Re: [Sipping] A new draft on P2P SIP

[Philip Matthews <[e-mail]>]

Hi Arjun:

Thanks for your comments.

I agree that our draft is a mixture of requirements
and high-level architecture. Initially, our goal was
to write a draft that called attention to a class of
applications in which P2P SIP has equal functionality
to existing telephony networks. Our impression is that
most of the discussion (in drafts and on the mailing list)
has revolved around what we called "ad-hoc" networks
in the draft.We used the term "industrial-strength" to
differentiate us. It was *not* our intention to disparage
anyone else's work. However, as you point out, its use
can be misconstrued, and it was perhaps not the best choice.

In the interest of keeping our draft focused and short,
we only mentioned features that had not been discussed
by others. We then decided to add the section on high-level
architecture because, in our experience, keeping the
P2P and SIP layers separate is very beneficial.

We totally agree with you that there needs to be more
discussion of the requirements for P2P SIP. It is our hope
to have a first cut of a requirements document out in the
next few weeks to help this process along.

You are right that our draft only concentrates on telephony
features. It is the application that we understand. However,
it seems like you have other applications of P2P SIP in mind.
I can only think of instant messaging.
What do you envision?

- Philip


Arjun Roychowdhury wrote:
> Philip:
> 
> some quick comments:
> 
> *) Is this document a requirements document or an archirecture
> document ? It seems to have a little of both. Also, before we have a
> requirements specification on what a P2P network must do, one needs to
> have a common understanding/agreement on what really is a SIP P2P
> network and what goals it acheives. Your draft seems to concentrate
> only on telephony features. I am not sure if this is the only
> motivation. Either way, I would like to see more discussion on the
> scope
> 
> *) I would suggest you not use the term that this draft addresses
> "Industrial-Strength" P2P while others may not. I believe one of the
> goals of the IETF is to specify real-life/stable solutions/protocols
> and I am sure this would also be the goal for any alternate drafts. If
> you believe that your I-D addresses a different network from others,
> do state it as so (I believe that is your intent)
> 
> *) I don't believe the abstract of the draft really summarizes what is
> the scope of the document. For example, the draft mainly focuses on
> the need to provide the same functionality a PBX has without really
> having one. However, it does not talk about challenges in user/node
> discovery, fault management, cross network interop issues, Firewall
> traversal   and such that a p2p network may address. If these are not
> addressed, then the scope of just replacing a PBX will intelligent
> phones in a defined/closed network becomes a much simpler issue.
> 
> *) Section 3 seems to reccomend a functional architecture at a
> simple/20,000 feet view. I am not sure if this should be a part of a
> document when the need/scope/requirements are yet to be fleshed out.
> Also, since Section 3 is at such a high level, it is difficult to be
> able to discuss/argue on the need/functionality of this 3 level
> architecture.
> 
> 
> 
> regds
> arjun
> 
> 
> On Mon, 14 Feb 2005 17:24:25 -0500, Philip Matthews
> <[e-mail]> wrote:
> 
>>My co-worker, Behrouz Poustchi, and I have submitted a brief draft
>>contributing to the discussion around P2P SIP. In it, we present some
>>features that we would like to see considered in any P2P SIP standard
>>(or whatever it is going to be), based on our experience in this area.
>>
>>We are new to the SIPPING working group, but we hope that our draft
>>is a positive contribution to the subject, and would be happy to
>>participate in any work that is done in this area.
>>
>>The draft is available at:
>>
>>http://www.ietf.org/internet-drafts/draft-matthews-sipping-p2p-industrial-strength-00.txt
>>
>>Comments on the draft are very welcome.
>>
>>- Philip Matthews
>>
>>_______________________________________________
>>Sipping mailing list  https://www1.ietf.org/mailman/listinfo/sipping
>>This list is for NEW development of the application of SIP
>>Use [e-mail] for questions on current sip
>>Use [e-mail] 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 [e-mail] for questions on current sip
Use [e-mail] for new developments of core SIP