Markku, Bob , Sorry for the multiple responses .. Let me see if I get this correctly: What you're saying, Markku, is in fact that Annex L is not applicable for a Voice Mail server if it external to the home GK, as in my example. The applicable interfaces, would then be primarily H.450.7, and possibly Annex K (Service Control Indication), which in both cases need to be intercepted by the home GK in case the terminal is not available. Again, thank you very much for these clarifications -- Boaz

-----Original Message----- From: Markku Korpi [mailto:korpim@SCN.DE] Sent: Monday, March 20, 2000 10:30 PM To: ITU-SG16@mailbag.cps.intel.com Subject: Re: MWI - Message Waiting Indication

Boaz, H.450.7 MWI was made made for the purpose you indicate. The MWI uses non-call associated signalling connection, i.e. a connectioon that looks like a normal H.323 call, does not have any logical channel and is (usually) immediately released.

The messaging server actually sends the MWI to an H.323 address and does not need to know how MWI is processed at the terminating side. A gatekeeper (or a separate feature server) may, of course, act on behalf of the terminal and intercept the MWI for specific terminals. And as indicated by Bob Gilman in his answer, the GK/Feature server can then use, for example, Annex L stimulus signalling to control the terminals display and MWI lamp or other indicator.

On the other hand, if I have an intelligent H.323/H.450 terminal, that processes the MWI, I probably want my gatekeeper to pass the MWI transparently to my terminal.

Additionally you can open an Annex K HTTP session within the MWI procedure - provided the terminal (or its feature server) supports Annek K.

One more point: H.450.7 follows (not by accident...) QSIG MWI procedures, so that seemless interworking with PBX networks is relatively easy to implement in the gateways.

Perhaps somebody else on the list can comment the status of ongoing SIP work with regard to MWI. I would just say that, in order to reach "world wide feature control", it would be very important to follow the same functional procedures and functional message content, regardless of the underlying protocol. This is already demonstrated by using same functional procedures in H.323 and in QSIG, and this principle should/could be also applied IMO to SIP.

Markku Korpi

-----Original Message----- From: Michaely, Boaz [mailto:Boaz_Michaely@ICOMVERSE.COM] Sent: Sonntag, 19. Marz 2000 12:13 To: ITU-SG16@mailbag.cps.intel.com Subject: MWI - Message Waiting Indication

Hi folks, Please forgive my ignorance in case this is already a done deal, but does anybody know where we are standing in regard to MWI ?

Two aspects seem to be of interest: 1. (to the GK ) - A method for a Voice Messaging System to signal the GK that a certain subscriber has a new message.

2. (to the terminal) - A method for the GK or the VMS itself to signal the terminal that the subscriber associated with it has a new message

Do we have any means in general for signalling call-unrelated information (other than RAS) , that should be used for this purpose? e.g. is Annex K (HTTP) a suitable candidate ? Is H.450 suitable for call unrelated information ? Is anyone aware of similar work done e.g. in SIP which we may reuse or consider to avoid interworking problems ahead of time ?

Best regards, -- Boaz

Boaz Michaely Senior System Architect, Corporate Comverse Network Systems Tel: +972 (3) 766-3844 , Mobile +972 (50) 39-30-17 < http://people.itu.int/~michaely <http://people.itu.int/~michaely> > < mailto:boaz_michaely@comverse.com <mailto:boaz_michaely@comverse.com> >