[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Sipping] Internet draft on missed calls / msgs notification
David,
> Of course even if you have that you need a data
> schema, which is mostly what this draft seems to be about.
Agreed. It was the 'server only' view that needs to be discussed and amended
IMHO.
>> What may the P2P SIP view be for these functions?
>Perhaps a fully-replicated distributed database among the Contacts for a
>particular AOR.
You have answered the question how this can be done without servers, and
there is more in-depth about serverless telephony in several papers that you
may already know:
http://www1.cs.columbia.edu/~library/TR-repository/reports/reports-2004/cucs
-044-04.pdf and more at http://www1.cs.columbia.edu/~kns10/publication/
>the purist model of no servers at all is the enemy of the
>sweet-spot model which avoids gratuitous use of servers when they add
>no value, introduce scalability and security problems, and just get
>in the way
Please don't attribute to me things that I have not said. The issue with
servers is:
>The cost of the servers may be less relevant for a service provider,
>compared to the operational costs for the servers, updates, etc..
For the operational costs you can include the cost of the staff to maintain
the servers, versus a self organizing server cloud.
> but that's nonsense.
And if you have a problem with the "purist model" then you would need to
explain why most of the Internet traffic is P2P anyway, as reported in
several places, such as
http://linuxreviews.org/news/2004/11/05_p2p/ and
http://www.cachelogic.com/research/slide1.php
"Purists" must be knowing something on how to avoid a costly "service
infrastructure" :-)
Thanks, Henry
-----Original Message-----
From: sipping-bounces at ietf.org [mailto:sipping-bounces at ietf.org] On Behalf
Of David R Oran
Sent: Saturday, September 24, 2005 2:33 PM
To: Henry Sinnreich
Cc: sipping at ietf.org; 'David A. Bryan'
Subject: Re: [Sipping] Internet draft on missed calls / msgs notification
On Sep 24, 2005, at 11:19 AM, henry at sinnreich.net wrote:
> It would be interesting to explain when and why message
> notifications are not endpoint functions and under what
> circumstances a server is preferable.
I you want in integrated view of all calls and not just the ones for
which the INVITEs reached a particular device.
For example, I have serial forking which rings my desk phone and only
if it doesn't answer does it ring my cell phone. I would however,
like a full view of the call log for my AoR, including those that
never rang anywhere (were screened, went to call coverage because my
phone was off or disconnected, were answered on one phone but I'd
like to have the call record on another phone so I can easily
originate a call from there, etc., etc., etc., etc., etc.)
> Commercial P2P products (Skype) and P2P SIP have no servers...
Henry, I love you, but that's nonsense. Of course they do, for
voicemail, skypeout, and likely tons of other things. As with most
things the purist model of no servers at all is the enemy of the
sweet-spot model which avoids gratuitous use of servers when they add
no value, introduce scalability and security problems, and just get
in the way.
> What may the P2P SIP view be for these functions?
>
>
Perhaps a fully-replicated distributed database among the Contacts
for a particular AOR. Of course even if you have that you need a data
schema, which is mostly what this draft seems to be about.
>
>
> The cost of the servers may be less relevant for a service
> provider, compared to the operational costs for the servers,
> updates, etc..
>
>
>
> Thanks, Henry
>
>
>
> From: Avasarala Ranjit-A20990 [mailto:ranjit at motorola.com]
> Sent: Friday, September 23, 2005 1:28 AM
> To: sipping at ietf.org
> Subject: [Sipping] Internet draft on missed calls / msgs notification
>
>
>
> Hi
> I want to propose an internet draft on notification for missed
> calls/ messages thru XML event package. Plz go thru it and give
> your comments.
>
> Thanks
> <<draft-ranjit-missed-calls-msgs-00.txt>>
>
> Regards
> Ranjit
>
>
>
>
> _______________________________________________
> Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
> This list is for NEW development of the application of SIP
> Use sip-implementors at cs.columbia.edu for questions on current sip
> Use sip at ietf.org for new developments of core SIP
_______________________________________________
Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors at cs.columbia.edu for questions on current sip
Use sip at ietf.org for new developments of core SIP
_______________________________________________
Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors at cs.columbia.edu for questions on current sip
Use sip at ietf.org for new developments of core SIP