier if you start on a March 1 date such as
March 1 1900. This is becase the months are 31,30,31,30,31 31,30,31,30,31
31,xx Putting February last makes leap year calculations easier.

Carl

-----Original Message-----
From: Mark Davis [mailto:markdavis34@home.com]
Sent: Monday, January 22, 2001 9:32 AM
To: Unicode List
Subject: Re: Time Intervals

This appears to have bounced the first time I sent it.

----- Original Message -----
From: "Mark Davis" <markdavis34@home.com>
To: "Unicore" <unicore@unicode.org>; "Unicode" <unicode@unicode.org>
Sent: Monday, January 22, 2001 08:04
Subject: Time Intervals

> After a request from Tex, I realized that I hadn't posted my notes on the
> issues with ISO 8601 on this list. (More exactly, issues with a previous
> version of Schema, but since ISO 8601 is not well-defined, it wasn't
> either.)
>
> The page is at http://www.macchiato.com/unicode/timeIntervals.htm
>
> Mark
>
> ----------
> http://www.macchiato.com
>



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:18 EDT