[arin-ppml] Draft Policy 2009-2: Depleted IPv4 reserves

[Ron Cleven]

Michael,

I think maybe you missed the sarcasm in my email.  I was reacting to the 
following statement by someone from TELUS, who did not appear to have a 
very optimistic view of IPv6:

    So to summarize, the large ISPs should take one for the team
    and force their customers onto a network standard that would
    cut them off from the majority of content they are interested
    in, incurring massive cost while doing so, and then every
    other ISP out there doesn't have to worry about IPv4 exhaust
    because it won't even happen.  When you put it that way, I
    will bring the business case to my superiors and get approval
    right away!

As is pretty obvious, his assertion was that the transition to IPv6 for 
broadband customers would involve "massive costs" and such a transition 
would "cut [the adopting customers] off from the majority of content".

I don't have any opinion on the cost issue, though I have certainly read 
plenty of posts here that dismiss that assertion as silly.  However, I 
obviously know that a properly-designed IPv6 transition for broadband 
customers should/would not cut those customers off from a majority of 
content, as there appear to be a few thousand different ways of 
implementing such a transition (ok, I might be exaggerating a bit), most 
of which would not have that effect.

However, the sarcasm was not really directed at the TELUS person.  If a 
marketing person had been in charge of the design of IPv6, there would 
only be ONE way of implementing the transition, because the top 5 goals 
for such a design would have been:

1) Clean interoperability of IPv6 and IPv4 to expedite adoption
2) Clean interoperability of IPv6 and IPv4 to expedite adoption
3) Clean interoperability of IPv6 and IPv4 to expedite adoption
4) Clean interoperability of IPv6 and IPv4 to expedite adoption
5) Clean interoperability of IPv6 and IPv4 to expedite adoption

Products need to be defined in a fashion that allows them to eventually 
get to market, and, in this market, you have to play nice with the 
billions of IPv4-only devices out there, at least for some non-trivial 
transitional period of time.

All those bits could have been put to good use.  If I "owned" the IPv4 
address 1.2.3.4, I should implicitly own the IPv6 address x:1:2:3:4:y. 
In short, at worst I should have to buy a fleet of shiny new Cisco 
routers that understand that and have enough capacity to deal with it.

Instead, after a decade, there still seems to be as many network designs 
to deal with IPv6/IPv4 interoperability as there are ISP's who have 
dabbled with it.  But, there I go exaggerating again.  Sorry.  It is 
amazing that with all the smart people involved, we still find ourselves 
playing chicken with binary arithmetic.

Ron



michael.dillon at bt.com wrote:

>>So, there is no way for a PC operating over IPv6 to access an 
>>IPv4 web site or mail server or ...
>>
>>Hmmmmm, sounds like a problem.  Maybe some really bright 
>>people out there can figure out a way to make that happen?
> 
> 
> Please don't make inaccurate statements. If you don't know the answer
> then either don't comment, or ask where you can find out how to allow a
> PC on IPv6 to access IPv4 servers.
> 
> There are several ways that this can be achieved, and many of them are
> documented from the ISP perspective at ARIN's IPv6 site
> <http://www.getipv6.info>
> 
> For solutions which work independent of the ISP supporting it, try
> Google. 
> These solutions also exist and have for many years now.
> 
> --Michael Dillon
> _______________________________________________
> PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>