Re: [v6ops] draft-moreiras-v6ops-rfc3849bis-00

[Arturo Servin <[e-mail]>]

On 8/21/13 1:35 PM, Fred Baker (fred) wrote:
> </chair>
> 
> On Aug 21, 2013, at 6:33 AM, "George, Wes"
> <[e-mail]> wrote:
> 
>> The main problem with extending 2001:db8::/32 to a /29, or
>> allocating 2xxx:db8::/xx is that it requires everyone to make
>> changes to their Martian routing and packet filters, unlike
>> pulling something from outside of 2000::/3. I don't view that as
>> "less expensive", at least operationally.
> 
> 
>> Also, if we're talking about your heartburn with a fairly large
>> requested prefix (e.g. a /20 or /27 or whatever), that lends
>> support to the idea of reusing space that otherwise is likely to
>> become the IPv6 version of IPv4 Class E space (e.g. 6bone space,
>> or 0200::) - formerly reserved, now reclaimable, but unusable as
>> "normal" space on account of currently existing filters (and
>> probably some dumb implementation hardcoding of "invalid"
>> space).
> 
> Well, that is how we came up with 10.0.0.0/8; that was the
> ARPANET's allocation. He could include that in the IANA section of
> the 6man draft as another alternative, and come to that conclusion
> in 6man.
> 

	I couldn't find the reference in RFC1918, Is it documented somewhere
else (besides ietf-mailing lists dicussions)?

Regards,
as