RE: [obix-xml] Alarm Acknowledgement

["Brian Frank" <bfrank@tridium.com>]

Dave,

Remember everything can be batched using the Sys.batch operation.

Personally I favor an operation directly on the alarm since it is more
flexible and keeps the spec simpler.  You can still implement it however
you want with creative use of URIs for alarms.

Brian


-----Original Message-----
From: Richards, Dave [mailto:drichards@trane.com] 
Sent: Thursday, March 23, 2006 10:51 AM
To: obix-xml@lists.oasis-open.org
Subject: RE: [obix-xml] Alarm Acknowledgement

Having the operation be against the alarm object itself leaves you
unable to
do a bulk acknowledgement.  You'd have multiple round trips rather than
one
and I don't know how important that is.  The user would not notice any
difference unless the network was unusually (or inherently) slow.

Also, are there considerations for the implementation of having
operations
against dynamic/transient objects?  A fixed service would seem simpler
in
some respects.

So to answer your question, I have a feeling, but I'm not yet sure what
it
is.

-----Original Message-----
From: ahansen@tridium.com [mailto:ahansen@tridium.com] 
Sent: Thursday, March 16, 2006 3:15 PM
To: obix-xml@lists.oasis-open.org
Subject: [obix-xml] Alarm Acknowledgement

I would like change how oBIX alarms are acknowledged.  The specification
currently defines an AlarmService (section 15.4) for doing this.  The
sole
purpose of the AlarmService is for acking alarms and the way it is done
is
by passing the service a list of alarm URI's.

Since alarms are required to have URI's and are therefore individually
reachable, I don't see the need for the level of indirection the alarm
service introduces.  Instead, I would like to add an ack operation to
the
AckAlarm (section 15.2.3) contract.

Does anyone have any feelings about this?