[arin-ppml] Just a reminder of some quick mathematicsfor IPv4that shows the long term impossibility of it

Ted Mittelstaedt tedm at ipinc.net
Mon May 16 17:49:43 EDT 2011

On 5/16/2011 10:09 AM, Chris Engel wrote:
> Ted,
>>> Even though I enjoy healthy debate as much as anyone, I'm not
>>> sure what the point or relevance of this thread is?
>> The point is that IPv4 isn't going to work to get the rest of the
>> world online.  Sorry it's degenerating into a NAT debate but the
>> NAT proponents seem to think that NAT will allow IPv4 to be used
>> forever on the Internet.
> Even if true, why does that matter?

If it WAS true or even remotely plausible then it WOULD matter.

But it isn't, which is why it doesn't matter, thus making your
question rhetorical.

> If 30 years from now if IPv6,
> for whatever reason, should prove insufficient to the planets
> internet needs does that mean that 30 years of policy regarding it
> will have been wasted?

The planet's addressing needs and the planet's internet needs are
different.  We aren't talking internet needs, we are talking
addressing needs.  And to say that IPv6 will wear out in 30
years is foolish and shows a lack of understanding about communication

We still to this day produce telephone gear that works on the POTS

Communication standards that reach a critical mass of popularity
usually don't go away easily.  We still use RCA jacks and plugs in
audio.  We still until very recently produced personal computers
with both parallel ports and serial ports on them.  The very fact that
we are having this discussion about IPv4 when the planet has not even
reached 50% penetration of Internet connectivity is proof enough of
how difficult it is to change a bad standard when a lot of people
use it.  (rs232 is a terrible serial port standard, BTW, as it is
not balanced)

When and if the planet ever reaches 100% market penetration of
Internet there will still be enormous amounts of IPv6 unused.  As an
addressing standard IPv6 is pretty much "it" as they say.

> No one posses a crystal ball to see with
> perfect clarity what the future holds. Generally policy should
> reflect immediate needs with the understanding that it may need to be
> modified in future if those needs change.
>>> Some participants here view universal end-to-end connectivity as
>>> an important goal and as such NAT being significantly harmful to
>>> the internet. Others of us believe that goal is not particularly
>>> desirable and possibly even harmful to the interests of a portion
>>> of the community....and thus NAT has significant utility that
>>> outweighs any potential harm.
>>> Much like politics or religion, I don't believe either side will
>>> be effective in changing the others beliefs no matter how much
>>> verbiage is expended in the effort. That seems evident by the
>>> number of times this particular discussion has taken place on
>>> this list.  Is it possible to simply agree to disagree on the
>>> utility/harm of NAT and set aside that portion of the
>>> discussion?
>>> Can we simply agree that at this particular point in time IPv4
>>> address space continues to have some value/use to a significant
>>> portion of the internet community?
>> So it's the "I've got mine Jack to bad there ain't any left for
>> you" approach?
> What are you talking about here? No one can manufacture more IPv4
> then currently exists to assign to those who don't have it yet. They
> can however present you with a choice as to whether you want IPv6
> space or whether you want some "limited subset" of IPv4 functionality
> behind something like CGN.

No, they can't - because to run the CGN you must have 'real' IPv4 and
when that is gone, what are you going to assign to the CGN's?

> If there is some genuine value to either
> then the dynamics of a free market will ensure that such choices are
> available.  In the long run, companies don't succeed by offering their
> clients inferior products.....and if they do, their competitors have
> only themselves to blame for not making a compelling enough case to
> consumers about the advantages of their own products. Would you
> prefer rule by fiat?

rule by fiat?  Are you arguing that this is always bad or something?

>>> If we can generally agree on that proposition, then it seems
>>> clear that ARIN still has some responsibility for setting
>>> policies in regards assignment of that space. The question of
>>> whether the rest of the worlds population of human's, llama's or
>>> house flies will be able to access the internet through IPv4
>>> strikes me as entirely tangential to that point.
>> Since ARIN has essentially completed assignment of that space,
>> there is really not much left to set as policy in the IPv4 realm
>> other than continued interference in transfers of IPv4 from one to
>> the other party.
>>> FWIW, my particular hope is that IPv6 see's a steady increase in
>>> adoption so that people who do value publically addressable space
>>> can get it, IF they want it....and that NAT&   IPv4 (and maybe
>>> even NAT66) continue to be available to those of us who prefer it
>>> as an option.
>> But those NATs will NOT continue to be available to those of us who
>> prefer them because they require IPv4 to go on the "outside" of the
>> NAT.
> Please explain, this isn't clear?

When the last of the RIR's assigns out the last IPv4 that will be the
end of it.  If a new ISP that has no numbering comes along where are
they going to get their IPv4 to get online?  Unless you think that CGN
automatically includes NAT64 also?

