BBC News - 'Fresh proposals' planned over cyber-monitoring

[Peter Fairbrother]

On 11/05/13 19:06, Ben Laurie wrote:
> On 11 May 2013 17:26, Roland Perry<lists at internetpolicyagency.com>  wrote:
>> In article<EADEBDCB-9FBA-4EAA-ADC8-444B93CAD478 at batten.eu.org>, Ian Batten
>> <igb at batten.eu.org>  writes
>>
>>>> Basically, the people who specified IPv6 screwed up, big time, in not
>>>> making it backwards compatible. They've got all sorts of excuses, that
>>>> only geeks who understand products, but not product management, would
>>>> relate to.
>>>
>>>
>>> It would be interesting to understand what such a protocol would look
>>> like.  The basic problem surely is that if you have two address spaces,
>>> one larger than the other, you can't have a 1:1 mapping between the two
>>> (and to do so would defeat the object of making the address space
>>> larger).
>>
>>
>> All that was necessary was making the bottom 0.1% (or whatever) of IPv6 map
>> onto the old IPv4 space.
>
> Oh yeah? So how would an IPv4 machine address the remaining space?
>
>

Reserve say two bytes of IPv4 space for IPv6 - so eg 
144.134.x.x.x.x.x.x.x.x is an IPv6 address, and anything which does not 
begin 144.134 is an IPv4 space.

Getting an IPv4 machine to address the extra space is forward 
compatibility, not backward compatibility, and it wasn't built into IPv4 
software or RFCs.

But then, no other mechanism for forward compatibility, necessary to 
address a larger space, was.


-- Peter Fairbrother