RE: Functions on Dates, Times and Durations

Dear Ashok,

> 1. Durations that contain both days and months
> do cause problems when used without a reference
> dateTime.

Well, this reflects exactly what I am saying,
in a more radical way, though:
Do not use a duration without a reference dateTime!

There are two (including the Software AG proposal
three) possibilities considered to avoid problems:

1) Split durations into two subtypes which can be
   added commutatively, associatively, etc.
2) Project durations to two numeric dimensions
   and then add numbers (that is commutative etc.).
3) Not use duration without dateTime

> We have chosen to address this problem
> by creating the 2 subtypes of duration
> which contain either days or months.

The taskforce decided for the first option.
And I know that the functions offered there
are sufficient to support my use case of regular
time series. It is more complicated, but feasible.

> Would you like this decision to be reopened?
> If you do, I would request you to provide
> a compelling usecase.

I admit that a simpler API alone makes no use case.
Nevertheless, when the taskforce will discuss the
SoftwareAG's proposal about the alternative 2
(to use numbers instead of duration subclasses),
it may be of interest to have my thoughts as
additional background material.

Actually, I would be interested to hear what the
use cases are behind the requirement to calculate
with (parts of!) durations in absence of a dateTime.
Does anyone know a URL?

Cheers,

Bernhard


References:
http://lists.w3.org/Archives/Public/public-qt-comments/2003Dec/0263.html
http://lists.w3.org/Archives/Public/public-qt-comments/2003Oct/0029.html

--- END ---


________________________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping" 
your friends today! Download Messenger Now 
http://uk.messenger.yahoo.com/download/index.html

Received on Friday, 19 December 2003 11:59:09 UTC