>>> The world is a diverse place, I don't see why the internet should
>>> not reflect that diversity in being able to cater to a varied
>>> and sometimes conflicting set of interests. Yes, that adds to
>>> the complexity of the system from an engineering
>>> standpoint....but so does manufacturing more then one size of
>>> shoe.
>> Sounds good so let's go ahead and run IPX on the Internet too...
>> since I like that old Netware protocol better than IP.  So I should
>> be arguing for ISPs to all enable it on their routers based on
>> backwards compatibility, using that logic.
> No one is stopping you...nor should they have the right to do so.  No
> one has the right to tell anyone else what packets to run on their
> own network segments. If what you are doing proves problematic, you
> may find your segment isolated and not much traffic routed through
> it... but if you can manage to make it work well enough for the
> people using it, why should anyone be able to dictate to you the
> details of exactly how you achieve that?

I didn't say run IPX on my own segment, I said run it on the Internet.
Your getting pretty desperate to keep this NAT thing justified when
you stoop to remarks like that.

>> The fact of the matter is that what other people choose to do DOES
>> affect you, the Internet is not some wild west network where there
>> is no law and governance and you Chris can do as you damn well
>> please.
>> Every time someone else brings up another AS it uses a piece of ram
>> in MY router.  Every time I subnet the advertisements of my own AS
>> and prepend some and not others to balance my load it uses a piece
>> of ram in YOUR router.  Like it or not, we are tied to each other.
> Um....as far as I am aware....we are responsible for how our
> individual routers work....we can certainly choose NOT to carry a
> particular route in it if we want.....and we can certainly choose to
> drop/block/filter particular packets going across our boundaries if
> we choose. There are consequences for us making poor choices of
> course....but that doesn't mean we don't have any choice.

It isn't reasonable to assume that an admin of an ISP that takes a
full feed would have the time to be able to research every single
entry in the table and decide if he was going to block it or not.

>> How well do you think the US highway system would work if every
>> state was allowed to set their own highway widths?  Or set their
>> own standards on what color vehicle brake lights would be?  Would
>> you like to get a ticket in my state for having an amber
>> directional signal on the back of your car instead of red?
> Actually, the US highway system works alot like that. Different
> states have different speed limits set, different regulations for
> what you can do while driving (in my state you can get ticketed for
> talking on the cell phone while driving or having tinted glass, in
> others you can't) and of course once you get off of PUBLIC roads and
> onto PRIVATE ones, there is an entirely different set of rules
> governing their use. Note that the US Highway system is PUBLICALY
> funded, which is a big differentiator.

In the US, only the old US highways are allowed to be noncompliant, any 
additions to the US highway system must meet design standards 
recommended by American Association of State Highway Officials (AASHO)

And of course, the Interstate highway system has a much higher
set of AASHO standards to meet.

>> This is why the Internet cannot reflect the world's diversity in
>> it's protocols.  You can be as diverse as you want with website
>> content and suchlike but the value of the Internet is that everyone
>> is talking with the same protocols.  We currently have a problem
>> with one of them right now and we have a plan in place to change it
>> that was set up a decade ago that all the major networks have
>> signed on to doing - and what is going on is a few malcontents out
>> there who were asleep at the switch and are too lazy to educate
>> themselves about how IPv6 works now want to derail that plan by
>> pretending CGN will allow us to ashcan IPv6 and keep IPv4 going in
>> perpetuity.
>> It is one thing to regard CGN as transitional and admit you have a
>> grotty infrastructure that needs it that you can't replace just
>> right now, but you are going to soon.  It is quite another to claim
>> that it is reasonable that CGN will allow IPv4 to be a permanent
>> future protocol on the Internet, but that is what your doing.
>> Ted
> I certainly have no ability to stop anyone from deploying IPv6 that
> wants to.... wouldn't try if I could. If people are CHOOSING
> something else over IPv6...then the only people who are to blame for
> that are the designers, proponents and vendors of IPv6 solutions for
> not offering a compelling enough reason/argument for people to CHOOSE
> IPv6.  Don't blame others for your inability to convince people of
> your vision for the future.

IPv6 isn't my vision, I didn't design it.  And people often choose
things that are self-destructive, why do you think the US has an
epidemic of obesity?

This isn't a case of people understanding both IPv6 and IPv4 and
choosing IPv4.  It's a case of people understanding IPv4 because
that is all they know, and not choosing IPv6 because they don't
understand it.  Given a chose between something they know and understand
and something they don't, people as a group will choose the first thing
even when it is self-destructive.

There are at current estimate over 6 thousand people walking around 
today who are alive because of vehicle air bags.  Yet, the general
public back in the 70's thought air bags were stupid when they first 
heard of them.  Fortunately the federal government ignored the idiots
in '84 and mandated them.  I suspect that those 6,000+ people are
probably in the pro-airbag camp nowadays.


> Christopher Engel (representing only my own views)

More information about the ARIN-PPML mailing list