On Thu, 1 Jan 2004 19:07:17   
 Carleton MacDonald wrote:
>I dono?=t like the idea of 24:00:00, because there is no 24:00:01.

Interesting perspective, quite valid point.

>   Using
>24:00:00 also means not using 00:00:00, an illogical situation because there
>is a 01:00:00, 02:00:00 etc.
>
Actually, perhaps what should happen with watches is to allow them to post 24:00 very 
briefly (fractions of a second) just to show the day has finally ended its 24 hours 
and start the new day *immediately* thereafter, i.e. evidently at 00:00, all within 
the first second of the "new" day!

>The changeover from 23:59:59 to 00:00:00 is simple and logical and
>understandable;

I must strongly concur with Carleton here.  Indeed.  The only comment I'd like to add 
that perhaps should solve this for once and for all is: from what **day** perspective 
are we alluding to this value?  The "old" one or the "new" one?

In other words, as somebody else already shared here.  Both:

24:00 2003-Dec-31, and
00:00 2004-Jan-01

refer to the exact same *position* in the "time circle" (another circle metaphor now 
applied to the day construct).

So, take your pick!  On the other hand, as far as watch construction goes, Carleton's 
point is dead on.

> midnight therefore belongs to the new day, not the old one (assuming it'd be a 
> hassle to show 24 or simplicity of construction is aimed at).

Therefore, if one is referring to midnight then perhaps it would be far more logical 
to refer to this as 31 Dec to be coherent/consistent (borrowing from my example above).

As for the other point below (the am/pm thing) I'll refrain to comment on it since I 
never *ever* use such crappy stuff, so I really don't care one way or another...  ;-)

Marcus

>Converted to 12-hr, this makes 12:00:00 am midnight, as it is immediately
>followed by 12:00:01; 12:00:00 pm is noon because it is immediately followed
>by 12:00:01 pm...


____________________________________________________________
Get 25MB of email storage with Lycos Mail Plus!
Sign up today -- http://www.mail.lycos.com/brandPage.shtml?pageId=plus 

Reply via email to