Re: [rtcweb] Appropriateness of bypass mechanisms (Re: HTTP Fallback draft)

"Cullen Jennings (fluffy)" <fluffy@cisco.com> Fri, 07 September 2012 14:23 UTC

From: "Cullen Jennings (fluffy)" <fluffy@cisco.com>
To: Harald Alvestrand <harald@alvestrand.no>
Thread-Topic: [rtcweb] Appropriateness of bypass mechanisms (Re: HTTP Fallback draft)
Thread-Index: AQHNjQRDzKB1tIUge060+/uXq+3R1A==
Date: Fri, 07 Sep 2012 14:22:51 +0000
Message-ID: <FE024D5B-3BC1-41E2-9740-93BE61741BF3@cisco.com>
References: <20120807180156.286e74d2@rainpc> <D5BDA7BE-FE55-47FA-99FD-1645084370B0@gmx.net> <20120807191226.5b8e7f32@rainpc> <BLU401-EAS2566333A7F4DEA0D3BFDBE593CD0@phx.gbl> <913383AAA69FF945B8F946018B75898A1477EDB9@xmb-rcd-x10.cisco.com> <CAD6AjGSU8mzbcdbOkgGtAms1tdHhjiuQn_NFXELwO2kjegJkCQ@mail.gmail.com> <502285B2.8080007@alvestrand.no>
In-Reply-To: <502285B2.8080007@alvestrand.no>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <7A873E7259F3AC438D1A6B1EB4DD7CC8@cisco.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "rtcweb@ietf.org" <rtcweb@ietf.org>
Subject: Re: [rtcweb] Appropriateness of bypass mechanisms (Re: HTTP Fallback draft)
Precedence: list

On Aug 8, 2012, at 9:28 AM, Harald Alvestrand <harald@alvestrand.no> wrote:

> Forking thread....
> 
> On 08/08/2012 05:05 PM, Cameron Byrne wrote:
>> 
>> 
>> Is this thread  really about the ietf engineering a way to by-pass network policy set by network operators?
>> 
> It's about getting stuff done in the presence of rules that hinder the "simple way".
> Those rules may be set that way deliberately, by ignorance, because of implementation limitations, or in error; there's no way to know.
> 
>> I do not believe that is acceptable.
>> 
> We've already done TURN, including TURN over TLS.

TURN was specifically designed to allow the network operator to be able to continue to control and enforce policies similar to what they currently do in their firewalls. You will note Kaufman strongly disagreed with this design principle and would have done the permissions in TURN significantly differently - that would have actually made TURN faster and easier to use but would have removed this property. We choice to stay on the path of "not violating or circumventing the will of network administrators". 

That said, I think the discussion about making media relay / tunnel protocol that masquerades as HTTPS is leaping of the edge of the slippery slope. I have mixed feelings about if that is wise or not but that's a long discussion.

> 
> Old quote: "This is not about starting down the slippery slope; this is about how far we slide into the muck at the bottom".

Agree with that :-)

[rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Hannes Tschofenig
Re: [rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Ted Hardie
Re: [rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Iñaki Baz Castillo
Re: [rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Tirumaleswar Reddy (tireddy)
Re: [rtcweb] HTTP Fallback draft Marc Petit-Huguenin
Re: [rtcweb] HTTP Fallback draft Bernard Aboba
Re: [rtcweb] HTTP Fallback draft Justin Uberti
Re: [rtcweb] HTTP Fallback draft Tirumaleswar Reddy (tireddy)
Re: [rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Lorenzo Miniero
Re: [rtcweb] HTTP Fallback draft Cameron Byrne
Re: [rtcweb] HTTP Fallback draft Iñaki Baz Castillo
Re: [rtcweb] HTTP Fallback draft Tirumaleswar Reddy (tireddy)
[rtcweb] Appropriateness of bypass mechanisms (Re… Harald Alvestrand
Re: [rtcweb] HTTP Fallback draft Justin Uberti
Re: [rtcweb] Appropriateness of bypass mechanisms… Cullen Jennings (fluffy)