[arin-ppml] Change of Use and ARIN (was: Re: AFRINIC And The Stability Of The Internet Number Registry System)

hostmaster at uneedus.com hostmaster at uneedus.com
Tue Sep 14 03:50:33 EDT 2021


I do not see IPv6 as a failure. In most networks where both protocols are 
available, more than 1/2 of the traffic flows the IPv6 way. That is NOT a 
sign of failure. Lots of work has been done to extend the lifetime of 
IPv4, and to drag as many unused IPv4 addresses back into active use.

There is nothing wrong with use of either IPv4 or IPv6.  Lots of people 
have had IPv6 turned on in their ISP's routers (often by its replacement 
because the old one went bad) and did not even notice that now more than 
1/2 of their traffic uses IPv6. This is because most of the 
biggest by volume websites happen to publish AAAA records, and the 
standard states that IPv6 will be tried first if available on a given 
network.

Either works well for HTTP(s), the main traffic on todays internet.  The 
only real advantage that IPv6 has over IPv4 is the address size and the 
lack of checksum calculations at every hop. Things can be done much more 
efficiently in IPv6, when NAT and other "hacks" are removed, and the nodes 
can all directly communicate with each other in a manner that we used to 
do in IPv4 all the time prior to NAT.

CIDR and NAT were the two technologies that saved IPv4 from exhaust more 
than 25 years ago.  Without those technologies, IPv4 would have run out 
much sooner.  Another factor that extended IPv4 was the decision of China 
to use IPv6 only on their academic network, instead of eating up the 
remaining IPv4 address space like they were proposing around the 2008 time 
frame.

IPv6 simplifies things.  There is but one network size, a /64.  No more 
sliding netmasks when a given site gets too big for a given size subnet 
like happens a lot in IPv4 based public address networks.  NAT is also 
generally not used, which means all that overhead of maintaining tables in 
routers is not required.  Even the calculating of checksums at each hop 
has been eliminated, allowing routers to process IPv6 packets

This makes creation of networks more standard and automatic, reducing the 
workload of the network admin.

Eventually, no mattter what steps are done, IPv4 will exhaust.  There are 
but a fixed number of addresses, and this will never change.

I see IPv6 as freedom, and the way back to an earlier time when every node 
had a public address.  In many ways, it is freer than even those with a 
class "A" of the past, since every network everywhere is larger than even 
that class "A".  We have come a long way with it, and most of the major 
sources of traffic have adopted it, with more being added each day.

By no means do I consider IPv6 dead, and at this point I do not think that 
any other protocol like an IPv8 is going to be developed at this point and 
take over IPv6.  In any case, even if that elusive IPv8 were developed, I 
doubt those that want to stick with IPv4 will move to that either for the 
same reasons they do not want to move to IPv6.

It was said that a 32 bit address was chosen, because it was good enough 
for an experiment. That experiment was called IPv4, and it has never 
ended.  Even after IPv6 gets greater use then IPv4, I doubt that IPv4 will 
ever be closed down, but I do predict that it will become an option that 
you will have to pay extra for on a standard internet connection, as at 
some point nearly every public service you will want to reach will be 
available on an IPv6 address.

Albert Erdmann
Network Administrator
Paradise On Line Inc.


On Mon, 13 Sep 2021, Ted Mittelstaedt wrote:

>
>
> On 9/13/2021 2:43 PM, Joe Maimon wrote:
>> 
>> 
>> Ted Mittelstaedt wrote:
>>> Helping them to find IPv4 just kicks the can down the road a bit more.
>> It helps them right now. What it doesnt help is the state of IPv6
>> deployment. Which most users could care less about. Its just your (most
>> noble) goal, and you would have us all ignore others' needs of today for
>> your vision in the who-knows-when future.
>> 
>
> I didn't invent IPv6.  Not my circus not my monkeys.  I'm just a guy
> who listened to the experts when they said "we invented this IPv6 thing"
> then learned about it and used it.  I didn't listen to the experts and
> try to think up excuses for not doing what they said or try to figure out 
> ways to snake IPv4 away from other people.
>
>> My goal here is to encourage people to step back and consider what has
>> actually happened and how and to learn from that. My hope is that the
>> self governance model survives this self inflicted disaster.
>> 
>
> The lesson is people shouldn't have dragged their feet on IPv6.  The experts 
> told them what was coming, the experts built a replacement, and
> the sheeple out there in networking and sysadmin land ignored it and
> now are being smacked around because they ignored it.
>
>>> 
>>> Just because kicking people's asses to get IPv6 deployed isn't going to
>>> help someone who doesn't have IP4 RIGHT NOW doesn't mean it's a wasted
>>> effort like you claim.
>> 
>> The fact that it has failed for 20 years means exactly that.
>> 
>
> It means nothing.
>
> Electric cars have been around for a century.  But claiming that they
> failed for 80 years means they are failing now is idiotic because
> they aren't failing now.
>
> People tried for well over 20 years to make a vaccine for coronaviruses.
> Lots of "experts" said when the pandemic started the money spend on trying to 
> get one would be a waste of money because they tried for 20
> years to get a working one and it never worked.  Then Moderna came along
> and said "up yours, experts"
>
> Charles Babbage invented the Difference Engine in 1821.  A machine based
> on it was finally built in 1854 by George Scheutz.  Babbage invented the
> Difference Engine 2 and it was finally constructed and currently sits in
> the London Science Museum and it can calculate to 31 digits.
>
> Welcome to how technology works, Joe.  Lots of times an idea that works and 
> is valid and proved out is not implemented until decades later.
> It is simply that the time wasn't right for it.
>
> The thing that's so incredible about IPv6 wasn't that it was thought up
> and built and works.  It was that it was thought up and built and worked
> decades before we ran out of assignable IPv4 and really needed it. Usually 
> inventions like it are the result of a madcap crash at the last minute and 
> thrown together.
>
>
> Ted
> _______________________________________________
> ARIN-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:
> https://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>


More information about the ARIN-PPML mailing list