Re: [VIPR] Agenda for 2011/09/23

Marc Petit-Huguenin <petithug@acm.org> Tue, 11 October 2011 19:07 UTC

Message-ID: <4E9493E6.8000804@acm.org>
Date: Tue, 11 Oct 2011 12:07:18 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.23) Gecko/20111010 Iceowl/1.0b2 Icedove/3.1.15
MIME-Version: 1.0
To: Dean Willis <dean.willis@softarmor.com>
References: <4E7B5199.5000409@petit-huguenin.org> <3175C4C5F682C145B25808EB5737F16C0ECA7F83@xmb-sjc-21b.amer.cisco.com> <1D062974A4845E4D8A343C653804920206805382@XMB-BGL-414.cisco.com> <4E89E235.6000908@acm.org> <1D062974A4845E4D8A343C6538049202068FE74D@XMB-BGL-414.cisco.com> <4E8F03CE.9000504@acm.org> <4E8F276B.5090200@softarmor.com>
In-Reply-To: <4E8F276B.5090200@softarmor.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Cc: vipr@ietf.org
Subject: Re: [VIPR] Agenda for 2011/09/23
Precedence: list

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/07/2011 09:23 AM, Dean Willis wrote:
> On 10/7/11 8:51 AM, Marc P
>> Yes, but the difficulty here is how to choose this route.  The proposal sent by
>> Hakim, i.e. sending the capabilities of the endpoint together with the SIP route
>> and ticket, would permit this, as only the VIPR domain the closest to the end
>> user would be able to send the real capabilities of the endpoint (extracted for
>> example from the SIP registrar).  The transcoding is interesting, as an
>> intermediary using it could eventually claim to better route than the direct
>> route.  Perhaps what we need is a new media feature tag that would indicate that
>> a capability set is "native" (i.e. built directly from the SIP registrar or
>> information coming from the endpoint), instead of been from a intermediate media
>> point, like transcoding.
> 
> 
> If we create a situation where there is an incentive for an intermediary to
> mis-state a capability set in order to "capture" more revenue-generating
> traffic, then the intermediaries are going to lie like politicians seeking
> re-election.
> 
> I think this is a MUCH harder problem than we currently realize.
> 
> Even Hakim's proposal requires that route-validation also validate the offered
> capabilities, not just evaluate them and choose the richest set. This is hard; a
> motivated intermediary could probably convince a test-bot that it is the target
> instead of an intermediary.

Perhaps one way to solve this would be to lower the priority of a route if a
subsequent SIP call made using it does not match the capabilities advertized.
It could also be done with an OPTIONS, but the response can also be faked.  It's
a little bit more difficult to fake the actual video (I suppose that wideband
audio can be faked by resampling the narrowband signal).

- -- 
Marc Petit-Huguenin
Personal email: marc@petit-huguenin.org
Professional email: petithug@acm.org
Blog: http://blog.marc.petit-huguenin.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iEYEARECAAYFAk6Uk+UACgkQ9RoMZyVa61fr6QCfWdjxaiKUBeHoFBu9xs54aWK0
98cAn3miNyYPjcpZZhCMQ0lBmVSCFoC1
=mZo4
-----END PGP SIGNATURE-----

[VIPR] Agenda for 2011/09/23 Marc Petit-Huguenin
Re: [VIPR] Agenda for 2011/09/23 Hakim Mehmood (naim)
Re: [VIPR] Agenda for 2011/09/23 Muthu Arul Mozhi Perumal (mperumal)
Re: [VIPR] Agenda for 2011/09/23 Marc Petit-Huguenin
Re: [VIPR] Agenda for 2011/09/23 Muthu Arul Mozhi Perumal (mperumal)
Re: [VIPR] Agenda for 2011/09/23 Marc Petit-Huguenin
Re: [VIPR] Agenda for 2011/09/23 Dean Willis
Re: [VIPR] Agenda for 2011/09/23 Marc Petit-Huguenin
Re: [VIPR] Agenda for 2011/09/23 Dean Willis