[Freedombox-discuss] VOIP application layer

Marc Petit-Huguenin marc at petit-huguenin.org
Wed Jul 20 19:37:00 UTC 2011


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

On 07/20/2011 12:13 PM, Marc Manthey wrote:
> 
> On Jul 20, 2011, at 4:16 PM,
> FreedomBox-Discuss.NeoPhyte_Rep at OrdinaryAmerican.net
> <mailto:FreedomBox-Discuss.NeoPhyte_Rep at OrdinaryAmerican.net> wrote:
> 
>> On Tue, Jul 19, 2011 at 1:44 AM, Marc Manthey - marc at let.de
>> <mailto:marc at let.de> wrote:
>>> <snip>
>>> So why we need all this NAT /SIP/STUN / punching stuff ?
>>>
>>
>> Then, how long has it taken major implementers, large Internet Service
>> Providers (ISPs), large network companies, and national governments,
>> to buy IPv6 equipment and place it into service?
> 
> 
> In france one of the biggest ISPs (free.fr ) offers ipv6 for every customer
> without any extra charge since years. http://ipv6pourtous.free.fr/rani/
> 
> Comcast has released source-code for the next-generation of IPv6 enabled home
> gateway device (a.k.a. home router). 
> This IPv6 solution is based on the  OpenWrt  platform.
> 
> www.comcast6.net/ <http://www.comcast6.net/>
> 
> http://sourceforge.net/p/dslite-6rd/home/
> 
> 
> So i predict everyone has ipv6 before 2012 ends , so why we need ICE/SIP and
> STUN in ipv6 when we have end-to end connectivity ?
> http://en.wikipedia.org/wiki/End-to-end_principle

There is multiple reasons.  The first one is that IPv4 will never disappear, so
there is a need for a mechanism that can select a route that works between two
endpoints, which is what ICE is doing.  Note that ICE is *not* a mechanism to
traverse NAT, as a lot of people think, but a framework to manage various
connection methods, which can be NAT traversal methods, IP versions (i.e.
between IPv4 and IPv6), interfaces (in case of a multi-home endpoint), etc...

Another reason is that IPv6 unfortunately does not signal the end of NAT.  One
reason is to be able to renumber networks without changing the IP address of the
endpoint.  There is some effort at the IETF to find a solution to this problem
without a NAT, but it is probably already too late.

Another reason is that some people think that a NAT adds security to a network.
 People like to believe that there is simple solutions to complex problems, and
NAT is one of these.  So even without the renumbering problem, people will
continue to buy this particular snake oil after IPv6.  You can't fix stupid.

For all these reasons, we will have to live with ICE for a very, very long time.

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

iEYEARECAAYFAk4nLloACgkQ9RoMZyVa61fZcwCfUom4TtbgmMti+3mNG4c5cNo/
LuMAoKB7t89G0ZDnDc1SieVWCM0tmIO9
=O3xO
-----END PGP SIGNATURE-----



More information about the Freedombox-discuss mailing list