Re: Cap Free-Busy Request; some compelling thoughts

[Bruce_Kahn@xxxxxxxxxxxxxxxx]

Craig wrote on 07/23/2003 01:30:36 PM:
>                    
          At the time the only option offered
by 
> the CAP spec was the iTIP approach (not real-time).

Umm, iTIP busytime REQUEST/REPLY can
be done "real-time"; there is no store-and-forward lag inherent
in iTIP.  iMIP is the store-and-forward binding (aka email binding)
of iTIP.  CAP is a superset of what started out as iRIP, the real-time
binding of iTIP.

In CAP it should just be a matter of
using an iTIP Section 3.3.2 REQUEST whose response is an iTIP Section 3.3.3
REPLY message.    iTIP VFREEBUSY REPLYs are defined as:

   The "REPLY" method in a "VFREEBUSY"
calendar component is used to
   respond to a busy time request. The method is sent by the recipient
   of a busy time request to the originator of the request.

so since the CAP server needs to generate
some kind of response for the VFREEBUSY REQUEST its logical that it generate
this.   

I have not been able to keep track of
the recent flury of emails this thread so perhaps I missed some discussion
on why this is NOT doable in CAP utilizing iTIP messages.  

BTW: Back on 3-Apr-03 I had proposed
we reinstate the CAP-Draft-03 (thru 05) text regarding CS and VFREEBUSY
creation/maintenance.  It was removed in Draft-06 w/o any WG discussion
and this went unnoticed for a while.  I think any design that expects
CUAs to maintain busytime data is inherently flawed and unscalable; the
CS is best suited to do this mainteance and this results in much more accurate
busytime data.  Exactly how the CS does this internally is not relevant
to CAP since thats an implementation detail. 

Yes Virginia that means someone could
code up a CS that statically calculates the data at some interval and thus
is not as consistanly accurate as a dynamically calculated/updated CS but
so??   We cannot design a protocol to deal with poor implementation
details and provide the functionality users want.   In any case, was
there any disent or thought to readding the prose back and would it address
some of the busytime concerns others have too?

Bruce
===========================================================================
Bruce Kahn                  
             INet: Bruce_Kahn@xxxxxxxxxxxxxxxx
Messaging & Collaboration            
    Phone: 978.399.6496
IBM Software Group                
        FAX: and nothing but the FAX...
Warning: Dates in Calendar are closer
than they appear.