[Ietf-calsify] PERIOD types

[Reinhold Kainhofer]

Am Dienstag, 12. April 2005 00:03 schrieb Doug Royer:
> This seems to be the same argument how to find the correct
> end time. If the argument is that is all the Attendee CUA
> has to do is add to DATE (not time), then why can't the
> Organizer CUA do that and send out everything in seconds?

Because it makes a difference when you use RRULE or RDATE (and one of the 
occurences fall on the daylight shift). 

> Existing 2445 implementation would process ##S correctly
> as would CALSIFY implementations.

But that's usually not what users would exists. If I have an event at 13:00 
and it lasts 1 day, I expect it to end on 13:00 on the next day, no matter if 
the night is one hour shorter or longer as usual.

Even if the "big two" (as you put it) don't think like that, I would say 
that's a bug in their implementation. 

> Lets deprecate all but 'S' so that everyone can
> use the same calculation method 

Hell, let's just get rid of all RRULE and RDATE (and the RECURRENCE-ID)! If 
each event is created separately with it's own DTSTART and DTEND, we won't 
have any problems </sarcasm> Just because we find some inconsistencies, 
doesn't mean it's best to simply remove that feature. That's throwing out the 
child with the bath.
I think it's more important to write down how to handle all these cases 
exactly (because that's what lead to the fragmentation: nothing in the spec 
said how to calculate the dtend). 


> Converting from 2445 DURATIONs:
>
> 	I think we could live with saying 1M == 60S.
>
> 	The big two vendor seem to say that 1D = 86400S.

Oh, great. Exchange also didn't use the minus sign in the alarms, so let's 
just disallow the minus in the alarm offset! </sarcasm> Just because one 
company chose one particular implemention (which in my eyes is not what is 
meant in rfc 2445), all others should be forced to use the same brain-dead 
implementation?
There are lots of places where Exchange is not really standard-compliant. Are 
you now trying to argue to adapt the standard to what Exchange implement? 
(And leave all other existing applications in the cold, when they try to get 
things right).


>         And lets say that 1W == 7D.

Yes, that's clear, because no weekdays are skipped or added, even in leap 
years.

Cheers,
Reinhold

-- 
------------------------------------------------------------------
Reinhold Kainhofer, Vienna, Austria
email: reinhold at kainhofer.com, http://reinhold.kainhofer.com/
 * Financial and Actuarial Mathematics, TU Wien, http://www.fam.tuwien.ac.at
 * K Desktop Environment, http://www.kde.org/, KOrganizer / KPilot maintainer
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.osafoundation.org/pipermail/ietf-calsify/attachments/20050412/d97ffc64/attachment.bin