[Ietf-calsify] Re: PERIOD types

Reinhold Kainhofer reinhold at kainhofer.com
Mon Apr 11 15:08:00 PDT 2005


Am Montag, 11. April 2005 23:44 schrieb Jonathan Lennox:
> On Monday, April 11 2005, "Reinhold Kainhofer" wrote to 
"ietf-calsify at osafoundation.org" saying:
> > Since in leap seconds the 59th minute has 60 seconds, in both cases the
> > DTEND would be 19990101T000000Z (since 1M should mean: just add 1 to the
> > minutes count, leave the seconds untouched).
> >
> > > (Note that if it just splits into hours, minutes, seconds etc, and then
> > > adds to the minute, it might looks like it handled leapseconds, while
> > > in fact it didn't. or the other way around.)
> >
> > I think that 1M doesn't mean "add 60 seconds", rather it means "add1 to
> > the minutes" (even if the minute doesn't have exactly 60 seconds).
>
> Given that, how would you handle my case D?
>
> DTSTART:19981231T235960Z
> DURATION:PT1M
>
> "Add one to the minutes, leave the seconds alone" would imply an end time
> of 1999-01-01 00:00:60 UTC, but there is no such time.

Intuitively I'd say 00:00:59 (i.e. the last second of the minute). Of course, 
that can't be deduced from any standard, but that's what I as a user would 
expect. Of course, for me as a programmer it would be simpler to just 
understand 1M as 60 seconds, but I doubt that this is what a user would 
expect.

> This is nit-picky for leap seconds, which nobody really handles anyway, but
> it's important for daylight saving time shifts.

Exactly, you run into the same problems. In particular, take (March 27, 2005, 
is the date when summer time was introduced everywhere in the EU):
DTSTART;TZID="Europe/Vienna":20050326T023000
DURATION:P1DT
Since the hour from 02:00 to 03:00 is skipped, there is no 02:30 on March 
27.... I'd say, the event should end at the last valid time before that time. 
That would mean it would end a 02:00. 

Actually this would make sense insofar as before 02:00 the event was still 
going on. Then the time jumps to 03:00, and then the event is not going on 
any more according to the calculation without time jump.

From a user's perspective: If the event starts at 2:30 and takes one day, I 
wouldn't expect it to end at 3:30, even if there's no 2:30.

(Actually, KOrganizer still shows an end time of 02:30, since the UI doesn't 
show time zone shifts.)

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/20050412/03663d11/attachment.bin


More information about the Ietf-calsify mailing list