RE: Recurring Anniversary Instances?

[Doug Royer <Doug.Royer@xxxxxxx>]

> From: Bruce_Kahn@iris.com
> Date: Fri, 30 Jul 1999 17:12:29 -0400
> 
> Doug shot back:
> >> Is DTSTART;VALUE=DATE:19990730 equivalent to 
> >> DTSTART;VALUE=DATE-TIME:19990730T000000Z?
> >
> >No, not at all!
> >
> >> Is DTEND;VALUE=DATE:19990730 equivalent to 
> >> DTEND;VALUE=DATE-TIME:19990730T235959Z (or should that be 
> > DTEND;VALUE=DATE-TIME:19990801T000000Z since its "non-inclusive")?
> >
> >No, not at all!
> 
> So, tell me then what is the effective 'starting time' (or 'ending time') 
> of each of the following:
> 
> 1: DTSTART;VALUE=DATE:19990730

It's only valid if you ask: "Is this event this day (19990730)?".
It has no time.

> 2: DTSTART;VALUE=DATE-TIME:19990730T000000

00:00:00

> 3: DTEND;VALUE=DATE:19990730

It's only valid if you ask: "Is this event this day (19990730)?".
It has no time.

> 4: DTEND;VALUE=DATE-TIME:19990730T235959

23:59:59

> (2 are freebies I know but included for refrence.).  When I used the term 
> equivalent I mean they refer to the same point in time; not that they are 
> necessarily interchagable (ie: mixing a VALUE=DATE w/a VALUE=DATE-TIME, 
> etc).  Until now I thought we had agreement on this but perhaps not. 
> 

In CAP we had said that you must ask for events in ZULU and not
local time. I think this is going to be a bug given the CS has
no idea what TZID the CUA is in, and the CUA has no idea what TZID
the CS is in.