Re: 17 April Mobility Conf call notes
Hi, Laurent:
Please see the earlier emails sent by me, Ed, and Paul as well.
I guess that you are not understanding the other side of the coin. It has been explained earlier.
No one is saying where the "services" to be EXECUTED using some mandatory criteria. You are jumping much ahead of everyone by saying that home CSCF will be used to execute the services. Who knows ahead of time how many ways a service can be executed? We are NOT standardizing "services" per se. We are standardizing the protocol to support mobility. We MUST Not be MANDATING that the protocol should be used to execute a service in a given way. We MUST not be mandating how a HLF will accessed by using ONLY a given particular call flow (for example, AT&T's contributions show 3 different ways, while your contribution shows only one way for a particular scenario). A protocol can used to create many services depending on the relationship between the service providers and users. No one should ever dream to "hard-wire" a protocol feature for execution of services. It will be totally against the "norm" of development of the protocol.
The important point is that a subscriber needs to store some information: Home/Visited/Visiting/Target GK/zone/network address (point of attachment). It is an OPTION. If you do not like this, you are NOT forced to use it. It is your choice.
Who has given the right that a subscriber will NOT be able to register the information related Home/Visited/Visiting/Target GK/zone/network address (point of attachment) if they want to? This information, like all other parameters, will be stored in the HLF/VLF as appropriate.
Contributions are there: D.354, MD-017, and MD-018 in explaining the benefits and subsequent emails have also explained (please also see Ed's and Paul Guram's emails) the benefits that could be realized. If Alcatel does NOT want to use those OPTIONs, no one will force it. It is your choice.
As I explained yesterday's conference (also Ed explained in his email), there is very little value to use term like home adm domain unless the information like Home/Visited/Visiting/Target GK/zone/network address (point of attachment) is stored in the HLF/VLF as appropriate.
So, if people disagree with this concept, contributions are needed to understand. So far, no one has provided any contributions in explaining the things why they are opposing the concept. In fact, like AT&T, Motorola has also submitted contribution in supporting the concept of Home/Visited/Visiting/Target GK/zone/network address (point of attachment).
Hope this will clarify further. If you need to have any more clarifications, please let me or other know.
Best regards,
Radhika R. Roy AT&T
-----Original Message----- From: Laurent Thiebaut [SMTP:Laurent.Thiebaut@ALCATEL.FR] Sent: Tuesday, April 18, 2000 10:34 AM To: ITU-SG16@MAILBAG.INTEL.COM Subject: Re: 17 April Mobility Conf call notes
I support Jaakko's comments: the notion of Home is needed (at least for subscription information storage and for central place that knows which GK/VLF is currently hosting an user: HLF ) In some configurations, service needs to be executed in the Home domain. This does not mean that it has to be executed in a Home CSCF It is not contributions telling "what terms SHOULD NOT be defined", but contributions telling "why some terms SHOULD be defined" that are needed. Best regards Laurent T.
V Laurent Thiebaut tel: +33 (0)1 3077 0645
A L C A T E L e.mail:laurent.thiebaut@alcatel.fr
Jaakko Sundquist jaakko.sundquist@NOKIA.COM on 18/04/2000 16:06:17
Please respond to Mailing list for parties associated with ITU-T Study Group 16 ITU-SG16@mailbag.cps.intel.com
To: ITU-SG16@mailbag.cps.intel.com
cc: (bcc: Laurent THIEBAUT/FR/ALCATEL)
Subject: Re: 17 April Mobility Conf call notes
Hi Radhika & Laurent,
See my comments embedded...
The contributions and explanations have been provided why "Home" is needed in the same context of "Home" domain.
The Home Administrative Domain came about because it seemed a good idea to bind the subscription of a User to some administrative entity (such as an operator). At the same time we have defined the HLF as the only place in which location information about the user can always be found (albeit clearer definitions about this information are still needed). Thus the domain containing the HLF of the user was in effect defined as the Home Administrative Domain of the user.
If we do not accept the "home" for GK, zone and home network address, we have serious problems to accept the word "home" in any place of mobility whether it is "home" Adm domain.
I do not follow your reasoning above at all. Concidering a home GK or zone, I can understand that they may be needed if the service execution is done in the "home" of the user, i.e. all calls for a user would be controlled by the same GK, namely the home GK of the user. For service execution in the Visited Administrative Domain, I do not see any reason, why the concept of a home GK would be needed. Furthermore, if a concept is not definitely needed, I will not want to define it. I also feel we need to examine and define the "service execution in the home environment" or "virtual home environment" model better, so that everyone understands, what it is all about. As for the need for a home Network Point of Attachment, I am waiting for your contribution that clarifies, why we would need it.
I have very a serious objection with your idea that the word home should not be used. If it is so, let us NOT use "home" anywhere in the document.
The point is that so far, if I understand Laurent correctly, neither he or I have accepted the concepts of home GK or home Network Point of Attachment. Those are the terms that we would not like to see in the picture.
It is an OPTION to use "home". If anyone does NOT like it, they may NOT use it. Time and again, it has been shown why it is needed.
It will be very difficult to make any meaningful progress unless we are in consistent in defining term terms.
I like to see contributions explaining why "Home" should NOT be used.
I do not want to see contributions telling what terms SHOULD NOT be defined, I want to see contributions telling why some terms SHOULD be defined. So far I haven't seen such contributions on home NPoA and I'm still not sure about the home GK either.
-Jaakko
participants (1)
-
Roy, Radhika R, ALARC