We serve many customers with Call Reports from a whole boquet of PBX's, from a myriad of vendors. I understand the purpose of using the notification typology for real-time circumstances. We do have customers with real-time requirements where we bar usage of a device if an expense quota has been reached. My concern is, what happens if a customer's server is down, and is not listening for notifications? Will a stream of notifications be sent when communications are re-established? Given the call record size, I understand why you don't allow a date range parameter query on the call record graph. Is there any way a GetCallIDList allowing date range parameters can be made available, which info can be used in turn to request each call record? This would allay fears that notifications have been missed, and wouldn't necessitate real-time notification for customers where there is no such requirement?