Re: AW: AW: [Sipping-tispan] TISPAN requirements, first requiements

[Miguel Garcia <[e-mail]>]

more inlince comments, just restricted to the point of discussion.

Schmidt, Christian wrote:

> Hi Miguel,
> 
> some more comments inline.
> 
> Best Regards,
> Christian Schmidt
> 
> 
> 
> -----Ursprüngliche Nachricht-----
> Von: Miguel Garcia [mailto:[e-mail]] 
> Gesendet: Mittwoch, 24. August 2005 12:02
> An: Schmidt, Christian
> Cc: [e-mail]; Alexeitsev, D; [e-mail]
> Betreff: Re: AW: [Sipping-tispan] TISPAN requirements, first requiements
> 

> I'll give you one example that came to my mind. Many of the 
> supplementary services in PSTN/ISDN assume that there is only one 
> line/terminal per user, so if the user is already busy from that 
> terminal/line, he won't be able to take a new call, thus, services such 
> as Call Forwarding on Busy will apply.
> 
> This is not exactly the same in SIP, where a user can be using several 
> terminal simultaneously, and there isn't a concept of a "line" or 
> "channel". Therefore, the fact that a user is busy from one terminal 
> does not necessarily imply anything with respect his busy condition. 
> This is a change with respect the PSTN/ISDN services.
> 
> Christian: This information should be added somehow, perhaps:
> "small variations are expected in some services (for example CCBS, because of different busy definitions) are expected.....".

Ok, we can add the clarification in the introduction.
> 
> 
>>3.1 General Requirements
>>"the user should receive the service without any degradation"
>>This seems to be somehow in contradiction with the "variation" passage in the overview section. Or do you mean variation as addon only?
> 
> 
> The requirement means that, if one user is in the NGN and the other in 
> the PSTN, the NGN user should not receive a degradation with respect the 
> NGN simulation service (read the "native network" in the text). 
> Similarly, the PSTN user should not receive a degradation with respect 
> the PSTN/ISDN supplementary service provided in the PSTN.
> 
> I think this isn't in contradiction with the "variation" passage.
> 
> Christian: OK, now I understand. But I am still a little concerned about the requirement. For example in case of CCBS. When the called PSTN subscriber get free, the CCBS call will be presented to the called subscriber with priority, to avoid other terminating calls to come first. How do you want to achieve this with SIP? Without this feature, this is a kind of service degradation for CCBS. Perhaps, you should modify to "without significant degradation".

I think the requirement is "without any degradation", as far as I 
understand. Whether that is achievable or not, that is another story, 
but the requirement is clear. We can discuss the implementation of CCBS 
when we reach that point, perhaps it is not the right time to do it now.

> 
> 
> 
>>There should also be a general requirement included:
>>GEN-3: "SIP UA not providing this simulation service should not be influenced, they are simple not able to provide the related service."
> 
> 
> I agree with the idea, but I don't think it is a requirement, but a 
> property or expectation of the system. But I can add some clarification 
> text indicating that we don't expect SIP UAs that do not implement the 
> services to provide the service to the user.
> 
> Christian: It is a requirement for the selection of a proper solution. Do not select a solution which is not backwards compatible. And the statement
> should avoid discussions, that with introduction of this services, all SIP UAs would have to be changed.

OK, I guess the requirement makes no harm, so I will add it. I am not 
quite convinced that we need it, but no problem, I will add it.

BR,

     Miguel
> 
> Best Regards
> Christian Schmidt
> 

-- 
Miguel A. Garcia           tel:[phone]
sip:[e-mail]
Nokia Research Center      Helsinki, Finland


_______________________________________________
Sipping-tispan mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/sipping-tispan